PDA

View Full Version : D&D 5th Edition: the fifth edition of the discussion thread



Pages : [1] 2 3 4 5 6

Kurald Galain
2012-06-05, 06:57 PM
Basically, Fifth Edition D&D is coming out in the near future, and has an ongoing public playtest right now. Discuss it here! Also, chocolate!

Useful links:
Playtest sign up (http://www.wizards.com/dnd/Article.aspx?x=dnd/4ll/20120109)
Enworld's info compilation (http://www.enworld.org/forum/showwiki.php?title=Books:D+and+D+Next)


Previous threads:
First edition (http://www.giantitp.com/forums/showthread.php?t=218549)
Second edition (http://www.giantitp.com/forums/showthread.php?t=231033)
Third edition (http://www.giantitp.com/forums/showthread.php?t=242069)
3.5th edition (http://www.giantitp.com/forums/showthread.php?t=245504)
Fourth edition (http://www.giantitp.com/forums/showthread.php?t=244672)

TuggyNE
2012-06-05, 07:43 PM
It occurs to me that this thread could reasonably be titled "Discussions & Debates 5e", or similar :smallwink:.

1337 b4k4
2012-06-05, 07:59 PM
Continuing from old thread:


20% constitutes a "reasonable expectation for failure," and warrants a roll.

If the rules say that, if you roll, you will lose 20% of the time to a dead guy, then those are bad rules.

Is it reasonable for the players to override the rules in this instance? Sure. But the point is that they shouldn't have to.

You only have a 20% chance of failure if you roll. You only roll if there is a reasonable chance of failure. If the only way to fail is to roll, you don't roll. In other words, the die roll is what you do when you don't already know the outcome of the given situation. The die role is subservient to the "reasonableness" of the task at hand, it does not determine the reasonableness. This is RAW, and arguing otherwise is equivalent to arguing that you should get a chance to fire your bow across 3 continents and snipe the BBEG on his morning constitutional because you always succeed on a 20. No rule system can prevent you from twisting the rules into something they don't mean, and personally, I don't want to slog through legalese to play a game because some players and DMs don't know how to act like civilized adults.


You: You can't compare a level 1 Fighter and level 20 fighter and complain they're not different enough, because they're both veterans
Me: But we were comparing a level 20 Fighter with a level 0 commoner. And the results still made no ****ing sense
You: They're both veterans you can't compare them!


Seriously... what?

The what comes from your quote:


They're level 20 fighter, with all the benefits we know of that we can expect from that, against level 1 character with a negative modifier to the roll. So yes, the comparison is 100% valid.

If you meant Level 0 Commoner, that isn't what you wrote.


Unless we want to say that there's no real big difference between a Wizard at first and 20th level, after all he spent way more time training to be a wizard than he's spent actually adventuring, so they should be practically the same. Just give him more 1st and 2nd level spells per day, it'll be all right.

In older editions, where wizards didn't get their spells automatically and at whim, this is very much true. While a wizard might have the power to cast, and have access to higher level spells that doesn't make him a demi-god except insofar as magic allows one to bend the laws of physics. Perhaps part of the problem is that while D&D has increased its level caps, its lost the changing game styles that came with the old BECMI rules, leading to power creep and "astral slimes replace green slimes" because we have what in older editions really would have been immortals slogging through the cosmos instead wandering around saving the world from a lich with delusions of grandeur.

Personally, I think D&D ought to go back to BECMI and 4e "tiered play" give each one its own distinct play style and core, with flat curves among them. That way you can play your demi-god heroes and not have to worry about how a one size fits all rule system breaks down at that level.


Replace the dead man with a cripple with 1 strength. And while yes, logically it makes sense there is no chance of failure for the fighter, the rules indicate there is a fair chance of that cripple winning. It isn't a sure thing, so it needs to be rolled out.

Replace the dead man with a cripple and the rules still say if there's no reasonable chance for failure, then you don't roll. So RAW, the cripple still loses every time, unless maybe he's crippled because his legs are missing and so he walks on his hands all day and has the arm strength of a normal man.


My bad about the threshold. It doesn't change that no check above a DC15 will be ignorable for any PC (we've been told PCs won't get beyond 20 in their attribute), and you can't ignore contests because there is no set DC.

They won't be ignorable if there is a reasonable chance of failure. Even without an explicit take 10/20 rule, you can see that the rules clearly allow a check to be skipped if the action is "appropriate and stress free". Meaning while your level 20 PC can't automagically ignore the DC for breaking out of manacles in the middle of being assaulted, they most certainly can ignore them for being captured by goblins and left to rot in a cell for a few days.

They maybe should be more explicit, but my reading of the rules and their description in the previous LL, leads me to believe the intention is that regardless of level, when you want to set a challenging DC, that you should use 20 something, and that its up to the DM and the players to determine whether a given task is actually something with a chance of failure anymore. I much prefer this play style. It's clear you don't.


But there is no set DC. The whole point of the contest is that you have an opposed roll. It is impossible to say "This guy will be below my threshold" unless his bonus + 20 is less than your strength minus 5. So I guess the guy with 1 strength against the guy with 20 strength may be an auto success. (His max roll is 15, which is within your auto success thing), but anything with better than a -5 modifier you have to play out the check.

You could handle this any number of ways. Assuming that you haven't already determined no reasonable chance of success or failure (again, see page 1 of the rules) you could say that ability scores differing more than 5 (so a STR 20 fighter wins over an STR 14 fighter), or you could declare the DC to be the 10 + ability as an average, or you could simply declare on of the rolls to be the DC. Presumably the check is in response to one person deciding to act, so the DC is just the other person's skill roll. Again, they could make this clearer, but the first rule of checks is don't roll checks if there's no reasonable chance of failure.


Yes this is extremely stupid, and most DMs will completely ignore results that don't make sense, but that isn't justification for having ****ty rules that don't work in the first place.


Except the rules do work, because the first rule is don't roll if it doesn't make sense to roll. Just because there aren't numbers in that rule doesn't mean it's not a rule.

Tehnar
2012-06-05, 08:23 PM
Continuing from old thread:



You only have a 20% chance of failure if you roll. You only roll if there is a reasonable chance of failure. If the only way to fail is to roll, you don't roll. In other words, the die roll is what you do when you don't already know the outcome of the given situation. The die role is subservient to the "reasonableness" of the task at hand, it does not determine the reasonableness. This is RAW, and arguing otherwise is equivalent to arguing that you should get a chance to fire your bow across 3 continents and snipe the BBEG on his morning constitutional because you always succeed on a 20. No rule system can prevent you from twisting the rules into something they don't mean, and personally, I don't want to slog through legalese to play a game because some players and DMs don't know how to act like civilized adults.


So where is a cutoff point? When do you roll? Why can't WotC come up with clear criteria?



They maybe should be more explicit, but my reading of the rules and their description in the previous LL, leads me to believe the intention is that regardless of level, when you want to set a challenging DC, that you should use 20 something, and that its up to the DM and the players to determine whether a given task is actually something with a chance of failure anymore. I much prefer this play style. It's clear you don't.

I can't speak for Seerow, but I think the ability check system is a bad system and should not be in the game. I am not looking how to fix by setting DCs; a system as random as that should not be in any game in the first place.



You could handle this any number of ways. Assuming that you haven't already determined no reasonable chance of success or failure (again, see page 1 of the rules) you could say that ability scores differing more than 5 (so a STR 20 fighter wins over an STR 14 fighter), or you could declare the DC to be the 10 + ability as an average, or you could simply declare on of the rolls to be the DC. Presumably the check is in response to one person deciding to act, so the DC is just the other person's skill roll. Again, they could make this clearer, but the first rule of checks is don't roll checks if there's no reasonable chance of failure.

Those are houserules. While they may be reasonable, they do not belong in a discussion about a mechanic. WotC designers should be smart enough to identify there is a problem here (that you are solving with houserules) and implemented their own solution.

Seerow
2012-06-05, 08:36 PM
Continuing from old thread:



You only have a 20% chance of failure if you roll. You only roll if there is a reasonable chance of failure. If the only way to fail is to roll, you don't roll. In other words, the die roll is what you do when you don't already know the outcome of the given situation. The die role is subservient to the "reasonableness" of the task at hand, it does not determine the reasonableness. This is RAW, and arguing otherwise is equivalent to arguing that you should get a chance to fire your bow across 3 continents and snipe the BBEG on his morning constitutional because you always succeed on a 20. No rule system can prevent you from twisting the rules into something they don't mean, and personally, I don't want to slog through legalese to play a game because some players and DMs don't know how to act like civilized adults.


1) Actually when someone is making an attack roll at an impossible DC, they DO always roll. They don't get to fire across 3 continents because range gets capped based on the weapon. In fact it was for the exact reason you specified that auto success on a 20 was removed from skill checks in 3.5.

The RAW says that the reasonableness of the task at hand is determined by the DC of the task vs your relevant attribute. If the DC is less than your attribute minus 5, you have to try to roll. Period. You could maybe argue that the other person needs to roll first, effectively setting the DC for you (so if they roll less than a 15, you no longer have to roll), but other than that yes you do need to roll short of DM fiat.




The what comes from your quote:



If you meant Level 0 Commoner, that isn't what you wrote.

Okay sorry, level 1 commoner not level 0 commoner. You will note however I didn't specify fighter. No fighter would have a -1 strength mod. Level 1 character was short for anything at all that is not high level. Because yes, higher levels do represent increased power and expertise.




In older editions, where wizards didn't get their spells automatically and at whim, this is very much true. While a wizard might have the power to cast, and have access to higher level spells that doesn't make him a demi-god except insofar as magic allows one to bend the laws of physics. Perhaps part of the problem is that while D&D has increased its level caps, its lost the changing game styles that came with the old BECMI rules, leading to power creep and "astral slimes replace green slimes" because we have what in older editions really would have been immortals slogging through the cosmos instead wandering around saving the world from a lich with delusions of grandeur.

Personally, I think D&D ought to go back to BECMI and 4e "tiered play" give each one its own distinct play style and core, with flat curves among them. That way you can play your demi-god heroes and not have to worry about how a one size fits all rule system breaks down at that level.

Sure, I could get behind that. I've advocated as such in the past. Not 100% flat math, but big jumps at tiers. So the difference between a the lowest Heroic fighter and the highest Heroic Fighter might only be +7. That's totally acceptable. But once the Fighter jumps to Paragon tier, he suddenly gains a decent boost to just about everything (like +3-5 across the board), so it is noticeable that he's a cut above. Those numbers might scale another 5-10 over the tier, before going up to Epic where you see another jump.

Of course tiers would need to mean more than just the number jumps (because honestly in 4e that's all they were, and not even really good number jumps at that). For example effects like teleport and flight could be reserved for paragon tier. Effects like Simulacrum and Wish would be reserved for Epic. Basically each tier would unlock a whole slew of new modes of transportation, status effects, and defenses.

As for BECMI, Basic would basically be level 0. Really squishy, almost no powers or abilities. Expert/Champion is basically Heroic, Master is Paragon, and Immortal is Epic.




Replace the dead man with a cripple and the rules still say if there's no reasonable chance for failure, then you don't roll. So RAW, the cripple still loses every time, unless maybe he's crippled because his legs are missing and so he walks on his hands all day and has the arm strength of a normal man.

You are however defining "reasonable chance for failure" as "What does the DM think is reasonable?" Also if the DM can say the chance of failure is low enough for the cripple to beat the fighter despite the 20% chance, what does that say about say the Fighter trying to Grapple the troll? Can the DM just decide "Sorry that doesn't work" while ignoring the numbers that say the Fighter has a pretty decent chance at it? DM Fiat as the basis for making a skill system make sense isn't something I want.




They won't be ignorable if there is a reasonable chance of failure. Even without an explicit take 10/20 rule, you can see that the rules clearly allow a check to be skipped if the action is "appropriate and stress free". Meaning while your level 20 PC can't automagically ignore the DC for breaking out of manacles in the middle of being assaulted, they most certainly can ignore them for being captured by goblins and left to rot in a cell for a few days.

Sure, because the Fighter can retry until he succeeds. Unless he fails by what... 10 or more? With a +3 on the roll, that's a very real possibility against a DC over 13. And yes, there is no take 10 roll because they gave it to the rogue as a class feature.



They maybe should be more explicit, but my reading of the rules and their description in the previous LL, leads me to believe the intention is that regardless of level, when you want to set a challenging DC, that you should use 20 something, and that its up to the DM and the players to determine whether a given task is actually something with a chance of failure anymore. I much prefer this play style. It's clear you don't.

DC 20 challenges as the default for a difficult challenge? Jeeze! Remember most characters only have like 3 skills trained, and there is a very real chance of nobody having something for the situation at hand. You're looking at an attribute mod of +3-+5 being applied to the roll. If you make any semi-difficult task a DC20 minimum, that means any time you call for a roll, the PCs will succeed less than 20% of the time. (And remember, you just got done saying how 20% chance of failing to arm wrestle the diseased guy with 1 strength was negligible enough to not bother rolling. So I guess we're doing the same thing here, so the party just auto fails at anything unless they have the correct skill trained)


You could handle this any number of ways. Assuming that you haven't already determined no reasonable chance of success or failure (again, see page 1 of the rules) you could say that ability scores differing more than 5 (so a STR 20 fighter wins over an STR 14 fighter), or you could declare the DC to be the 10 + ability as an average, or you could simply declare on of the rolls to be the DC. Presumably the check is in response to one person deciding to act, so the DC is just the other person's skill roll. Again, they could make this clearer, but the first rule of checks is don't roll checks if there's no reasonable chance of failure.

But once again you're setting 'reasonable chance of failure' at "Whatever the DM feels like" which is absolute bull**** because every DM ever will rule these things differently. Seriously put a question of "Where is the reasonable line" to 10 different DMs kept apart, they'll come back to you with 15 different possibilities, with widely varying results. You are literally feeding Mearl's bull**** about how rules are bad because a DM will handle it with this line of argument.


Except the rules do work, because the first rule is don't roll if it doesn't make sense to roll. Just because there aren't numbers in that rule doesn't mean it's not a rule.


And Rule 0 in AD&D was that the DM can change anything. I guess that makes AD&D the perfect system, and nobody can complain about what was wrong with it.

Seriously DM fiat does not replace rules. If WotC is going to give us a rule set, it needs to be rules that can actually be followed not "Go with your gut feeling on it"

Fatebreaker
2012-06-05, 09:28 PM
You only have a 20% chance of failure if you roll. You only roll if there is a reasonable chance of failure. If the only way to fail is to roll, you don't roll. In other words, the die roll is what you do when you don't already know the outcome of the given situation. The die role is subservient to the "reasonableness" of the task at hand, it does not determine the reasonableness.

So, in summation:

"You only have a chance of failure when you roll and you only roll when there is a chance of failure."

...really?

Granted, I removed the word "reasonable" from that summation, but "reasonable" is not a word with a clear and consistent meaning from player to player, DM to DM, or game to game. It's a word bereft of value in that context. If a group has a problem with determining what is or is not reasonable, then this rule does not help them. If they do not have such a problem, then they don't need the rule in the first place. The DC is supposed to be an arbiter of reasonableness. If it is not, what good is it?

I don't need a game mechanic to tell me to "be reasonable." I can do that on my own. For free.

Dublock
2012-06-05, 09:32 PM
Seriously DM fiat does not replace rules. If WotC is going to give us a rule set, it needs to be rules that can actually be followed not "Go with your gut feeling on it"

Based on this I honestly find it hard to really judge how its going to pan out in terms of skills/rolls.

With what we see right now I get the impression there isn't enough of a difference between skilled and unskilled characters about a skill, but I feel like without more information like feats/magic items (I think Christmas tree with always be around in D&D), etc will impact and ultimately help with the rolls to the point I will be happy with. If we do have all the information I might add a few things.

I would love some high level information to see how it scales, their mindset, and help figure out the skill stand point. But we might not until release :(

Seerow
2012-06-05, 09:37 PM
By the way, just as an aside, since I did a lot of negative ranting in my last post: I really did like the suggestion of "If the attributes are within 5, it's a challenge. If not the higher attribute guy wins". I'd still like to see more attribute scaling with level (the idea of characters being capped at 20 as their maximum really doesn't sit well with me), but something like that would solve the problem of opposed checks coming up with unbelievable results. I could accept that as a standard rule, but that's the thing: it would need to be a standard rule. Because right now it's not, it's something you made up.

1337 b4k4
2012-06-05, 09:45 PM
So where is a cutoff point? When do you roll? Why can't WotC come up with clear criteria?


The cutoff point is at whatever point the DM and players decide it is. Do you and your DM think that a level 20 fighter should be able to break any manacles at any time, then go for it. Or are you seriously suggesting that if the rules said that level 20 fighters could only break manacles 20% of the time, that would be better? And if they chose something more reasonable, like 90% of the time, what about people who don't think level 20 fighters should be gods? They're not happy now either. The beauty of a flat system like what WotC appears to be going for, is that without an explicit rule for "how tough are manacles" we can all be happy. If you want all level 20 fighters to break all manacles, you don't bother with a check at all, if you want level 20 fighters to have a small chance, you pick a high DC, if you want them to have a really good chance, but not a guarantee, you pick a low DC. No need to rip out rules or try to recalculate DCs depending on the level of the party, if I want a tough task, I chose one DC, if I want an easy task, I choose another, same numbers, no matter the level. That makes my life as a DM much easier.


Those are houserules. While they may be reasonable, they do not belong in a discussion about a mechanic. WotC designers should be smart enough to identify there is a problem here (that you are solving with houserules) and implemented their own solution.

You see it as a problem, I don't, since I seem to interpret the rule about reasonableness of checks as overriding all other rules concerning checks. If it's not reasonable, I don't care what the check rules say. The mechanic is encompassed in more than just the specific rules on how to conduct a check. Part of the mechanic is determining if the check is necessary in the first place.


Actually when someone is making an attack roll at an impossible DC, they DO always roll. They don't get to fire across 3 continents because range gets capped based on the weapon. In fact it was for the exact reason you specified that auto success on a 20 was removed from skill checks in 3.5.

And this is why I think we will never reach an agreement on this discussion, because you seem to be saying the reason that a 3e character can't fire across a continent is because of the rules for weapon range, rather than the simple fact that such an action would be impossible. If there were no range rules for a bow in 3e, would you say the rules allow you to shoot across a continent?


Okay sorry, level 1 commoner not level 0 commoner. You will note however I didn't specify fighter. No fighter would have a -1 strength mod. Level 1 character was short for anything at all that is not high level. Because yes, higher levels do represent increased power and expertise.

Well yes, increased levels do represent increased power and expertise. My argument was that level 1 to level 20 is more akin to an E1 to an O8 or a rookie NFL player vs a 10 year veteran player, rather than Joe Couchpotato to the 10 year veteran.

As for level 1 commoner, I have two thoughts on this. One I think levels are only relevant within your class, just like someone with a PhD in mechanical engineering and someone with a PhD in Underwater Basket Weaving aren't directly comparable, neither is a level 20 fighter and a level 20 cleric.

The second is that I think D&D screwed up big time when they gave NPCs and monsters "levels", aside from the above, there's no reason why NPCs and monsters have to be generated the same way as PCs, or have the same stats, because they serve a different purpose. This incidentally is why I hope they take a page from Swords and Wizardry and drop the stats and associated saves for monsters and just give them a single Save value, with a potential modifier for a prime attribute.


Of course tiers would need to mean more than just the number jumps (because honestly in 4e that's all they were, and not even really good number jumps at that). For example effects like teleport and flight could be reserved for paragon tier. Effects like Simulacrum and Wish would be reserved for Epic. Basically each tier would unlock a whole slew of new modes of transportation, status effects, and defenses.


I think we're in agreement here. The problem is, a mechanic that works at really low levels breaks down at really high levels without a lot of extraneous math (a la 3e or 4e) that would be much better simplified by just acknowledging that different "tiers" of play have different rules and/or mechanics.


You are however defining "reasonable chance for failure" as "What does the DM think is reasonable?"

I am defining it as "what do the players and DM think is reasonable for the setting, world, story and current situation at hand". The key is that I expect DMs and players to actually talk to each other and behave like civilized adults. And if one side or the other can't do that, I expect the rest to stop playing with them until they can. There's no reason for the DM and the players to be adversarial to each other. If you and I can agree that arm wrestling a cripple is a silly thing to roll for, why should we need a rule that either a) tells us it's a silly thing to roll for or b) requires us to roll, even though the result is guaranteed?


Sure, because the Fighter can retry until he succeeds. Unless he fails by what... 10 or more? With a +3 on the roll, that's a very real possibility against a DC over 13. And yes, there is no take 10 roll because they gave it to the rogue as a class feature.


I don't see anything about failure by 10 or more in the rules, unless I'm missing it. The only thing I see is on page 4 under multiple checks, where again you are asked to be reasonable. Is it a task that allows multiple attempts (picking locks), then allow it, just be aware of the passage of time. If it doesn't allow multiple attempts (convincing the guard you work for the king) then disallow it.

I really have to ask at this point, and I don't mean to be rude, have you actually read through the play test rules, and actually read and parsed the rules; or have you just skimmed them, assumed things from other editions to be the norm or in the rules and dismissed them out of hand; or are you simply going off what you've heard second hand? I'll be honest, there's a lot of stuff I keep seeing pop up and I seriously wonder whether I'm reading the same play test document that everyone else is some times.


DC 20 challenges as the default for a difficult challenge? Jeeze!

Eh, I misremembered, but you're missing the forest for the trees. The point was that a DC of X always means the same thing. DC 20 is always so hard, DC 5 is always so easy.


And remember, you just got done saying how 20% chance of failing to arm wrestle the diseased guy with 1 strength was negligible enough to not bother rolling. So I guess we're doing the same thing here, so the party just auto fails at anything unless they have the correct skill trained

No, I said that you only had a 20% chance of failing if you needed to roll in the first place. The decision whether or not to roll is entirely separate from the chance of failure once you have decided to roll. By deciding to roll, you have said "for this given task, there is a significant (to the story or situation) chance of failure, and it is reasonable for the situation at hand to have an uncertain and randomized outcome, therefore we should roll to determine the outcome. We will roll using this mechanic as outlined in the rules, which provides us with chances of failure based on these criteria". The first step is making the decision to roll.

Or to put it another way, in D&D Next, you only roll when a random outcome will add something to the game, not every time you want to do something. Walking and chewing bubblegum doesn't require a DEX check.


But once again you're setting 'reasonable chance of failure' at "Whatever the DM feels like" which is absolute bull**** because every DM ever will rule these things differently. Seriously put a question of "Where is the reasonable line" to 10 different DMs kept apart, they'll come back to you with 15 different possibilities, with widely varying results. You are literally feeding Mearl's bull**** about how rules are bad because a DM will handle it with this line of argument.

And I am absolutely 100% ok with different DMs deciding their world is a little bit different. D&D is not a sport, it's not a competition or a contest. I have no stake in two different DMs worlds behaving identically. In fact, I personally think it would be rather dull if every single DM played everything exactly the same way. If I wanted a perfectly consistent experience across every DM, I would simply DM the game myself, or get a computer to DM for me.


Seriously DM fiat does not replace rules. If WotC is going to give us a rule set, it needs to be rules that can actually be followed not "Go with your gut feeling on it"

They are rules to be followed. The rule is "If you come across a scenario where you and the DM can not reach an agreement on what the outcome should be for the story and the environment, or where you think the result should be random, then you should roll for the result using these following mechanics. Otherwise, go with what fits."

I don't need or want a rule to tell me that a fighter will win an arm wrestling contest with a dead man. I don't want a rule to tell me that a thief can always open a pair of chinese finger cuffs. I don't need a rule that tells me how much damage a mountain deals when it falls on your head from space. I don't need a rule that tells me that arrows can't be fired across continents. And I certainly don't need a rule that tells me that if I think a level 20 character is a demigod with the strength of hulk, that a wooden door doesn't pose a challenge to him.

I do want rules for when I do decide that something should be a challenge, if I choose a DC of X type, it will be Y% challenging.


If a group has a problem with determining what is or is not reasonable, then this rule does not help them. If they do not have such a problem, then they don't need the rule in the first place.

If a group has trouble determining what is or is not reasonable, no rule will help them. If the rules say that it's reasonable for a wooden door to pose a challenge to a character, and the DM agrees with the rules, and players don't, unless the DM has been 100% strictly by the (numerical) rules with his game (and if you find such a DM, let me know so I can avoid them) then the DM has no more leg to stand on than if the rules didn't declare a DC one way or the other.


I could accept that as a standard rule, but that's the thing: it would need to be a standard rule. Because right now it's not, it's something you made up.

Sure, it would be nice to see it spelled out, but to be honest, it's to some degree already the same rule. Consider, if I have a DC that's 10, I succeed with an attribute of 15. Similarly, if I have a DC of 15, I succeed with an attribute of 20. Well if you view the opponent's roll as your DC, and the average on a d20 as 10 (or 11 if you round up), than a character with a 10 Attribute has an average DC of 10-11, and a character with a 15 attribute (+2) gives an average DC of 12 (13). It's not a precise match, but you do still beat the average DC with your 20 attribute.

Seerow
2012-06-05, 10:02 PM
Okay here's a question for you: You don't like rules. You say you don't want rules for a lot of things that are perfectly logical. Sure I don't need to know about a Fighter firing his bow across continents, but I do need to know where the line is drawn. Can he fire it 100 yards? 300? 1000? I would argue you need a rule to define that, and that rule by its definition prevents the cross continent shot. Similarly, I don't need to know the damage of a mountain falling from space, but a boulder dropped off a cliff? That's something that comes up. Now if a mountain happens to be dropped from space (I guess someone got hit by a meteor?) you can extrapolate that rock falling damage rule to fit with the mountain falling damage.



And this is why I think we will never reach an agreement on this discussion, because you seem to be saying the reason that a 3e character can't fire across a continent is because of the rules for weapon range, rather than the simple fact that such an action would be impossible. If there were no range rules for a bow in 3e, would you say the rules allow you to shoot across a continent?


I would ask why there were no rules for it. Why am I paying for a game that doesn't have rules for basic things in the game world. If I buy a bow I expect to know how far it will shoot, and not rely on GM rule of thumb to tell me exactly where it cuts off. This gets increasingly important as power levels leave human behind. Because yes, in 3.5 a character CAN shoot his bow to any point he can see with an epic feat. So he could fire his bow at the moon. Or with a high enough spot check or some divinations, conceivably fire at a continent on the other side of the world. Is this realistic? No! But that's the joy of epic play, you are leaving realism behind for something super human.

In fact, this is a great example of why DM fiat as a mechanic doesn't work. In 3.5 rules, I can in fact shoot my bow thousands of miles. If you were my DM, it wouldn't matter what level I was or how awesome I got, that shot would be too hard to make, because you think it's silly. Rules make it so everyone has a common foundation to approach the game from, rather than 5 people coming to the table with different ideas of what is acceptable at a given level of play.


I don't see anything about failure by 10 or more in the rules, unless I'm missing it. The only thing I see is on page 4 under multiple checks, where again you are asked to be reasonable. Is it a task that allows multiple attempts (picking locks), then allow it, just be aware of the passage of time. If it doesn't allow multiple attempts (convincing the guard you work for the king) then disallow it.


Page 3 of the DM's guide, Hazzards. If you fail by 10 or more on the check, you get a hazzard.

I guess that should answer your follow up question on whether I've read the playtest rules.




Okay, now that I'm done responding to that I'm getting back to my initial point, since all of the rest of your post just brings me back to it: You really seem to hate rules. You want a game where everything is extremely narrative, and apparently relatively low in power level. That's fine, you can totally have a game like that. But the best part is, you don't need to pay anyone for a rule book to play that game.

All you need to do is have everyone write down their archtype on a sheet, write down their 6 ability scores (rolled or not, whatever you prefer), and use ability checks to improvise everything that you don't think should be automatic. I mean, you don't need a rule book to tell you what spells your wizard can cast, you can figure out what's appropriate for yourself. You can decide if you think it's okay for the Fighter to wrestle that troll or not. If that's what you and your group like, that's great... but D&D is traditionally a rules heavy game, and that means yes, you are going to have rules telling you what you can and cannot do. And most importantly, you don't need to shell out 100 dollars on core books to make the game work exactly the way you like it.

1337 b4k4
2012-06-05, 10:54 PM
You don't like rules.

I do like rules, I hate legalese and rules lawyers, and I don't like having so many rules that the game devolves into competitions between char-op people, rules lawyers, and the players as a whole vs the dm. Equally, I don't like having so many rules that altering any one rule destroys the delicate balance of the system. I'd rather my rules be a lego set rather than a house of cards.


Sure I don't need to know about a Fighter firing his bow across continents, but I do need to know where the line is drawn. Can he fire it 100 yards? 300? 1000?

How far do you and your DM want to allow him to fire? But we have a misunderstanding here. I don't a rule with a given DC or max for something. What I mind is that the max or DC is so baked into the rest of the rules that if I wanted to change it, it would screw other things up to. I don't mind a rule that says a 2 ton boulder falling on you from 100 feet up will kill you, or do 100 HP of damage, what I do mind is a rule that says "Falling objects do damage at a rate of 5*weight in tons d (height/10)" because then someone eventually turns around and exploits that into something unreasonable, like saying that the half ton falling rock from 50 feet up only does 5d5 damage, so my 2nd level fighter will survive having a half ton rock dropped on his head every time, never mind that logically you wouldn't survive being hit dead on with a half ton rock landing on your head, it's in the rules. One is an example that one can adjust or ignore as needed for the world, the other is an exploitable mechanic that can break down, and while isn't in this case, may be tied to other mechanics that break if removed. See also healing surges, where if you don't like them, too bad because healing skills, undead abilities and so much more rely on them being in the system. By comparison, I see very little so far that relies on the HD system in D&D Next (with the obvious caveat that it may change, and get baked in just as badly).


In fact, this is a great example of why DM fiat as a mechanic doesn't work. In 3.5 rules, I can in fact shoot my bow thousands of miles. If you were my DM, it wouldn't matter what level I was or how awesome I got, that shot would be too hard to make, because you think it's silly. Rules make it so everyone has a common foundation to approach the game from, rather than 5 people coming to the table with different ideas of what is acceptable at a given level of play.

Not at all, if I were your DM, we would have already had a discussion about how epic this campaign and world could get. Whether you could or couldn't be that epic should have nothing to do with the rules. The rules should provide me with a framework to use to construct and support such epicness.


Page 3 of the DM's guide, Hazzards. If you fail by 10 or more on the check, you get a hazzard.

I guess that should answer your follow up question on whether I've read the playtest rules.


Fair enough, but hazzards don't prevent you from trying again per the rules, which was your original claim.


You really seem to hate rules.

Again, I don't hate rules. I hate rules that restrict, rather than allow for construction.


If that's what you and your group like, that's great... but D&D is traditionally a rules heavy game, and that means yes, you are going to have rules telling you what you can and cannot do. And most importantly, you don't need to shell out 100 dollars on core books to make the game work exactly the way you like it.

D&D may not be the most rules light system out there, but rules heavy it aint. D&D is a walk in the park (yes even 2e 3e and 4e) vs something like GURPs

But while I may not need to shell out money for books, sometimes I enjoy inspiration from those books. And sometimes they provide ideas and structure I don't have. I can also download labyrinth lord for free and play as much as my heat desires. I still bought the book.

I look at rules as a lego set. Sure, I can melt down some plastic and use a 3d printer to construct my own building blocks and do whatever the heck I want without paying a dime to lego. Or I can spend some money on some lego, build a model off the box, and then customize it into something else, maybe even something I didn't think of before.

The rules shouldn't be a legalese filled contract between me and my players.

Menteith
2012-06-05, 11:06 PM
I've been gaming with the same group since middle school, and we've all finished our degrees now. We're good friends, and we enjoy gaming together, but we still have disagreements about rules in RPGs. It's not that we're trying to undermine each other, or that we want to break the game, but it comes up. Sometimes an ability doesn't work the way you'd expect it to, sometimes two characters will work together in a way that's disgustingly strong, sometimes something just isn't covered by the rules.

It's useful to have a neutral source that we can go to when we do disagree. I'd rather have a well thought out, clearly stated ruling on everything in the game. If I disagree with a ruling I can still change it, but there's no harm in having mechanics clarified, and quite of bit of good in doing so. If a system has large gaps that require me to write significant amount of house rules for consistency in my worlds, that's not a good thing. Having rules doesn't prevent me from making my own if I have to, but a strong base covering most situations makes a system much stronger for me.

Ashdate
2012-06-05, 11:10 PM
I think rules should always be approached from the viewpoint of a new GM, one who may be reluctant to take the reins and adjudicate (this is pretty common in my experience). It's easy enough to say "leave it to the DM to figure it out" but that will only leave a new GM up for criticism when their decision is taken badly, without guidelines to help adjudicate those decisions.

The point shouldn't to have a rule for every situation, just one that the DM can turn to when the rules provided are murky (as they often can be in weird situations).

That we're having a discussion about when to use skill rolls and when not to is a perfect example of how a "bad call" can be made; really, the problem with using a d20 for success has always been a problem since at least 3e, but the protection of skill ranks/training made it more palatable I think. While I don't object to the idea of using a Stat modifier in place of all Str/Dex/etc. related skills, the system needs to (in my opinion) either adapt to the d20 or replace it. I won't repeat or get into the criticisms people have levied at it thus far, except to say that rather than leave it up the DM to determine when a skill check is necessary, model the rules instead to make "easy" tasks just that: easy to succeed in on a d20 roll. Others have done this math or something similar, but to repeat:

The DM guidelines in the playtest suggest a DC "10 or lower" should be a 'trivial" task. As many have pointed out, a DC 10 is not "trivial" when the chance of failure is higher than 25%. I would suggest that therefore, that if they want to simulate something being "trivial" that they assume that it should be something like a DC 5. This would mean that someone with a good ability score (say, a 14, or +2 bonus) would have a really good chance of succeeding. Having a good ability score and training would therefore mean you "would" auto-succeed by the rules; experience DMs (as well as players) can see the math, and new GMs can see how the math plays out.

A DC 8-10 would therefore become the new "moderate" challenge, where (as per the text) training would be required to reliably succeed. Therefore, having a +2 stat modifier would allow you to succeed 60% of the time, where training in the skill would bump that up to a 75% success rate. A good stat/training would likely allow for a player to auto-succeed. This would make sense in that even with zero training, it would become appropriate for a novice character to at least having roughly 50/50 odds of succeeding. It also would properly peg the challenge as being "worth" rolling for, and could become the standard default for a DM that is unsure whether a skill roll is appropriate or not.

For contested skill rolls, I think you could use the advantage/disadvantage system to greater effect. The player with the higher ability score has "advantage" on a contested skill roll (thus rolling twice). Admittedly, I don't know if this math works well, but it would help alleviate the problem of the 8 strength wizard beating an ogre in a contest of muscle, as well as allow for more opportunities to utilize the "advantage" mechanism.

tl/dr: You can have your cake and eat it too when it comes to creating rules that help new GMs, but don't slow down experienced ones.

Crow
2012-06-05, 11:18 PM
Just gotta say, I love how the 5E is Ending thread has a final comment noting how the discussion has "devolved" into a discussion about GNS theory.

As far as newbie GM, or Experienced GM: I would like to see the new edition look something like 2e in that you had a simple core chassis, along with loads of optional rules for those who wanted it. The 5e playtest for me seems like a good core chassis so far. i can't wait to see how they tweak it.

Seerow
2012-06-05, 11:27 PM
Just gotta say, I love how the 5E is Ending thread has a final comment noting how the discussion has "devolved" into a discussion about GNS theory.

As far as newbie GM, or Experienced GM: I would like to see the new edition look something like 2e in that you had a simple core chassis, along with loads of optional rules for those who wanted it. The 5e playtest for me seems like a good core chassis so far. i can't wait to see how they tweak it.

Honestly something like that can work, but they need to go farther still at simplifying things if that's the intent. Like Vancian Magic needs to come off the casters, with the casters being given a basic magic blast and being told to improvise everything else. They can then get their resource back via themes just like the martial characters. This sets everyone on even footing as the baseline, and makes it easier to sub different magic systems in and out as new books/modules are released.

1337 b4k4
2012-06-05, 11:30 PM
I think rules should always be approached from the viewpoint of a new GM, one who may be reluctant to take the reins and adjudicate (this is pretty common in my experience). It's easy enough to say "leave it to the DM to figure it out" but that will only leave a new GM up for criticism when their decision is taken badly, without guidelines to help adjudicate those decisions.


And there are guidelines. People seem to be taking objection to the fact that there's no explicit rule telling you that arm wrestling a cripple is not something you generally need to roll for. That seems like taking things a bit too far.

And ultimately, the only way to get better at adjudicating is to actually adjudicate. Better to be able to adjudicate with smaller things and get the hang of it before your players come up with some world breaking combo.


I won't repeat or get into the criticisms people have levied at it thus far, except to say that rather than leave it up the DM to determine when a skill check is necessary, model the rules instead to make "easy" tasks just that: easy to succeed in on a d20 roll.

And that's what it appears 5e is attempting to do. We can argue over whether the chosen DC ranges are appropriate, but that's not what the discussion has been about so far.

It's also worth noting that RAW, "An adventurer can almost always succeed automatically on a trivial task." which is anything DC 10 or lower.


Like Vancian Magic needs to come off the casters, with the casters being given a basic magic blast and being told to improvise everything else. They can then get their resource back via themes just like the martial characters. This sets everyone on even footing as the baseline, and makes it easier to sub different magic systems in and out as new books/modules are released.

Honestly, as much as I would like this myself, I don't see it happening. Vancian Magic is pretty much a as much a staple of D&D as the 3 basic classes are. I don't even particularly care for it (I actually like the M20 system), but even I was disappointed to find it missing from 4e (or more accurately, co-opted and warped). There's also the fact that magic in D&D even since the beginning occupied more space than just what a combat oriented "blast" would occupy. You need healing and control magic as well.

Still, it would be nice to see magical system as plugin modules rather than a baked in part of core, especially given how different some other magic user classes are in general.

erikun
2012-06-05, 11:44 PM
If I may weigh in on the current conversation:

I don't need a rulebook to tell me to get along with my friends. I can do that all on my own, thanks. Telling me to make a collective decision at the table also isn't necessary, because this is what happens when there's a question about the rules anyways. I can do that with any system, I do do that with any system, and I am certainly not paying $100+ for a system to tell me to do just that.

Rather, what I want to know is: How do certain mechanics interact with the system, and what are the real-world values of objects represented in the game?

I do not know how to play D&D Next. I haven't played it, and I certainly haven't played the completed version. I do not know what a "good" value is for a specific roll, nor do I know what a "bad" value is. As such, it is up to the game system to tell me so. If a situation is considered unreasonable and shouldn't be rolled for, then the system should simply tell me so - because I most certainly do not know what the system values mean. That is the key point. 1 Strength could me a cripple, someone who is barely capable of walking with their clothing on, or 1 Strength could mean a person 10% stronger than average. I have no way of determining the reasonability of a 1 Strength vs 20 Strength contest without looking at the numbers, and if I'm seeing only d20-4 vs d20+5, then it certainly does not appear to be an impossible contest with an obviously predetermined result.

The comparison only gets worse when we move away from the obvious physical situations and into the vaguely magical. What should when a 1st level Wizard with 20 INT casts an illusion at the 20th level Cleric with 4 INT and a defensive advantage? Should the 20th level character "obviously" succeed? Should the 20 INT attack "obviously" overcome the 4 INT defense? Should the defensive advantage "obviously" grant protection against the assault?

I would also want to know how rather mundane actions interact with the ruleset, so that the DM can properly adjudicate situations. If the rules say a longbow can fire 300 yards and gains a -1 penality for every fifth of max range, we can use that. If we decide that longbows really have a range of 1000 yards, we can still use the rest of the rules (-1 penality for every fifth of max range) and adjust accordingly. However, if the rules just say "longbows can be used to attack at a distance", then what can we use? Do we just say they are perfectly accurate until 1000 yard, and suddenly lose the ability to hit anything? Are we supposed to perform fieldtests on longbows to determine accuracy drops?

Remember, we're paying one hundred dollars, quite possibly each, for this ruleset. Unless the D&D Next ruleset possesses some miraculous preparation ability, simply putting together a game take up a large enough part of time. I don't have time to be testing/researching ever sword, bow, and piece of armor myself to come up with accurate data. I expect the system to do it for me. That's why I'm giving them my money. (Also, note that this applies to fantastical and magical equipment even moreso - I have no way of testing or studying the properities of an "Elven Feywood Farbow", and certainly have no way to determine what the designers had in mind.)

Fatebreaker
2012-06-06, 12:35 AM
If a group has trouble determining what is or is not reasonable, no rule will help them. If the rules say that it's reasonable for a wooden door to pose a challenge to a character, and the DM agrees with the rules, and players don't, unless the DM has been 100% strictly by the (numerical) rules with his game (and if you find such a DM, let me know so I can avoid them) then the DM has no more leg to stand on than if the rules didn't declare a DC one way or the other.

A group can have trouble determining what is or is not reasonable because the people themselves are not reasonable, yes; however, an otherwise reasonable group can also have trouble determining what is or is not reasonable because they have no basis from which to make a reasonable call.

Plenty of players are not blacksmiths, archers, locksmiths, swordsmen, cavalrymen, sailors, or otherwise in the know regarding skills or abilities which might be a) relevant during a game, b) observable in the real world, and c) easily modeled with elegant rules.

Telling these players that they should just "be reasonable" in no way helps them. Whether they are already reasonable (hooray!) or they are not (boo!), either way they lack knowledge which they were relying on the rules to provide or simulate. It's like telling someone who can't swim that the way to swim is to "drown less (http://penny-arcade.com/comic/2007/11/19)." It's good advice, but it's not actually helpful.

Rules do not have to be complicated to be effective, but bad rules will surely complicate the game.


And there are guidelines. People seem to be taking objection to the fact that there's no explicit rule telling you that arm wrestling a cripple is not something you generally need to roll for. That seems like taking things a bit too far.

No, people object to the fact that the rules allow for clearly ludicrous outcomes.

If Bob the Barbarian loses an arm-wrestling contest against Disease-Riddled Commoner on an infrequent but reliable basis, that's a degree of failure which says we should be rolling, but the scenario tells us we should not. So either the mechanic does not properly model the scenario as intended, which means the mechanic is broken, or it does, which means the game expects for Disease-Riddled Commoners to reliably (even if infrequently) beat healthy Barbarians in contests of strength, which is an odd direction for a heroic fantasy adventure game to go.

Do you see the conflict there?

darkelf
2012-06-06, 12:43 AM
So where is a cutoff point? When do you roll? Why can't WotC come up with clear criteria?

because it is not their job to come up with clear criteria in a game where literally anything can reasonably be expected to come up at the table. the game has a thinking, reasoning DM for a reason, let him or her do their damn job.

Menteith
2012-06-06, 12:49 AM
because it is not their job to come up with clear criteria in a game where literally anything can reasonably be expected to come up at the table. the game has a thinking, reasoning DM for a reason, let him or her do their damn job.

Speaking purely as a DM, it actually is their job to come up with rules for common circumstances. I'm fine with making rulings when I have to (what happens when Metafaculty is used on a Vecna-Blooded target or something), and I'll record them for consistency, but if a system demands that I house rule core mechanics to make them functional/realistic, then it's a bad system. If I want to make my own homeruled system, then I'd do that. But if I'm paying a significant amount of money to use this system, then it should be comprehensive, and cover a reasonable of situations.

EDIT

Fatebreaker said it better than me.

Seerow
2012-06-06, 12:50 AM
because it is not their job to come up with clear criteria in a game where literally anything can reasonably be expected to come up at the table. the game has a thinking, reasoning DM for a reason, let him or her do their damn job.

Why is it the DM's job to make rules and balance the game on the fly, when he has all the work of actually running the game to do? Especially when there is a team of people being paid for months or years to develop a system that is supposed to be providing the rules. I really don't get this insistence of putting everything on the DM rather than the game designers.

I know Mearls is pushing it hard, because if people buy into it it makes his job way easier, but why do we expect the DM to do so much? I mean it's this exact sort of attitude that keeps people away from DMing. So much responsibility gets placed on the DM that most players are afraid to take it. I loved 4e because it made the DM's job so much easier, people around my area who had previously refused to DM were running games, and doing so easily. Returning to the olden days where rules are archaic and the DM needs to decipher everything puts the onus for everything back onto him, and that's something most people simply don't want to deal with.

Craft (Cheese)
2012-06-06, 12:59 AM
Of course tiers would need to mean more than just the number jumps (because honestly in 4e that's all they were, and not even really good number jumps at that). For example effects like teleport and flight could be reserved for paragon tier. Effects like Simulacrum and Wish would be reserved for Epic. Basically each tier would unlock a whole slew of new modes of transportation, status effects, and defenses.

If you read through the core books, this is actually how the 4E tiers were supposed to work. In practice it didn't really work out because it was up to the DM to provide these things to players (the extremely limited nature of the ritual and power systems means the players had no way to get it themselves) and DMs would give these things to players whenever they felt it would be appropriate for the game, regardless of level.


Still, the intent was definitely there.

Seerow
2012-06-06, 01:05 AM
If you read through the core books, this is actually how the 4E tiers were supposed to work. In practice it didn't really work out because it was up to the DM to provide these things to players (the extremely limited nature of the ritual and power systems means the players had no way to get it themselves) and DMs would give these things to players whenever they felt it would be appropriate for the game, regardless of level.


Still, the intent was definitely there.

Right. Like I said, I agree with the design goal of 4e, I just acknowledge it fell really flat and left high level play feeling like more of the same. I would want the play to really feel different. Like people mentioned in the last thread how 4e actually expects all doors to be a level appropriate challenge, I'd make it clear how that's not the case and yes you should just breeze through most doors past a certain point, because doors are typically not an epic level challenge. A Epic Level dungeon crawl would be a completely different beast from a heroic level one, to the point where a heroic tier character being there might not even realize that he is in a dungeon crawl.

jaybird
2012-06-06, 01:07 AM
because it is not their job to come up with clear criteria in a game where literally anything can reasonably be expected to come up at the table. the game has a thinking, reasoning DM for a reason, let him or her do their damn job.

I buy books because I'm lazy and don't feel like making up my own "clear criteria". I think and reason because the PCs need a guy to tell them what's happening, or else the heroes of light become murder hobos. I certainly expect the book to tell me exactly how hard it is to break down an adamantium door, as well as provide general guidelines on how much adamantium is in that door should my players decide to take the door with them and sell it.

Craft (Cheese)
2012-06-06, 01:31 AM
Right. Like I said, I agree with the design goal of 4e, I just acknowledge it fell really flat and left high level play feeling like more of the same. I would want the play to really feel different. Like people mentioned in the last thread how 4e actually expects all doors to be a level appropriate challenge, I'd make it clear how that's not the case and yes you should just breeze through most doors past a certain point, because doors are typically not an epic level challenge. A Epic Level dungeon crawl would be a completely different beast from a heroic level one, to the point where a heroic tier character being there might not even realize that he is in a dungeon crawl.

Well, while I think it's a noble idea, I think there's one very critical point that needs to be solved before this goal can be implemented effectively: The Spore problem.

If you haven't played Spore, spore basically is 5 different games wrapped up into 1. As you guide your species past different stages of evolution, the gameplay completely changes.

What's the problem with this? The problem is that Spore is not a very good game. This is primarily for 2 reasons:

1. The included mini-games are so different as to practically be in different genres. Chances are, even if there's one or two of the 5 different games that you really like, there'll be at least one of them that you'll absolutely hate because it just won't appeal to your taste. Problem is you can't really ignore the one you hate, because you switch through them as the game progresses.

2. Because the different stages of evolution are so different, development attention had to be spread thin. Rather than putting all of their resources into making one really good experience, they had to make 5 lackluster ones.

Now, the first problem can be somewhat solved by getting together a group with the same preferences and just playing through a limited level range, but I don't see any clean way to avoid the second problem except to just multiply the design work.

Stubbazubba
2012-06-06, 01:44 AM
If an arm wrestle between a cripple and a level 20 Fighter comes down to 1d20-4 vs. 1d20+5, and that difference of 10 is supposed to be ruled out as 'no chance of failure' out of pure reason, then when the dragon with a +15 STR has that same Fighter pinned against the ground with a massive claw, should the Fighter get a chance to try and extricate himself? If so, then the "reasonableness" rule is inconsistent and needs to be standardized; the line has to be drawn somewhere. If not, then 20 levels is unnecessary and probably kind of boring, since what you can accomplish besides to-hit and damage capped out a long time ago. If D&D 5 is going to be that much more tame in terms of power escalation, I'd be OK with that, but a lot of people wouldn't, and it would need to come out and say it now before everyone gets their feelings hurt.

Either way, the "reasonableness" rule is unreliable because we have no frame of reference for how fast a dragon can move, or how wise one is, other than what the numbers and dice tell us, so those rules need to explicitly tell us where all the lines are drawn. I'm OK if it says, "If the difference in Ability modifiers is greater than 10, no roll is necessary," but it needs to say exactly that, not rely on "reasonableness." Because trying to trick a dragon with illusion magic is entirely unreasonable where I come from; there are neither dragons nor magic.

Kurald Galain
2012-06-06, 04:08 AM
I agree that it's WOTC's job to write down guidelines for common situations. Rare situations can always be decided by the DM, but I want a central rule for how to break a wooden door, or how many damage a falling rock does. This improves playing speed as well as internal consistency of the game world.


(edit) Also, I don't have a problem with a dragon pinning the wizard or cleric and giving him no chance at all to escape physically (unless e.g. they have MC'ed to rogue for an escape artist ability). This is still a team game, so your friends can bail you out; or perhaps you could use magic to escape.

Yora
2012-06-06, 04:11 AM
I've been gone for a week. Anything new and important happening in the meantime?

Kurald Galain
2012-06-06, 04:16 AM
I've been gone for a week. Anything new and important happening in the meantime?

Lots of people arguing, mostly. Well, many people have played the playtest now with highly mixed reactions. Pretty much every mechanic and design principle is loved by some people, hated by others.

king.com
2012-06-06, 04:55 AM
Lots of people arguing, mostly. Well, many people have played the playtest now with highly mixed reactions. Pretty much every mechanic and design principle is loved by some people, hated by others.

Its almost as if the design goal of bringing everyone together was a bad idea.....

prufock
2012-06-06, 06:47 AM
I have, to my disappointment, not been able to get my group together to try out the Next playtest material. I have nothing else to say, only expressing my displeasure.

1337 b4k4
2012-06-06, 07:00 AM
Apparently I'm not making myself clear. I have no problem with the game including examples and base rules for how difficult a given task normally is, especially relative to other tasks. That the 5e play test comes with a series of common DCs doesn't bother me in the slightest. What I do have a problem with is thinking that I need a rule to tell me that arm wrestling a dead guy shouldn't be rolled, or that no matter the relative charisma scores, the king is going to believe his long time friend and trusted advisor over a random PC who just walked in off the street and killed his daughter, without some astronomical proof of why the king should believe the PC.

I am perfectly OK with the core mechanics as outlined in the rules only being able to adjudicate a subset of all the possibilities in the world. I don't want or expect a mechanic that can perfectly model every possible aspect of every possible uncertain outcome in every possible situation. Never mind that such a mechanic doesn't exist, if it did, it would be so unwieldy and so damn complicated, no one would actually play it as designed, and every game and world and scenario would be just as inconsistent as if the rules didn't exist in the first place.

I just honestly can't imagine any situation that would come up where a check (as outlined in the play test) would be necessary, that can't be reasonably adjudicated. Obviously, no one here thinks it's reasonable that a dead person has any chance of wining an arm wrestling contest, so no check is necessary. When the dragon lands on your PC, the question is, does it seem reasonable that you PC has a non-0 chance of simply throwing the dragon off him? If not, no check needed. You don't need to have real world experience with dragons to do this, you just need to decide how powerful STR 10 is in your game world, and how powerful dragons are in your game world. Are your PCs reluctant farm boys or hercules? Are dragons mere flying lizards or practically gods unto themselves. The check rule isn't there to resolve all these situations, it's there to resolve situations where you've decided that both sides should have a chance of defeating the other.

Tehnar
2012-06-06, 07:27 AM
Apparently I'm not making myself clear. I have no problem with the game including examples and base rules for how difficult a given task normally is, especially relative to other tasks. That the 5e play test comes with a series of common DCs doesn't bother me in the slightest. What I do have a problem with is thinking that I need a rule to tell me that arm wrestling a dead guy shouldn't be rolled, or that no matter the relative charisma scores, the king is going to believe his long time friend and trusted advisor over a random PC who just walked in off the street and killed his daughter, without some astronomical proof of why the king should believe the PC.

I am perfectly OK with the core mechanics as outlined in the rules only being able to adjudicate a subset of all the possibilities in the world. I don't want or expect a mechanic that can perfectly model every possible aspect of every possible uncertain outcome in every possible situation. Never mind that such a mechanic doesn't exist, if it did, it would be so unwieldy and so damn complicated, no one would actually play it as designed, and every game and world and scenario would be just as inconsistent as if the rules didn't exist in the first place.

I just honestly can't imagine any situation that would come up where a check (as outlined in the play test) would be necessary, that can't be reasonably adjudicated. Obviously, no one here thinks it's reasonable that a dead person has any chance of wining an arm wrestling contest, so no check is necessary. When the dragon lands on your PC, the question is, does it seem reasonable that you PC has a non-0 chance of simply throwing the dragon off him? If not, no check needed. You don't need to have real world experience with dragons to do this, you just need to decide how powerful STR 10 is in your game world, and how powerful dragons are in your game world. Are your PCs reluctant farm boys or hercules? Are dragons mere flying lizards or practically gods unto themselves. The check rule isn't there to resolve all these situations, it's there to resolve situations where you've decided that both sides should have a chance of defeating the other.

I think we are pretty much in agreement what the core rules should do: resolve common situations that come up in the game and provide a guideline to resolve those situations that are not covered by the rules. I don't think anyone is disputing that.

What I am disputing is that the principle resolution of 5E, the single d20 ability check resolution is bad. Wrestling a dead man is a extreme example why that rule mechanic does not work.

Wouldn't it be better if they designed a mechanic that didn't require a DM to fiat things? A mechanic that prevented children from winning strength contests with ogres, or at least made it extremely unlikely?

The question still stands, where do you draw the line. Obviously a child shouldn't be able to wrestle an ogre, but should a fighter be able to? What about a fighter vs storm giant? If he can't wrestle with him, why can a monk damage him with unarmed strikes? These are all questions that can be answered with solid mechanics and not DM fiat.

Resolving things on a DM fiat is bad even with reasonable DM and reasonable players. I have been gaming with my current group for 10 years, that is to say we are past the point of any silly arguments. But at 2 am, after drinking 6 beers and playing DnD for 6-8 hours, mistakes start seeping in. My judgment is not what it was. If the PCs encounter a strange circumstance I often pull a resolution out of my ass, and often its not consistent with my resolution when the same PCs find themselves in the same situation two sessions down the line.

Now for most part that is ok, and a part of table top gaming, as long as it doesn't happen to often. If I have to pull resolutions out of my ass every hour while we play, something is wrong. Why did I pay money for a product if I have to come up with rules/mechanics/resolutions to play?


EDIT: A example of a system that does it right; Shadowrun 4E. A child will almost always lose to a adult human, who will almost always lose to a troll and who will almost always lose to a dragon in a tug of war. Assuming average statistics, and the almost always is >98%. There is no need for a DM fiat there. *

*Assuming you use straight STR+BOD contests. I am also not advocating Shadowrun as the best system evah; it has its problems but it does this pretty well.

Clawhound
2012-06-06, 09:13 AM
If one is starting RULES LIGHT, then it makes sense to begin with the most generic rules possible. Put the arbitration on the DM. The designers can then make more detailed rules later on.

Later on in the playtest, after they've sorted out feedback, they may (or may not) revisit the issue. That's the hazard of a playtest.

You can not cover all possibilities in a game system. Infinite variability yields infinite possibilities. If the generic rules can cover 80% of all possibilities in the adventure game, then they make sense enough. What you want is the fewest rules possible covering the most frequent challenges.

Of the remaining 20%, ad-hoc rulings should cover another 80% with few problems. Can you wrestle a bucket of water? How about a boulder? Can you wrestle a boulder? How about the wind? Those are all technically valid possibilities, but it's just not worth writing the rules for those rare circumstances.

Wrestle a dragon? Yeah, that's worth writing the rules for.

The rules do already account for some of this. Wrestle a corpse? Fine, the first combatant to 0 hp loses. Go. The corpse instantly loses because it's at 0 hp.

When a generic rules adequately covers a circumstance, you don't need specialized rules. Child wrestles a troll? Let's assume that the child wins 20 times, causing 40 hp of damage. The troll wins once, rending the child into mincemeat. Again, the rules have successfully modeled the encounter even with absurdly lopsided die rolls.

Seerow
2012-06-06, 09:50 AM
If one is starting RULES LIGHT, then it makes sense to begin with the most generic rules possible. Put the arbitration on the DM. The designers can then make more detailed rules later on.

Here's the big problem I have with this: The rules light thing isn't fairly distributed. We have mundanes who have to magic tea party literally everything ever, while Casters get a full subsystem and spells that take up a large chunk of the 'how to play' packet, right out of the gate in the 'rules light' version.

This doesn't indicate to me they're going for a rules light game where everyone starts off with ****, and you gain more complexity through modules. It indicates to me that they want to recreate 2e-3e style D&D. Except here stuff is even -worse- for the mundanes.

It doesn't even matter if they get a better skill system or a full resource system for non-casters later, or if casters get some other rules lite system as a module. Because what we're looking at are the default assumptions of the game, and those default assumptions aren't rules light. So no, we can't just handwave it all away as "it's a rules light game" UNLESS they take out Vancian Casting, which I have a very strong feeling they won't ever do.




The rules do already account for some of this. Wrestle a corpse? Fine, the first combatant to 0 hp loses. Go. The corpse instantly loses because it's at 0 hp.

When a generic rules adequately covers a circumstance, you don't need specialized rules. Child wrestles a troll? Let's assume that the child wins 20 times, causing 40 hp of damage. The troll wins once, rending the child into mincemeat. Again, the rules have successfully modeled the encounter even with absurdly lopsided die rolls.

Um, are there some grappling rules in the playtest packet I didn't see? Why are the rules based around HP and doing damage, rather than restraining the target? Or are you making stuff up, and passing it off as a reasonable rule despite the fact that no rule at all exists in the game?

Kerrin
2012-06-06, 09:55 AM
I wish WotC had done a little better job framing the playtest - explaining the boundaries, what's not included and why, etc. It doesn't feel like they did enough expectation setting with this playtest as well as proving rough roadmap information for where things will go from here.


On another note, I wonder what development process they are using.

Did they start with a particular edition and do adding, subtracting, and changing? Did they enumerate all the mechanics of each system then stitch together 5e making tweaks as they went? Did they read the volumes of commentary about each edition and start from there? Did they do an analysis of other RPG systems as the starting point?

There are so many approaches that can be used and it'd be really interesting to know how they've approached the project.

This is one of the meta topics I tackle when starting a new software project - we have to figure out what development process, practices, and tools to use that are appropriate to the project before we start developing a solution because they form the framework of how we do our work.

Clawhound
2012-06-06, 10:42 AM
How else should you playtest the base rules?

It seems to me that the designers aimed for the safest harbor of D&D and started the playtest with that. I don't blame them. Make sure that the foundation is working mostly as intended. Make sure that the feel is right.

Will there be other pieces? Surely. We know there are missing things. We know that they are still arguing and designing follow-on pieces. If they had plopped a fully formed game on us, then this whole iterative playtest process would be a sham.

They had to start somewhere. I accept that this is it. At this point, I don't judge the game by what is or isn't in it. I can only judge the game by the rules that they've shown.

Doug Lampert
2012-06-06, 11:04 AM
I have, to my disappointment, not been able to get my group together to try out the Next playtest material. I have nothing else to say, only expressing my displeasure.

I just told everyone I'd be running an extra session for playtest on Memorial Day afternoon and anyone who wanted to show up could.

That way it didn't interfere with the ongoing campaign. It worked fine, we got me and all six players.

I sugest picking an otherwise open time, I'll probably go with another test on the 4th Saturday of this month (there are regular activities on the first three saturdays, but number 4 is open). So we'll play, then break for dinner, then have our usual evening session.

Tyndmyr
2012-06-06, 11:22 AM
Here's the big problem I have with this: The rules light thing isn't fairly distributed. We have mundanes who have to magic tea party literally everything ever, while Casters get a full subsystem and spells that take up a large chunk of the 'how to play' packet, right out of the gate in the 'rules light' version....
Um, are there some grappling rules in the playtest packet I didn't see? Why are the rules based around HP and doing damage, rather than restraining the target? Or are you making stuff up, and passing it off as a reasonable rule despite the fact that no rule at all exists in the game?

The grappling didn't come up in my test, but the HP/doing damage thing did. There was much looking for flanking or other tactical options.

The rogue in particular felt a bit shafted, given the lack of clarity on skills.


I'm not disappointed in the edition overall, but after playtesting, I do feel like the initial playtest packet was not well put together. We had something like a page of feedback(typewritten, 8 pt font) before a single die got rolled.

1337 b4k4
2012-06-06, 12:07 PM
Here's the big problem I have with this: The rules light thing isn't fairly distributed. We have mundanes who have to magic tea party literally everything ever, while Casters get a full subsystem and spells that take up a large chunk of the 'how to play' packet, right out of the gate in the 'rules light' version.

You'd have less of a problem if you interpret "Rules Light" as "Rules Light 'Iconic' D&D". Vancian magic is one of the most basic and long term elements of D&D, which means that there will be spell lists. Heck, even M20, arguably the most "Rules Light" version of D&D still has spell lists, and expects you to rely on the SRD for details if you want them. The nature of Vancian Magic requires that it will take up a substantial chunk of the paper and rules. The only way they could really have gotten a lighter magic would have been to scrap Vancian and go with something like 4x5 magic.

TheAbstruseOne
2012-06-06, 12:25 PM
Actually, there are rules for grappling in the system as it stands. It's a Str vs. Str/Dex contest depending on whether the grapplee is trying to power out of the grapple or wriggle out of it.

As far as the casters taking up most of the rules, welcome to every edition of D&D ever aside from 4e. The ratio for each edition seems to be between 1/3 to 1/2 of the Player's Handbook being taken up by the chapters on the magic system and spell lists - not even counting all the other places rules for magic pop up, like in feats, magic item creation, skills, combat situations involving magic, rituals, etc. We have in 1st Edition AD&D 60 out of 130 pages, 2nd Edition 148 out of 320 pages, 3rd Edition 128 out of 275 pages, 3.5...th Edition 135 out of 322 pages, and even the more modern design philosophy in Pathfinder giving us 168 out of 575 pages (noting that the Pathfinder Core Rulebook is the PHB and DMG rolled into a single book).

I can't speak for any of the OD&D games as I don't own any of those books, though, but I'd bet that they spent just as much space on magic as well.The only edition I can think of where they didn't spend a large portion of the page count on magic and spells is 4th Edition, and that's because aside from Rituals, the spells were in the class descriptions just like the other class's At-Will/Encounter/Daily powers.

Seerow
2012-06-06, 12:26 PM
You'd have less of a problem if you interpret "Rules Light" as "Rules Light 'Iconic' D&D". Vancian magic is one of the most basic and long term elements of D&D, which means that there will be spell lists. Heck, even M20, arguably the most "Rules Light" version of D&D still has spell lists, and expects you to rely on the SRD for details if you want them. The nature of Vancian Magic requires that it will take up a substantial chunk of the paper and rules. The only way they could really have gotten a lighter magic would have been to scrap Vancian and go with something like 4x5 magic.

Sorry I don't buy into this. If Vancian Magic is a core rule, then we should also have a skill system, and a resource system for Mundanes as a core rule.

I have absolutely no interest in a game where the default assumption is people who don't cast spells are second rate characters. This was literally the biggest problem in 3e, and here, they're taking it a step backwards and making it even worse.

Right now, in 5e, the way any martial character does anything is ask the DM if that's reasonable, DM agrees or disagrees. If the DM agrees, he sets a check DC, and the Fighter rolls. In a true rules light system, everybody would follow this exact resolution mechanic. And yes, that means that a Wizard will have "Magic strike +6 2d6+5 damage" as his major class feature, and doing anything else he asks the DM, they work something out, and the Wizard rolls.

The problem with giving Magic Users a discrete subsystems and abilities to choose from and nothing to non-casting characters, is that they are in effect playing two different games. One set of characters is able to do cool things by default, the other depends entirely on how his DM feels. That is a huge ****ing discrepency that should not exist in core rules.

Yora
2012-06-06, 12:35 PM
Lots of people arguing, mostly. Well, many people have played the playtest now with highly mixed reactions. Pretty much every mechanic and design principle is loved by some people, hated by others.

I said new or important. :smallbiggrin:

kyoryu
2012-06-06, 12:39 PM
Its almost as if the design goal of bringing everyone together was a bad idea.....

Really, it might be.

Peoples' ideas of "D&D" are very different. The 1e grognard is *not* playing the same game as the 3.x high-powered gamer, is *not* playing the same game as the 4e player.

And I don't mean in trivial ways like rules. The structure and assumptions of the way they play are *very* different. I don't know if "one ruleset to rule them all" is an achievable goal.

Yora
2012-06-06, 12:47 PM
So what? From everything I've heard, the vast majority of people is quite happy how things are looking. And if a few don't, then they don't have to play it and can play any of ther other ten or so versions of D&D that are around.

TheAbstruseOne
2012-06-06, 12:53 PM
Sorry I don't buy into this. If Vancian Magic is a core rule, then we should also have a skill system, and a resource system for Mundanes as a core rule.

I have absolutely no interest in a game where the default assumption is people who don't cast spells are second rate characters. This was literally the biggest problem in 3e, and here, they're taking it a step backwards and making it even worse.

While Linear Fighters/Quadratic Wizards was a big problem with 1st through 3.5 and is something the developers should be keeping in mind, we can't really say if it's going to be a problem in D&D Next as of yet. Right now, we can only see five specific builds of four different classes up to level 3. We can extrapolate some idea of what the classes will be like, but we can't know for sure what the specifics will be later down the road.

From what the designers are saying in the Q&As, blogs, and on Twitter; it looks as though damage and HP are going to scale through levels but not AC/to-hit/etc. This means that the balance between casters and non-casters may very well be that non-casters are going to do damage more consistently, while casters are going to do more damage in big bursts but less damage outside that. So your big 3rd level Fireball may do 6d6+12 damage*, you can only do it on one round per 3rd level spell slot while my fighter can do 2d8+6 damage* every single round.

Something like this would curb the overpowering nature of casters at higher levels, but would bring back even more the 10 Minute Workday problem. Why should we push forward and have the wizard and cleric save all their big spells when we can just kick in the door, have them blast everything, loot the bodies, then let them take a nap to get those spells back? It's a tough problem to solve and the only time I can recall that anyone took a real stab at it was in 4e. Fourth edition "fixed" the problem by causing other ones in return (making non-casting classes just as complex as casters and putting a heavy emphasis on the encounter rather than the day in adventure design). And even that didn't really fix the problem because everyone could just blow their dailies in the first encounter then take an extended rest to get them all back. The Milestone mechanic was a good try, but it just didn't have enough weight to compete. Thankfully, the problem just didn't seem as widespread as it did back in the old 1st/2nd/3rd edition days.

Anyway, after all that rambling, my point got lost. Anything we say about how power will scale over levels and how classes are balanced is complete speculation because we just don't have enough data at this point to make a call. Until we see different builds at higher levels for pregenerated characters or the actual character generation rules, we're really just guessing.

* These damage numbers are complete guesses and meant for illustrative purposes only. I have no earthly idea what damage or anything else may look like beyond 3rd level in Next.

1337 b4k4
2012-06-06, 12:57 PM
Right now, in 5e, the way any martial character does anything is ask the DM if that's reasonable, DM agrees or disagrees. If the DM agrees, he sets a check DC, and the Fighter rolls. In a true rules light system, everybody would follow this exact resolution mechanic. And yes, that means that a Wizard will have "Magic strike +6 2d6+5 damage" as his major class feature, and doing anything else he asks the DM, they work something out, and the Wizard rolls.

And there would still be someone complaining about how giving wizards "Magic Strike" is giving them a super special skill that mundanes don't get. The only way to actually get what you want is to either provide every single class with a full listing of every single power and ability they can have a la 4e (which a lot of people, myself included, disliked, not the least of which because it makes character creation and leveling a slog), or do whittle the core down completely so that in core, every character is a mundane fighter, and make all skills and powers plugin modules. Basically, every character is fighter, and fighter is no longer a class, and your class (Wiz, Cleric, Barbarian, Rouge) becomes a module on top of that base template. Which again goes back to maybe they should just plain scrap the fighter.


Something like this would curb the overpowering nature of casters at higher levels, but would bring back even more the 10 Minute Workday problem. Why should we push forward and have the wizard and cleric save all their big spells when we can just kick in the door, have them blast everything, loot the bodies, then let them take a nap to get those spells back?

Honestly, TSR had this solved way back in 0e. Random encounter tables, and upkeep costs. Random encounters (especially rolled once every hour or so) used up resources, and prevented full rest, and upkeep costs, both for cost of living, and for paying your hirelings meant spending time going back to town after each encounter was a losing battle. Even with KotB which caves of chaos comes from, IIRC the caves are nearly 8 hrs travel from the keep, going back and forth to town after every battle eats up your resources, and players do need to rest or they will suffer penalties, so humping it to the caves for 8 hours and then immediately going into battle will incur lots of problems.

Seerow
2012-06-06, 01:05 PM
Limited uses per day as a balancing point has been tried. It doesn't work at high levels, because those limited uses become effectively unlimited uses, as no party is going to go on long enough for the Wizard to burn through 50 spell slots.

Also worth noting: They are using some daily abilities for the Fighter too. the 2/day bonus action is just going to make Fighters also want to rest after 1-2 fights, thus exacerbating the 5 minute work day issue.

And of course that completely ignores that what makes spells broken isn't the damage per round they can cause, but the versatility granted. Mundanes get stuck with a non-scaling ability check for the vast majority of things they might want to do. Wizards get access to a whole new tier of spells every other level, giving them new non combat abilities as well as better damage. I don't trust WotC to not release a lot of awesome non combat powers, or combat powers that are more useful than damage... and even if I did trust that and wizards did it, everybody would be upset at how limited magic was.

The core problem is Casters are getting lots of options. Mundane options are "Ask your DM if you can do that". In a real rules light system, you wouldn't have that dichotomy, both sets of characters would be stuck asking the DM if they can do something. Then as a module both sets would gain access to more discrete powers that could be used in and out of combat.


Instead, what we have is Casters with a full fledged spellcasting mechanic, Fighters who can't do anything more than attack, and the promise that in the future the Fighter will have the ability to trade out his Theme to gain access to special combat maneuvers. If the Martial Classes need to waste their themes to get a resource system, why do casters get them for free? Once again, it's an indication of the 5e developers not just falling into old habits of wanking to wizards and making fighters suck, but actually taking a step backwards from what we had before.

The only way this is actually an improvement is for people who thought fighters had too many options before, or people who genuinely think that raw damage is enough to make a character viable. By going with a core system that includes vancian magic but no-ability fighters, they are instituting a disparity from the get-go, in a system that is being advertised as being customizable to any style of game. Can you not see how that is wrong?

Menteith
2012-06-06, 01:06 PM
Is it unreasonable for the rules to cover general tactical maneuvers and what's allowable? Look at the rules for a Dirty Trick (http://www.d20pfsrd.com/gamemastering/combat#TOC-Dirty-Trick) in Pathfinder. It wouldn't be that difficult to create standardized rules for what's reasonable and what isn't.

@kyoryu

I agree wholeheartedly. It's more than just rules differences; there are fundamental assumptions about the game that differ drastically between editions. They've stated that they want people to be able to create characters using different systems and all play at the same table, which is a cool idea. But when my idea of a Lv20 is a minor god who's capable of destroying nations and yours is an experienced soldier, we have a disconnect that can't really be fixed through the rules. One of us is going to be disappointed.

@Yora

Enh, from what I've seen, it's been a mixed reaction. I haven't seen anyone who's just over the top gushing with praise right now (bearing in mind that I've been watching reactions via GitP forums and Enworld, not WotC's forums). I don't know how the modules will affect it, and I'm hoping that they do a lot to make it more attractive for me.

Seerow
2012-06-06, 01:11 PM
And there would still be someone complaining about how giving wizards "Magic Strike" is giving them a super special skill that mundanes don't get.

In this scenario though "Magic Strike" would basically just be a ranged attack, that the magic user could then improvise to do whatever they and the DM thinks is appropriate. If DMs prefer a lower magic game, they'll restrict the improv more, just like they're more likely to restrict the fighter from doing awesome stuff.



The only way to actually get what you want is to either provide every single class with a full listing of every single power and ability they can have a la 4e (which a lot of people, myself included, disliked, not the least of which because it makes character creation and leveling a slog)

How is it more of a slog than 3.5 character creation when playing any spell casting character? Or are you saying you want half the classes in the game to be simple because occasionally you can't be assed to pick powers? If so why not have the magic user class as the one that gets shafted on options? All the wizard players out there would hate it, that's why. Yet somehow people who prefer non-casters get told to suck it up and deal with it over and over again. **** that ****.

If your complaint is that every single class ever had its own ability list, I can get behind that. I liked the overlap of spell lists in 3.5. If martial characters had a shared list that they all drew from, where different martial characters overlapped with one another, that would be fine by me. But would this address your complaint of 'slog'? Or do you think it's a slog to have to make the choices at all, regardless of how the choices are shared among classes?

Because if your position is actually that half the classes in the game should get no options so that they're easier to stat up, I don't know what to say. Because that's about the most stupid and selfish thing I've heard on a topic that invites a lot of stupid and selfish comments.

TheAbstruseOne
2012-06-06, 01:25 PM
Honestly, TSR had this solved way back in 0e. Random encounter tables, and upkeep costs. Random encounters (especially rolled once every hour or so) used up resources, and prevented full rest, and upkeep costs, both for cost of living, and for paying your hirelings meant spending time going back to town after each encounter was a losing battle. Even with KotB which caves of chaos comes from, IIRC the caves are nearly 8 hrs travel from the keep, going back and forth to town after every battle eats up your resources, and players do need to rest or they will suffer penalties, so humping it to the caves for 8 hours and then immediately going into battle will incur lots of problems.

They never solved the problem really. I've been running a home-conversion of the D1-3 modules in Pathfinder using the same random encounter tables from the original module (with updated monsters) and they come up very rarely. Also, the group can easily take precautions against a monster randomly walking up to them and attacking. If they're in a dungeon or even a place like the Caves of Chaos, they could simply clear out one area of low level monsters that's isolated (like the kobold caves) and put in place traps and/or barricades to prevent any interruption in their sleep. When the higher level spells come into play like Rope Trick, Magical Hut, and all the various "here's a pocket of interdimensional space you can take a nap in" spells; that becomes even more worthless at a time it's needed the most. And even if they were effective, not all the random encounters were combat encounters on the older tables.

Then there's the backlash against random encounters. They add nothing to the story and they're just ways of eating up gametime. With a system like OD&D/1st/2nd Edition or what we've seen so far of Next, it's not a big deal because combat's pretty quick. But I've been running 3rd/4e/PF games for the past decade and change, so I've still got a knee-jerk reaction to random encounters as eating up 30-60 minutes of my game session. It left a bad taste in my mouth and I really don't like them. It also removes a level of control from me as DM in determining things like pacing and PC advancement that I really like having.

Really, the only sure-fire way to fix the 10 Minute Workday problem is to stop playing with metagaming munchkins. I don't think there's a rule in the book that will actually take care of the problem in an effective way. If your players are pulling tactics like that, they're not acting in-character because no one who chooses the adventuring life would approach it that way. Either talk to them and explain that all this is going to do is force them to fight bigger, badder monsters every single encounter in order to keep the game fun or find some other way to use Rule Zero to punish them for it.

TheAbstruseOne
2012-06-06, 01:29 PM
How is it more of a slog than 3.5 character creation when playing any spell casting character? Or are you saying you want half the classes in the game to be simple because occasionally you can't be assed to pick powers? If so why not have the magic user class as the one that gets shafted on options? All the wizard players out there would hate it, that's why. Yet somehow people who prefer non-casters get told to suck it up and deal with it over and over again. **** that ****.

If your complaint is that every single class ever had its own ability list, I can get behind that. I liked the overlap of spell lists in 3.5. If martial characters had a shared list that they all drew from, where different martial characters overlapped with one another, that would be fine by me. But would this address your complaint of 'slog'? Or do you think it's a slog to have to make the choices at all, regardless of how the choices are shared among classes?

Because if your position is actually that half the classes in the game should get no options so that they're easier to stat up, I don't know what to say. Because that's about the most stupid and selfish thing I've heard on a topic that invites a lot of stupid and selfish comments.

Wow, you've already seen the character creation rules that they haven't released outside internal playtesting to know what a fighter or rogue character of a different build type than the pregens will look like? Awesome! I have so many questions for you about how the various classes, themes, and backgrounds will interact with each other to produce different character build types!!

Unless, you know, you're making assumptions based on rules that no one outside WotC and maybe the friends and family playtest have seen and are getting very upset and aggressive over those rules you haven't even seen yet...

demigodus
2012-06-06, 01:32 PM
When the dragon lands on your PC, the question is, does it seem reasonable that you PC has a non-0 chance of simply throwing the dragon off him? If not, no check needed. You don't need to have real world experience with dragons to do this, you just need to decide how powerful STR 10 is in your game world, and how powerful dragons are in your game world. Are your PCs reluctant farm boys or hercules? Are dragons mere flying lizards or practically gods unto themselves. The check rule isn't there to resolve all these situations, it's there to resolve situations where you've decided that both sides should have a chance of defeating the other.

Reasonable to whom? What if the DM and the PC in question disagree? What if the DM in question, and PC in question, and the other 3 PCs at the table have 5 different ideas for what the PC's odds should be?

It isn't like your only options might be farmboy or hercules. You could pick anything in between. What if the PC asked to be a grappling master, and the DM agreed to it? Does "grappling master" sound like a reasonable description for a PC in a DM fiat run system? Now, what if when the DM heard the term he was imagining a PC who did well in the Olympics, while when the player said the term, he was envisioning a character that leaped onto a dragon's neck, grabbed it, and using pure skill grappled it unconscious? In one of those cases, the PC would auto-fail this check. In the other case, the PC would auto-succeed.

Now what if the PC actually gave that description of grappling dragons with ease, but then it turned out that the DM's idea of a dragon differed greatly from the player's idea of a dragon?

If you have 5 different people, they will have 5 different ideas of what is reasonable. Just saying "be reasonable" and expecting that to solve problems is a HORRIBLE system. You might as well just write that whatever the DM thinks is most convenient to advance his plot line will happen, and it will be just as accurate.

Menteith
2012-06-06, 01:32 PM
Really, the only sure-fire way to fix the 10 Minute Workday problem is to stop playing with metagaming munchkins. I don't think there's a rule in the book that will actually take care of the problem in an effective way. If your players are pulling tactics like that, they're not acting in-character because no one who chooses the adventuring life would approach it that way. Either talk to them and explain that all this is going to do is force them to fight bigger, badder monsters every single encounter in order to keep the game fun or find some other way to use Rule Zero to punish them for it.

I disagree with this. There are ways to pressure players in realistic ways that drain resources from players without having to resort to DM fiat. It's not the player's fault if they use the abilities of their characters in intelligent ways. Being a good DM means knowing what they're capable of, and presenting challenging encounters regardless.

I'll admit that Plane-Shifting to fast time places is a problem, though....

1337 b4k4
2012-06-06, 01:37 PM
In this scenario though "Magic Strike" would basically just be a ranged attack, that the magic user could then improvise to do whatever they and the DM thinks is appropriate. If DMs prefer a lower magic game, they'll restrict the improv more, just like they're more likely to restrict the fighter from doing awesome stuff.

But it's still the magic user getting a toy that the fighter doesn't. The BBEG can take away the fighter's sword and bow. How does he stop the mage from magicing up the place? I suppose you could require that the MU carry an implement (I TOLD YOU TO TAKE HIS STAFF!), but then I think we're just back to option 2, where every character at the base is a fighter, since the implement is essentially a bow with unlimited amo. It's not a bad option, but it would radically change how we view D&D classes.


How is it more of a slog than 3.5 character creation when playing any spell casting character?

Honestly, I thought 3.5 was a slog too.


Or are you saying you want half the classes in the game to be simple because occasionally you can't be assed to pick powers? If so why not have the magic user class as the one that gets shafted on options? All the wizard players out there would hate it, that's why.

I'm saying I want the option to do all of the above. I want a system where I can be a lazy SOB and not have to do calculus to level up my character and still be a viable character. I want the option to choose from spell lists, and choose from combat moves, and I want the option to be able to just build spells and moves out of basic materials. I have no dog in who has to start with a basic skill, and add on modules to get cool toys, the only thing I have interest in is that magic feels "magical". 4e, magic was not at all different in feel from any of the martial powers. Equally, there wasn't a different feel between the fighter dancing and weaving and slashing his way through the enemies, and the wizard just zapping them all with a magic missile.


Because if your position is actually that half the classes in the game should get no options so that they're easier to stat up, I don't know what to say. Because that's about the most stupid and selfish thing I've heard on a topic that invites a lot of stupid and selfish comments.

Not quite. I think all the classes should have a "quick mode" for creation and leveling. Let's face it, character creation and leveling is a slog, and nothing turns off a new player faster than handing them a stack of text books to read before they even get a chance to play. A simple quick overview of the classes and their distinguishing features, with a viable simple progression system is what I want. If you happen by Zack S' blog, check out his posts on his idea for "Type V". That's something like what I want.


Reasonable to whom? What if the DM and the PC in question disagree? What if the DM in question, and PC in question, and the other 3 PCs at the table have 5 different ideas for what the PC's odds should be?

It isn't like your only options might be farmboy or hercules. You could pick anything in between. What if the PC asked to be a grappling master, and the DM agreed to it? Does "grappling master" sound like a reasonable description for a PC in a DM fiat run system? Now, what if when the DM heard the term he was imagining a PC who did well in the Olympics, while when the player said the term, he was envisioning a character that leaped onto a dragon's neck, grabbed it, and using pure skill grappled it unconscious? In one of those cases, the PC would auto-fail this check. In the other case, the PC would auto-succeed.

Now what if the PC actually gave that description of grappling dragons with ease, but then it turned out that the DM's idea of a dragon differed greatly from the player's idea of a dragon?

If you have 5 different people, they will have 5 different ideas of what is reasonable. Just saying "be reasonable" and expecting that to solve problems is a HORRIBLE system. You might as well just write that whatever the DM thinks is most convenient to advance his plot line will happen, and it will be just as accurate.

Then you have 5 people playing 5 different games and even having explicit rules aren't going to solve that because rule 0 is that the DM can override any rule. That's why I keep saying everyone at the table needs to come to an agreement about the type of game being played ahead of time. Sorry, if your player says "I want to be a grapple master" and you just say yes without getting an idea for what the player had in mind, that's a mistake on your part as a DM. Your options at that point are either to negotiate with the player when your visions disagree (as reasonable adults do), give the player what they want for now, and talk to them about it after the game (as reasonable adults do), state that you are the DM and for now you are making a ruling, and you will discuss it with the player after the game (as reasonable adults do), let the player have their way, silently sulking over how "they ruined your game" and passive aggressively punishing them (as unreasonable DMs do), or you can run roughshod over your player, telling them to suck it, and have them storm out of your game in a huff (as unreasonable DMs and players do).

Do you people seriously sit down to game with your players and never discuss anything about the world or the power level of the game?

Doug Lampert
2012-06-06, 01:48 PM
Really, the only sure-fire way to fix the 10 Minute Workday problem is to stop playing with metagaming munchkins. I don't think there's a rule in the book that will actually take care of the problem in an effective way. If your players are pulling tactics like that, they're not acting in-character because no one who chooses the adventuring life would approach it that way. Either talk to them and explain that all this is going to do is force them to fight bigger, badder monsters every single encounter in order to keep the game fun or find some other way to use Rule Zero to punish them for it.

Huh? MY CHARACTERS want to live and save the town/city/country/continent/world, they do not want to be party number 23 that failed and died as an example for party 24.

It's totally in character and not metagaming at all to rest when you're obviously impared and a relatively short easy rest will fix this. I rest if I think it will help more than pushing on in real life all the time.

The fix is SIMPLE and EASY, it can be implemented almost trivially.

Long term resources don't come back overnight! Seriously, does ANYONE who believes that recovery is that fast ever been tired? Hint, overnight does not give 100% recovery to real people. Pulling back and resting is a negligable benefit, the adrenaline will wear off and you'll stiffen up and you'll probably perform WORSE the next day.

Rest for purposes of recovering spell slots, healing surges, or any other "long term" resource should START on the second or third day of rest, and then be fairly slow from that point on.

People do what works, that's not metagaming, that's playing my character, if you want my character to push on then DON'T have vital resources that are available for a full recharge anytime I take 6 hours off. Make recovering spell slots or serious fatigue or damage take MONTHS, and people won't back out and rest 6 hours at the drop of a hat.

But don't tell me playing a superintelligent wizard as if he were smart enough to notice the most basic things about how his own powers work is metagaming.

TheAbstruseOne
2012-06-06, 01:51 PM
I disagree with this. There are ways to pressure players in realistic ways that drain resources from players without having to resort to DM fiat. It's not the player's fault if they use the abilities of their characters in intelligent ways. Being a good DM means knowing what they're capable of, and presenting challenging encounters regardless.

I'll admit that Plane-Shifting to fast time places is a problem, though....

You've never played with die-hard rules lawyer munchkins have you? I wanted to run a traditional good-vs-evil, classic 1st Ed style Greyhawk campaign in Pathfinder. One of my players wanted to play a ninja. I said no. He spent the next two months trying to convince me to let him play a ninja. He dug up every oddball module and novel and interview with Gary Gygax he could find to try to justify me letting him play a ninja even after I said it didn't fit with the stye of the campaign.

This is just to give you an idea because he does this with anything that happens in the campaign that he doesn't like. He's tried to argue for third party feats that are incredibly broken, for items that don't exist in the game world (he wanted a friggin' revolver for his ranger), for skills to cover things that make no sense for them to cover because he had a bigger bonus. I swear if I ordered a pizza with sausage on it, he'd use his smartphone to look up a third-party 2nd edition netbook to prove that I should've ordered pepperoni.

That is the sort of player that Rule Zero and a mechanical way to prevent the 10 Minute Workday is needed for. Most reasonable players either won't play that way or they'll see quickly the sort of frustrations that can cause. But for some players, it's really the only option other than just threatening to kick them out of the group.

Seerow
2012-06-06, 01:51 PM
Wow, you've already seen the character creation rules that they haven't released outside internal playtesting to know what a fighter or rogue character of a different build type than the pregens will look like? Awesome! I have so many questions for you about how the various classes, themes, and backgrounds will interact with each other to produce different character build types!!

Unless, you know, you're making assumptions based on rules that no one outside WotC and maybe the friends and family playtest have seen and are getting very upset and aggressive over those rules you haven't even seen yet...

I'm extrapolating from the designer's past proven tendencies, the information given in the playtest packet, and the information from their blogs/columns. Yes, there is room for me to be wrong, but it seems more and more likely that I am right. And yes, being right about it does make me angry, because seeing a game regress in terms of design is something that bugs me a lot.

Could WotC somehow miraculously go against everything they've shown us thus far and pull a great system out of nowhere? Sure. I however won't be holding my breath.

TheAbstruseOne
2012-06-06, 02:01 PM
Rest for purposes of recovering spell slots, healing surges, or any other "long term" resource should START on the second or third day of rest, and then be fairly slow from that point on.

Except that's not remotely how the game works or has ever worked. And even if you did extend a "long rest" from one night to three, it still wouldn't change this cheesy tactic. To paraphrase a rather well-known webcomic, "Anyone have anything they want to do for the next three days?"

As to the non-quoted parts of your post, I stand by the position that the 10 Minute Workday is metagaming. It is taking advantage of rules from the system in a way your character in-game would not. No adventurer is going to kick down a door, point the wizard at the monsters like a bazooka, blast everything, then spend the next 23 hours 59 minutes sitting around with the doors locked waiting to be able to sleep again to regain spells. Maybe once or twice in a dire situation, when there are multiple strong enemies and no time limit, but not several times.

And just to preemptively counter another argument, by the time the players are of a level where the 10 Minute Workday really gets abused, no one's worrying about having enough gold to buy rations. No one's going to pull this schtick at 1st or 2nd level and probably not even at 3rd level. It's when they get to 5th level and higher when all those juicy insanely powerful area spells and rays start popping up with only 2-3 spell slots to hold them. By that time, they can sell that +1 magic dagger they found back at 2nd level for a half ton of trail rations.

TheAbstruseOne
2012-06-06, 02:08 PM
I'm extrapolating from the designer's past proven tendencies, the information given in the playtest packet, and the information from their blogs/columns. Yes, there is room for me to be wrong, but it seems more and more likely that I am right. And yes, being right about it does make me angry, because seeing a game regress in terms of design is something that bugs me a lot.

Could WotC somehow miraculously go against everything they've shown us thus far and pull a great system out of nowhere? Sure. I however won't be holding my breath.

You mean the designers who gave you 4th Edition which does exactly what you're saying you want Next to do by giving fighters/rogues/etc. just as many power options as wizards and clerics? Or the blogs/columns/Q&As which state that there will be options for the fighter to have 4e style combat maneuvers to push/pull/slide enemies in the advanced combat modules and that they're also testing a change to the fighter class to give them two themes instead of just one and encouraging playtesters dissatisfied with the lack of options for the fighter to add the defender ability from the Moradin cleric? Or the playtest material that proves you wrong already since the fighter build already can do something that no other class can do out of the five pregens in getting two attacks per round at 3rd level through cleave?

Fatebreaker
2012-06-06, 02:09 PM
Apparently I'm not making myself clear.

No, you are being clear. Understand that there is simply no level of clarity which will make your point suddenly become convincing.

Even if we leave out the issue that even reasonable people can have very legitimate disagreements about what, exactly, "reasonable" is, the mechanical aspects alone are deeply flawed.

The rules of the game act as a series of conflict-resolution mechanic. That's why we have them. When we encounter a scenario where rolling would be ridiculous, but the mechanics produce an effect whereby failure is a reasonably likely outcome, then we have a flawed mechanic, and since we're talking about a product which we pay money for, we have a very real financial investment in whether or not the mechanics are any good.

In a game which uses a d20 as a conflict-resolution generator, any action with a chance of success or failure equal to or greater than 5% is worth rolling for. If the game tells me that a Strength 20 Barbarian loses an arm wrestling contest with a Strength 1 Diseased Commoner 5% (or more!) of the time, then mechanically that's worth rolling for, even though we know that's blatantly ridiculous.

Out of 400 potential combinations on two d20's, 45 result in the Strength 1 Diseased Commoner winning, and 11 result in a draw. That's an 11.25% chance of winning, and a 2.75% chance of a draw, for a combined total of a 14% chance to not lose for the Diseased Commoner -- well worth rolling for!

Is it silly? Yes. Is that why people are objecting to the mechanic? Yes.

The answer is not to ignore the mechanic. The answer is to get a better mechanic.

Put another way, if I'm the Diseased Commoner, and you tell me that I can't arm-wrestle the Barbarian because the odds of me winning are just unreasonable, I'm going to be very pissed when another character gets to make a roll with a 14% or less chance of victory.

And again, this is before we even enter into whether or not reasonable people will always agree on what is reasonable, or whether or not our table of reasonable people even have the information necessary to make a reasonable ruling, or whether or not "wing it and hope you get lucky" is a mechanic worth paying real money-dollars for.


You'd have less of a problem if you interpret "Rules Light" as "Rules Light 'Iconic' D&D".

Slapping the phrase "iconic" on something doesn't make it good.


On another note, I wonder what development process they are using.

I would also be rather curious to have sat in on their development process. I'm sure it would have been fascinating.

I would especially have loved to hear any discussion where disagreement was either very strong or absent. I'm sure there were fascinating debates with disagreement on all sides which ended amicably, but I'm especially interested in cases where they encountered mutually exclusive design goals (and which ones they picked as winners vs. which ones they picked as losers), and cases where they didn't even question whether something should be changed.


I have absolutely no interest in a game where the default assumption is people who don't cast spells are second rate characters.

I have absolutely no interest in a game where the default assumption is people who don't cast spells are second rate characters and the game pretends that the two are equal.

I'm okay with a game saying "this is better than that," so long as the game actually says it. Pretending otherwise is dishonest and poor game design.


Peoples' ideas of "D&D" are very different. The 1e grognard is *not* playing the same game as the 3.x high-powered gamer, is *not* playing the same game as the 4e player.

This is so true it baffles me why anyone would believe that a unified system is even possible. Basic assumptions about the game have changed to such a degree across editions that they can't realistically exist at the same table because they're simply not the same game.

obryn
2012-06-06, 02:13 PM
(he wanted a friggin' revolver for his ranger)
Well, in fairness, that's not completely out of the question for Greyhawk. :smallwink: (See: Murlynd; also: Barrier Peaks)

-O

demigodus
2012-06-06, 02:18 PM
Then you have 5 people playing 5 different games and even having explicit rules aren't going to solve that because rule 0 is that the DM can override any rule.

Hyperbole are not going to convince anyone of your point.

Rule 0 does not exist so that the DM can use it at every opportunity things don't go his way. It CAN be used that way, yes, but a) that isn't reasonable b) that is horrible DM'ing. Having explicit rules does solve such problems, because when a rule covers a situation, you don't HAVE TO use rule 0 on it.

And no, that isn't 5 people playing 5 different games. That is having 5 people who's instincts are different.


That's why I keep saying everyone at the table needs to come to an agreement about the type of game being played ahead of time.

To the point of covering scenarios that even the game designers who spent months/years of paid time couldn't be bothered to? You say that character creation is a slog fest, and yet a discussion anywhere near substantial enough to make sure issues like this don't come up with either a) involve numbers that can then be simplified into such characters anyways b) be so long, they are a slog fast compared to what character creation would be in 3.5.

Without the game providing guidelines, you can either hope your instincts agree, or make them up yourself. The problem with instincts, is the game running on knee jerk reactions is, in my opinion a horrible game. I like to play a game where I know what my character can do. I don't like to play a game where I have to guess at my DM's instincts to figure out what I can do. Alternative is to go through a very long list of options asking if I can do them, what is the DC, etc. But that tends to annoy people very quickly, very understandably.


Sorry, if your player says "I want to be a grapple master" and you just say yes without getting an idea for what the player had in mind, that's a mistake on your part as a DM.

Or a mistake on the part of the system where I can't just look at the player's grapple mod, and tell exactly what the term means there.


Your options at that point are either to negotiate with the player when your visions disagree (as reasonable adults do)

You argue for a rules light system to avoid a slog fest in making a character. All I see here is multiplying the size of that slog fest, and shoving it straight into valuable game time.


give the player what they want for now, and talk to them about it after the game (as reasonable adults do), state that you are the DM and for now you are making a ruling, and you will discuss it with the player after the game (as reasonable adults do)

I'm not sure how a system that forces you to go with these options is considered a good system...


Do you people seriously sit down to game with your players and never discuss anything about the world or the power level of the game?

Out of curiosity, how do you discuss power levels with your players? 3.5 has terms like "level" "character-tier" "high/mid/low-op", etc. that make such discussions quick and simple. How would you do such a discussion in a rules light system? And remember, rules light is supposed to cut down on character creation time, so this power level discussion + character creation has to be faster then the 2~3 minutes you spend describing the power level for 3.5 rather accurately (if you know the system), followed by its character creation.

TheAbstruseOne
2012-06-06, 02:22 PM
Well, in fairness, that's not completely out of the question for Greyhawk. :smallwink: (See: Murlynd; also: Barrier Peaks)

-O

I treat those modules like I do the Matrix sequels and the Shadowrun game for the 360 - they don't exist in my world and no amount of arguing is going to make it so.

Frankly, I shoved the party in Ravenloft three sessions ago just because I got sick of their monkeying around (my "good-vs-evil traditional classic style game" ended up with a LN monk, CN cleric, LE wizard, NE rogue, NE ranger, TN sorcerer who acts Chaotic Stupid, and a CN barbarian who managed somehow to find the random obscure rulebook that put the broken-ass Monkey Grip feat back into Pathfinder...and that's after I talked four of them out of their original characters which were a drow sorcerer (when I'm running a Vault of the Drow conversion), a gunslinger, a ninja/assassin, and a necromancer). Greatest setting ever made for an annoyed DM with frustrations against the players to work off outside Paranoia.

Menteith
2012-06-06, 02:23 PM
That is the sort of player that Rule Zero and a mechanical way to prevent the 10 Minute Workday is needed for. Most reasonable players either won't play that way or they'll see quickly the sort of frustrations that can cause. But for some players, it's really the only option other than just threatening to kick them out of the group.

I've never had anyone that bad, but that's not an issue with the 10m day, that's an issue with an incredibly irritating munchkin. When someone says a 10m adventuring day, I think of DMs not taking things like Rope Trick or Teleportation into account, or thinking that a player will never retreat. If you don't take them into account for X/day abilities, then you end up with poorly balanced encounters. But it's not that difficult to account for ways to refresh X/day abilities, and I don't have an issue with them for a game' balance. S'all I'm saying.

Yes, a traditional dungeon crawl isn't an appropriate adventure for people with these capabilities. But if a Wizard knows his limits, why would he continue after he's exhausted himself? It's not metagaming, that's common sense for a character.

kyoryu
2012-06-06, 02:26 PM
Huh? MY CHARACTERS want to live and save the town/city/country/continent/world, they do not want to be party number 23 that failed and died as an example for party 24.

It's totally in character and not metagaming at all to rest when you're obviously impared and a relatively short easy rest will fix this. I rest if I think it will help more than pushing on in real life all the time.

The fix is SIMPLE and EASY, it can be implemented almost trivially.

Long term resources don't come back overnight! Seriously, does ANYONE who believes that recovery is that fast ever been tired? Hint, overnight does not give 100% recovery to real people. Pulling back and resting is a negligable benefit, the adrenaline will wear off and you'll stiffen up and you'll probably perform WORSE the next day.



The fundamental problem with the 10 minute workday is that in many games, "days" are useless resources. Think about how many times we've been told that "things happen at the speed of plot."

If the bad guys don't shore up their powers, and partially invaded castles don't rework and reinforce, and there's no timeframe, and there's no resources such as food to manage, then there's absolutely *no* reason not to have a ten minute workday.

Unfortunately, most of these aspects have been discarded because they're "not fun", without removing other daily resources.

So long as you can exchange a worthless resource (days) for a resource with value (spell slots), players will do so at every opportunity.

Random encounters are another possible fix, as they force the characters to fight with depleted resources without even really getting anywhere for it - all of hte risk, none of the reward.

Your solution works so long as "more days" has more of a cost than just "days".

Oracle_Hunter
2012-06-06, 02:30 PM
Random encounters are another possible fix, as they force the characters to fight with depleted resources without even really getting anywhere for it - all of hte risk, none of the reward.
This is the worst way to fix 15 minute work days.

Frame it this way:
Problem: My Players use up all their resources in the first encounter of the day and then rest.

DM Solution: Force my Players to have additional encounters so that they fight on diminished resources.

Player's Response: Rest more frequently to be at top efficiency for random encounters.

Menteith
2012-06-06, 02:41 PM
I try to find ways so that my players want to continue past their limits. Sometimes this is accomplished by having a strict timeline - the palace guards are distracted by the festival tonight, the demon lord's ritual will be complete at the end of the week, the curse upon you will take your life by the end of the month, etc; sometimes with variable environments - if the Lord of Pain is attacked, he'll retreat into his impregnable fortress, so you only have one chance, if the evil mage learns who you are he'll use Scry and Die tactics, meaning you can't be discovered; and sometimes it's by letting the characters provide their own motivations - your father's killer is but a day's travel away, you must leave now!

Random monster encounters seem like a poor way to do it.

Scowling Dragon
2012-06-06, 02:43 PM
Frame it this way:
Problem: My Players use up all their resources in the first encounter of the day and then rest.

DM Solution: Force my Players to have additional encounters so that they fight on diminished resources.

Player's Response: Rest more frequently to be at top efficiency for random encounters.

Games are more complex then that and a scenario like that would rarely play out.

kyoryu
2012-06-06, 02:51 PM
I try to find ways so that my players want to continue past their limits. Sometimes this is accomplished by having a strict timeline - the palace guards are distracted by the festival tonight, the demon lord's ritual will be complete at the end of the week, the curse upon you will take your life by the end of the month, etc; sometimes with variable environments - if the Lord of Pain is attacked, he'll retreat into his impregnable fortress, so you only have one chance, if the evil mage learns who you are he'll use Scry and Die tactics, meaning you can't be discovered; and sometimes it's by letting the characters provide their own motivations - your father's killer is but a day's travel away, you must leave now!

Random monster encounters seem like a poor way to do it.

I agree in general. The best way to handle the problem is to make "days" a valuable resource - which is everything you've described above.

Random encounters are the weakest way to solve the issue, and only even work when there's an overriding goal anyway, and especially if you've given "days" some amount of value in the first place.

TheAbstruseOne
2012-06-06, 03:04 PM
Most players will actually avoid the 10 Minute Workday by playing in a smart and effective way. And in many cases, the plot itself will impose limitations such as a time limit. But that's not always the case.

Also, the 10 Minute Workday problem isn't just the players saying "Oh, the wizard and cleric are out of spells, we have to rest" and the DM replying "NOOOO! You must keep fighting because I finally have you on the ropes and then my monsters will win!" It's also not the cast if the party goes into one room and meets a very strong monster, then goes into another room and meets a different very strong monster and, due to the difficulty of the two encounters, have already expended all their healing and spell resources for the day.

It's a specific tactic in which a party who has the abilities and resources to beat an encounter while still holding high level spells in reserve for later still insist on using all their highest level spells and daily abilities in one burst immediately in order to more easily clear a room and then immediately stop and rest (spending the 16 hours they have to wait to sleep for 8 hours taking security precautions against wandering monsters and the like) to recover those spells. Not because the encounter demands it because of its difficulty, but because it's easier and less dangerous than actually attempting to face encounter as it was designed.

Regardless of whether or not it is a sound tactic, it's game-breaking as it will either remove all challenge from the game or it will force the DM to make every single encounter a make-or-break deadly fight in order to maintain the challenge to the players. It's also cheesy, munchkiny, and it makes no sense in-character. No party of adventurers would ever spend a month going through a dungeon one room per day, spending 1 minute or less fighting what's in the room and then 23 hours and 59 minutes sitting around in a locked, barricaded, and alarmed room before heading to the next room.

Fatebreaker
2012-06-06, 03:09 PM
Games are more complex then that and a scenario like that would rarely play out.

Really? Why not? All you've done is given them an even greater incentive to be ready for anything at any moment. Random encounters encourage resource hoarders to hoard more resources.

If a group rests after every encounter, and your response is random encounters, then all you've done is given the group a non-plot-relevant encounter to rest after.

This is the beginning of a hilariously vicious cycle.


I agree in general. The best way to handle the problem is to make "days" a valuable resource - which is everything you've described above.

Random encounters are the weakest way to solve the issue, and only even work when there's an overriding goal anyway, and especially if you've given "days" some amount of value in the first place.

Very much agreed. If days are not a valuable resource, then they will be sacrificed in the name of resources which are valuable.

Menteith
2012-06-06, 03:12 PM
Regardless of whether or not it is a sound tactic, it's game-breaking as it will either remove all challenge from the game or it will force the DM to make every single encounter a make-or-break deadly fight in order to maintain the challenge to the players. It's also cheesy, munchkiny, and it makes no sense in-character. No party of adventurers would ever spend a month going through a dungeon one room per day, spending 1 minute or less fighting what's in the room and then 23 hours and 59 minutes sitting around in a locked, barricaded, and alarmed room before heading to the next room.

I think we just have very different playstyles. If there's literally no pressures on the players, it's not their fault for taking the safest, more effective tactics. I can honestly say that I've never had a problem with the strategy, since I will almost always have a dynamic environment that responds to the players, and in the odd situation I don't, it's really unlikely that the situation is a straight dungeon crawl (Do people actually do straight no plot dungeon crawls still?)

kyoryu
2012-06-06, 03:15 PM
Regardless of whether or not it is a sound tactic, it's game-breaking as it will either remove all challenge from the game or it will force the DM to make every single encounter a make-or-break deadly fight in order to maintain the challenge to the players. It's also cheesy, munchkiny, and it makes no sense in-character. No party of adventurers would ever spend a month going through a dungeon one room per day, spending 1 minute or less fighting what's in the room and then 23 hours and 59 minutes sitting around in a locked, barricaded, and alarmed room before heading to the next room.

Part of what's missing is also the reactivity of the world - those goblins that were killed (or whatever) aren't just a static encounter - they're part of an ecosystem. Other critters know them, and will be aware of the noises. Sitting in a room and waiting 24 hours is a good way to, eventually, bring down the whole "dungeon" on your head.

Scowling Dragon
2012-06-06, 03:45 PM
Really? Why not? All you've done is given them an even greater incentive to be ready for anything at any moment. Random encounters encourage resource hoarders to hoard more resources.

If a group rests after every encounter, and your response is random encounters, then all you've done is given the group a non-plot-relevant encounter to rest after.

This is the beginning of a hilariously vicious cycle.


Because its metagaming. No roleplayer would go "Its OK to rest in the Volcano of mount voldoom because the GM will not send anything after us". Most people try to avoid this stuff in my experience.

TheAbstruseOne
2012-06-06, 03:46 PM
I think we just have very different playstyles. If there's literally no pressures on the players, it's not their fault for taking the safest, more effective tactics. I can honestly say that I've never had a problem with the strategy, since I will almost always have a dynamic environment that responds to the players, and in the odd situation I don't, it's really unlikely that the situation is a straight dungeon crawl (Do people actually do straight no plot dungeon crawls still?)

Yes, they do. I don't run them very often, but I have run them before when I couldn't figure out a good story or was running short on time before the session.

However, in many of my games, there are times in which there is no time pressure on the players built into the plot. If they have to rush to defeat the evil whatever before it does whatever or gets the MacGuffin, sure. But sometimes there are long stretches in which either nothing is going on, or the players don't know there's something going on.

For example, my current game is a Pathfinder game where I'm running a conversion and modification of the old D1-3 Vault of the Drow module. It started with the players retaking a keep from a bunch of goblins who took it over (no time limit as they could route out the goblins whenever they felt like it). The goblins had been displaced from their home in the Underdark by a troglodyte army which was preparing for an invasion of the surface world (time limit - they only had so many days until the army marched and the battle started, or they only had so much time to accomplish tasks during the siege or the keep would fall). They found out that the troglodytes were being supplied and encouraged by a specific drow house and went into the Underdark to investigate or retaliate against them.

This is the plot they've currently been in since sometime in March in a weekly session. They've made it to the drow city, scouted around, then picked up on a side plot thread. They're under no time pressure whatsoever right now. They are monitoring the drow house through bribed spies so they'll know if they mobilize or otherwise pose a threat to the village. They don't know that the drow plans for the troglodytes was a test run for an invasion of their own (including a ritual to darken the skies) that's on a 15 year timetable. They also don't know the only pressing clock is political tensions between their "home base" village and a city that's several days' ride away. So they're justifiably taking their time as there's no story-based time limit. I could flat out tell them about the brewing tensions between the city and the village or I could have the drow army ready to invade in a few days, but either one of those would break the story ideas I have in mind for future plots.

So even though I'm telling an intricate and involved story with several narrative threads, there's not really any time pressure whatsoever on the PCs at the moment (well, kind of...I dumped them in Ravenloft to take a break from tromping around the Underdark and I haven't decided when I'm going to bring them back yet). If I didn't have a few players (including the wizard) who feel as strongly as I do about the 10 Minute Workday, the ranger's player and possibly the cleric would probably take advantage of it on the few times I do present them with more traditional dungeon crawls.

Craft (Cheese)
2012-06-06, 03:56 PM
If your players are pulling tactics like that, they're not acting in-character because no one who chooses the adventuring life would approach it that way.

Unless you're trying to roleplay someone with a death wish, it's perfectly reasonable to rest, regroup, and prepare for the next battle after every fight. So reasonable, the only time where it's even thinkable to do that is when you can't due to the circumstances. Which is how the old RPGs handled the problem: You couldn't regain your resources unless you went all the way back, and you'd fight a bazillion random encounters again anyway so the only way to advance further was to just play better.

The real problem is that daily resources just no longer make sense in the modern style of game.

kyoryu
2012-06-06, 03:59 PM
The real problem is that daily resources just no longer make sense in the modern style of game.

Lots of things don't make sense in the post-DragonLance game style. Many, if not most, of the "problems" we see in D&D is the impedance mismatch between a game designed around pre-DL assumptions, and games run with post-DL assumptions.

This is actually one of the things I *liked* about 4e - it worked better in the post-DL world. It's also one of the things I hope that 5e clarifies - what the "overall game structure" is *supposed* to be, and ensuring that the game mechanics support that structure.

Scowling Dragon
2012-06-06, 03:59 PM
The real problem is that daily resources just no longer make sense in the modern style of game.

Yes they do.

Menteith
2012-06-06, 04:07 PM
snip

I don't have a good answer for you, then. All I can say is that I've never been in a session or DMed a session where it was a gamebreaking issue. /shrug

TheAbstruseOne
2012-06-06, 04:07 PM
Lots of things don't make sense in the post-DragonLance game style. Many, if not most, of the "problems" we see in D&D is the impedance mismatch between a game designed around pre-DL assumptions, and games run with post-DL assumptions.

Sorry, not seeing where Dragonlance gets involved...could you explain it to me? I don't mean that in a rude or condescending way, I literally don't see where the Dragonlance campaign setting fits into the debate...

Fatebreaker
2012-06-06, 04:10 PM
Because its metagaming. No roleplayer would go "Its OK to rest in the Volcano of mount voldoom because the GM will not send anything after us". Most people try to avoid this stuff in my experience.

*shrug* Some roleplayers will legitimately argue that, as a professional adventurer, they will behave in the manner most conducive for both survival and profit, and of the two, survival is more important. If that means going nova, then escaping to an alternate dimension where they are can recover their expended resources, and then chipping away at a dungeon at their leisure, then that's what they'll do. If they have no incentive to go quickly, why take the risk?

You could just as easily argue that pressing onwards in spite of a) no need to so and b) depleted resources is metagaming, because it relies on thinking, "Let's bank our lives on the leniency and sense of fair-play of the Great Dungeon-Master in the Sky!" It presumes that things like "encounter levels" and "challenge ratings" are a thing, or that they'll always find just enough monsters to be challenging without actually being a danger, which is a fairly game-aware attitude to take, don't you think?

Either way, if you have a group which is already prone to resting after encounters, then yes, more encounters will only encourage them to rest more, because that's what they do -- rest after encounters. If you want to kill the 10-minute workday, you need to give a reason for the party to not indulge in it. The plot is a good place to start.

--

Shadowrun features the Renraku Arcology, a huge "dungeon" in the middle of Seattle controlled by a rogue A.I. and his machine-army. How is the military taking back the Arcology?

Ten minute workdays, man. Ten minute workdays.

Go in, chip away at a section. Claim it. Secure it. Rest, rearm, recover. It's not fast, it's not pretty, but it's the best hope for survival and victory. That kind of attitude makes total sense when it's your life on the line.

kyoryu
2012-06-06, 04:17 PM
Sorry, not seeing where Dragonlance gets involved...could you explain it to me? I don't mean that in a rude or condescending way, I literally don't see where the Dragonlance campaign setting fits into the debate...

Because it's not the campaign setting. It's the campaign.

DragonLance was the first set of adventures that really took a party from 1st level through high levels, with a set plot, to a finish. It even introduced plot armor - giving DMs guidelines on how to bring players back into the game if "dead".

It's also the first case that I'm aware of where the *plot* took precedence over the *world*. The plot of the first novels was paramount, and the world existed primarily to support it. This differs from previous settings where the setting existed first, and had stories *within* it.

In many ways, it is the prototype for the modern game, and broke away from many, many core assumptions of earlier play styles. That doesn't necessarily mean it's bad, just that the rules were originally not designed with that in mind.

Really, look at the number of debates even in these forums about whether characters should be allowed to die. In a heavily story-based game, it doesn't make sense - and so the rules *shouldn't allow it to happen*. In a more simulationist game (which pre-DL was), of *course* it makes sense. Nobody's a special snowflake. Some number of people that go exploring in dangerous areas *will die*.

Even things like association rules make more sense in a game where you might have multiple characters, each of which is presumed to be mortal. Bob wants to play a paladin, and you want to be an assassin? No problem! We'll do one group one week, another group at another time. Problem solved! These rules don't work if you assume that you will have exactly the same set of characters, and no others, throughout the whole campaign.

EDIT: To be clear, I'm not espousing the superiority of pre- or post-DL styles of gameplay. I'm stating that the rules need to be written with one style or the other in mind, and they need to support that style, and that style of game needs to be relatively explicit in the rules. Even since 1st edition, the rules have been focused on running combats and encounters, with comparatively little information on what it means to run a campaign.

Craft (Cheese)
2012-06-06, 04:20 PM
Sorry, not seeing where Dragonlance gets involved...could you explain it to me? I don't mean that in a rude or condescending way, I literally don't see where the Dragonlance campaign setting fits into the debate...

I don't know enough to tell you how Dragonlance fits in, but I was referring to the shift from encounters-as-taxes to encounters-as-puzzles. In the former, encounters are basically the same thing used over and over again, and your challenge is to find a strategy that lets you stay alive through as long of a slog as possible. While in the latter, the experience of fighting a monster is supposed to be interesting and exciting all by itself, and the challenge is figuring out how to "solve" the given encounter rather than a meta-encounter strategy.


Daily resources make sense in the former system because the challenge isn't getting through 1 Kobold with 3 Magic Missiles, it's getting through 100 Kobolds with just 3 Magic Missiles. If you got your Magic Missiles back after every kobold you slaughtered then the game would be trivial (which is why the 15-minute adventuring day is a problem).


EDIT: Aaand Ninja'd. His explanation is better than mine anyhow, so I defer you there.

darkelf
2012-06-06, 04:23 PM
I haven't seen anyone who's just over the top gushing with praise right now (bearing in mind that I've been watching reactions via GitP forums and Enworld, not WotC's forums).

you could count me in that camp. but its not fashionable and some fellow gamers are dreary buzzkills.

kyoryu
2012-06-06, 04:25 PM
you could count me in that camp. but its not fashionable and some fellow gamers are dreary buzzkills.

I wouldn't say I'm gushing, but I like where they're heading with the system. There's too many open questions to gush, just yet.

And these forums are very heavily 3.x-biased. Any deviation from 3.x is likely to be spat upon :smallbiggrin:. The only real way I'd expect to see mass gushing here is if 5e were essentially a revision of 3e.

Fatebreaker
2012-06-06, 04:38 PM
you could count me in that camp. but its not fashionable and some fellow gamers are dreary buzzkills.

It's a playtest. This is exactly the time to express dislike, exchange ideas for improvement, and provide feedback.

If you like something, say what you like. Maybe you found something which other people did not.

Petey7
2012-06-06, 04:44 PM
I really don't see the issues that some of you see. To use an example previously mentioned, a Str 1 Commoner arm-wrestling a Str 20 Fighter. Naturally it seems silly for the Commoner to win, so just say the fighter automatically wins. Even if the commoner does technically have a, what was it, 11% chance of winning when you consider nat 1s and nat 20s, its still easy to say that the commoner winning sounds rediculous. In fact, the possibility of the commoner winning by getting a nat 20 or the fighter getting a nat 1 is exactly why it says in the DMs guide to sometimes just say whether someone automatically succeeds or fails.

I think they need to go into more detail about the ability check system, but I'm not a big fan of the skill system from 3.X.

TheAbstruseOne
2012-06-06, 04:52 PM
Enh, from what I've seen, it's been a mixed reaction. I haven't seen anyone who's just over the top gushing with praise right now (bearing in mind that I've been watching reactions via GitP forums and Enworld, not WotC's forums). I don't know how the modules will affect it, and I'm hoping that they do a lot to make it more attractive for me.

I'm on WotC, Reddit, Enworld, RPGnet, and here. Pretty much every discussion I've seen in every...single...friggin'...thread...has gone pretty much like this one has. Or worse. I've had to explain how clerics cast spells in Next something like 15 or 16 times at this point (not counting the four or five times I had to explain it to the same person) because no matter how I stated it, he could not understand that they cast like sorcerers in 3.x did.

The reaction's pretty much what you'd expect, though. Some people like it but agree it's got kinks if not major problems that need to be worked out, some people don't like what they're seeing but are keeping an open mind, some people flat out can't stand it, and a very vocal minority is threadjacking every discussion they can to rant about how horrible it is and how D&D is dead and how Wizards of the Coast killed their puppy and so on and so forth. Frankly, this has been the most civil and informative discussion I've participated in the past two weeks.

demigodus
2012-06-06, 04:55 PM
I really don't see the issues that some of you see. To use an example previously mentioned, a Str 1 Commoner arm-wrestling a Str 20 Fighter. Naturally it seems silly for the Commoner to win, so just say the fighter automatically wins. Even if the commoner does technically have a, what was it, 11% chance of winning when you consider nat 1s and nat 20s, its still easy to say that the commoner winning sounds rediculous. In fact, the possibility of the commoner winning by getting a nat 20 or the fighter getting a nat 1 is exactly why it says in the DMs guide to sometimes just say whether someone automatically succeeds or fails.

I think they need to go into more detail about the ability check system, but I'm not a big fan of the skill system from 3.X.

that 11% success rate is without assuming nat 1s are auto-fails, or nat-20s are auto successes. The commoner could win without either of them rolling a nat 20 or a nat 1. So not really sure why you are bringing up nat 20's or nat 1's.

Also, you seem to be completely missing the point of the example. The point is that, the system as written, allows for ridiculous situations, and then tells you to rule 0 it if gives ridiculous results. Saying "use rule 0 when our rules suck" doesn't make the rules suck less. Not to mention, an 11% chance of success is a pretty decent chance. A lot of times you will be doing stuff that aren't obviously auto-fails, where your chances might be that low. Why is it fair to the commoner that he gets to auto-fail, but at a different time, someone with similar chances gets to make a check?

NichG
2012-06-06, 05:05 PM
The 10 minute workday speaks to me of a deeper problem in the design of these games, which has been true for as far as I can tell all editions of D&D.

I don't think I've run a single game session with four combat encounters in it in the last three years.

Combat encounters take a lot of time. Once initiative is rolled, I can expect to sink at least an hour into it. If its a non-trivial encounter, three hours. If I'm playing for 8 hours every week, do I want to spend my time doing something interesting, or fighting off that next random encounter? I could split up a single 'day' into multiple game sessions, which I'd do for an intensive dungeon crawl, but something like that is pretty rare.

So what that ends up meaning is that, combat-wise, my players can expect at most two combat encounters in a given day. And thats a reasonable thing that improves the quality of my game. If I ran the four encounters recommended by the rules and designed into the system balance, it might be more challenging but it would correspondingly be far more tedious.

I'll admit though, I like the idea of a resource-limitation subgame for certain situations and feels, and I want the system to support it. Here's a crazy thought that's going to be very unpopular: what if all spells had moderately expensive material components that you couldn't easily eschew within the rules? If every Fireball, every Fly, etc costs 500gp of loot, then they'll be cast lest often. Furthermore, it scales with level as loot value increases.

Petey7
2012-06-06, 05:10 PM
And did 3.5 not allow for the same situations. I remember the first time I played, very first day, as a halfling rouge with a strength of 6, I successfully grabbled the parties fighter (I started off as a minor antagonist who was supposed to join the party after the first fight we had) who had a strength of 18, and held him down for 5 rounds. Also, in 3.5 a nat 20 is an automatic hit and a nat 1 one an automtic failure on attack rolls, so technically one can shoot the moon with a crossbow 5% of the time, and a seasoned warrior will fail to hit the side of a barn that they are standing two feet from 5% of the time.

TheAbstruseOne
2012-06-06, 05:13 PM
that 11% success rate is without assuming nat 1s are auto-fails, or nat-20s are auto successes. The commoner could win without either of them rolling a nat 20 or a nat 1. So not really sure why you are bringing up nat 20's or nat 1's.

Also, you seem to be completely missing the point of the example. The point is that, the system as written, allows for ridiculous situations, and then tells you to rule 0 it if gives ridiculous results. Saying "use rule 0 when our rules suck" doesn't make the rules suck less. Not to mention, an 11% chance of success is a pretty decent chance. A lot of times you will be doing stuff that aren't obviously auto-fails, where your chances might be that low. Why is it fair to the commoner that he gets to auto-fail, but at a different time, someone with similar chances gets to make a check?
Any attempt to abstract rules is going to have ridiculous situations come up because they're rules that are attempting to emulate reality, not reality itself.

Under 3.x/PF, it is impossible for a 1st level character to kill a 20th level character no matter what the circumstances. My 1st level character could have the 20th level character tied up, conscious, and bare-ass naked. I could then walk up with a knife and attempt to kill the 20th level character by slitting his/her/its throat...and there is absolutely no iteration of the rules in which that action is physically possible. The 20th level character is going to have far too many HP and far too high of a Constitution save against a coup de gras for me to be able to kill him/her/it, and even if successful the best I get is an automatic critical hit.

Other fun examples from games I've actually played in: Jumping over 150 ft from a stand-still with no magic or supernatural abilities involved, falling from low orbit with no armor, parachute, or anything else to slow descent or absorb the impact and surviving, climbing up a 80 ft sheer wall in under 6 seconds and kill four people with a sword once at the top (again with no magic or supernatural abilities involved), shooting a sniper through his scope at a distance of 500 yards with a handgun twice, timing a fall through a helicopter's blades so that he made it through undamaged and landed on the cockpit, saying "I went that way" to a group of guards and having them all believe it.

That's why there is a living, breathing person behind the DM screen to make these sorts of judgement calls. If the rules covered every single situation and did so flawlessly, there would be no need for a DM/GM/Storyteller/Referee and we'd all just say what our characters did and consult charts to find out what happens like in the Legend of Drizzt board game.

PairO'Dice Lost
2012-06-06, 05:23 PM
Any attempt to abstract rules is going to have ridiculous situations come up because they're rules that are attempting to emulate reality, not reality itself.

Under 3.x/PF, it is impossible for a 1st level character to kill a 20th level character no matter what the circumstances. My 1st level character could have the 20th level character tied up, conscious, and bare-ass naked. I could then walk up with a knife and attempt to kill the 20th level character by slitting his/her/its throat...and there is absolutely no iteration of the rules in which that action is physically possible. The 20th level character is going to have far too many HP and far too high of a Constitution save against a coup de gras for me to be able to kill him/her/it, and even if successful the best I get is an automatic critical hit.

Other fun examples from games I've actually played in: Jumping over 150 ft from a stand-still with no magic or supernatural abilities involved, falling from low orbit with no armor, parachute, or anything else to slow descent or absorb the impact and surviving, climbing up a 80 ft sheer wall in under 6 seconds and kill four people with a sword once at the top (again with no magic or supernatural abilities involved), shooting a sniper through his scope at a distance of 500 yards with a handgun twice, timing a fall through a helicopter's blades so that he made it through undamaged and landed on the cockpit, saying "I went that way" to a group of guards and having them all believe it.

That's why there is a living, breathing person behind the DM screen to make these sorts of judgement calls. If the rules covered every single situation and did so flawlessly, there would be no need for a DM/GM/Storyteller/Referee and we'd all just say what our characters did and consult charts to find out what happens like in the Legend of Drizzt board game.

Note that in 3e, characters are only really "real life" characters up through level 6ish, then they hit superhero/mythical hero/demigod levels of power. Even in AD&D with its flatter power curve, when you stopped getting HD at 10th level and settled down to rule a kingdom (in theory, if you didn't just start a new game), you were more than mortal by level 9ish and were expected to kill gods in the mid teens.

That progression is well-known by now and kind of expected--of course a level 1 character can't kill a level 20 character with a dagger, any more than Joe Average Guy could expect to do a thing to harm Cú Chulainn or the Hulk. Whether you like it or not is another issue, but if you think of level 20 characters as just level 1 characters with bigger numbers when many Greek heroes were mid-level at best, of course you're going to be disappointed.

navar100
2012-06-06, 05:29 PM
Huh? MY CHARACTERS want to live and save the town/city/country/continent/world, they do not want to be party number 23 that failed and died as an example for party 24.

It's totally in character and not metagaming at all to rest when you're obviously impared and a relatively short easy rest will fix this. I rest if I think it will help more than pushing on in real life all the time.

The fix is SIMPLE and EASY, it can be implemented almost trivially.

Long term resources don't come back overnight! Seriously, does ANYONE who believes that recovery is that fast ever been tired? Hint, overnight does not give 100% recovery to real people. Pulling back and resting is a negligable benefit, the adrenaline will wear off and you'll stiffen up and you'll probably perform WORSE the next day.

Rest for purposes of recovering spell slots, healing surges, or any other "long term" resource should START on the second or third day of rest, and then be fairly slow from that point on.

People do what works, that's not metagaming, that's playing my character, if you want my character to push on then DON'T have vital resources that are available for a full recharge anytime I take 6 hours off. Make recovering spell slots or serious fatigue or damage take MONTHS, and people won't back out and rest 6 hours at the drop of a hat.

But don't tell me playing a superintelligent wizard as if he were smart enough to notice the most basic things about how his own powers work is metagaming.

Then that falls into territory of players never using their resources because it takes "forever" to get them back. Everything will be Elaine Bennis' spongeworthy problem because they can't be 100% sure they need to use their rare resource for something more important later.

The solution has already been found - encounter resources. Note to self ignoring my dislike of 4E for the moment, I do like the concept it presented of encounter powers. I preferred 3E's Tome Of Battle idea of having a recovery method so as not to be forced to spam a minor inefficient attack because you ran out of the good stuff, but if unlike in 4E (as my perception) the "at will" attack was at least subjectively decent enough, then if recovery within an encounter is too offensive then not having one will be alright I guess. 3E Spells > Tome of Battle Maneuvers but crusaders, warblades, and swordsages were definitely not complaining about it. Given 5E will give spellcasters back their Vancian spell slots, give the warriors encounter abilities. Spellcasters will have enough spell slots they won't run out, and the DM can easily have more than one fight a day to prevent Nova attacks except perhaps in the BBEG combats where they're necessary.

demigodus
2012-06-06, 05:30 PM
Under 3.x/PF, it is impossible for a 1st level character to kill a 20th level character in one hit no matter what the circumstances. My 1st level character could have the 20th level character tied up, conscious, and bare-ass naked. I could then walk up with a knife and attempt to kill the 20th level character by slitting his/her/its throat...and there is absolutely no iteration of the rules in which that action is physically possible. The 20th level character is going to have far too many HP and far too high of a Constitution save against a coup de gras for me to be able to kill him/her/it, and even if successful the best I get is an automatic critical hit.

fixed it for you. I'm actually fine with 20th level characters being tough enough to survive a quick slice across the throat. At least the first few times. They really should be that tough.


Other fun examples from games I've actually played in: Jumping over 150 ft from a stand-still with no magic or supernatural abilities involved falling from low orbit with no armor, parachute, or anything else to slow descent or absorb the impact and surviving, climbing up a 80 ft sheer wall in under 6 seconds and kill four people with a sword once at the top (again with no magic or supernatural abilities involved), shooting a sniper through his scope at a distance of 500 yards with a handgun twice, timing a fall through a helicopter's blades so that he made it through undamaged and landed on the cockpit, saying "I went that way" to a group of guards and having them all believe it.

So you are saying 20th level characters are walking, human shaped bottles of epic? I fail to see a problem with all of those except the helicopter thing, because in that case how fast you fall isn't a factor of your skill, but of gravity, so no matter how skilled you are you can't fall fast enough to pull that off.


That's why there is a living, breathing person behind the DM screen to make these sorts of judgement calls. If the rules covered every single situation and did so flawlessly, there would be no need for a DM/GM/Storyteller/Referee and we'd all just say what our characters did and consult charts to find out what happens like in the Legend of Drizzt board game.

bolded: doesn't really have much to do with the argument at hand. Deciding that people that want the rules to cover more then you do want them to cover every possible situation, is called a strawman argument. It is great for making the person you are arguing with look ridiculous. It is horrible for convincing your opponent of anything.

No, the DM is there to give you a setting, move the plot along, tell you what happened in the rest of the world, cover the sections that the rules intentionally leave vague, because a reasonable group could agree with a reasonable DM's interpretations. There are things however, where a reasonable group could still heavily disagree, because they do not have the appropriate experience to make proper judgements. (examples include anything involving magic, converting stats to abilities, any profession none of them are familiar with, just how epic should a lvl 20 character be, etc.)

Rulebooks exist to cover these situations. If a rulebook doesn't cover situations the players expect it to cover, it is a poorly written rulebook. Naturally, this means that the same rulebook might be poorly written for some groups, and excellent for others. The point of the commoner vs lvl 20 fighter argument is that, it indicates the system going too far to the rule 0 side of the Rule 0 vs Specific Rules scale for some people. Too far to either side of the scale is bad really, it is just different people draw the ideal point at different places. That doesn't mean anyone is wrong. Even if they disagree with you.

Am I making sense here? Does this explain why some of us take issue with the game's "only roll when reasonable" attitude?

Seerow
2012-06-06, 05:34 PM
And did 3.5 not allow for the same situations. I remember the first time I played, very first day, as a halfling rouge with a strength of 6, I successfully grabbled the parties fighter (I started off as a minor antagonist who was supposed to join the party after the first fight we had) who had a strength of 18, and held him down for 5 rounds. Also, in 3.5 a nat 20 is an automatic hit and a nat 1 one an automtic failure on attack rolls, so technically one can shoot the moon with a crossbow 5% of the time, and a seasoned warrior will fail to hit the side of a barn that they are standing two feet from 5% of the time.

You're right that at 1st level this could happen in 3.5. And yes, it was just as absurd in 3.5 when a Fighter lost to someone far weaker than him. But in 3.5 at least you could count on the fact that leveling would make such a thing impossible, in 5e it will always be possible. It's embarrasing enough for a 1st level Fighter to lose to a 8 str halfling in a grapple, it's way worse for a 20th level Fighter to do so.

Also, no you can't shoot the moon with a crossbow, because range increments tell you flat out you can't shoot past 10 range increments. You can't just say "I'm taking a -1000 to hit to shoot out 1000 range increments" you can take up to -10 to shoot out to 10x your range. That gives most people about a 1/5th mile shot with a pretty hefty penalty to hit, which seems reasonable enough to me.

And yes, crit fumbles are stupid.


The point is not that 3.5e is flawless, it's that 3.5e had its own flaws, and the changes being discussed take those flaws and make them worse. That isn't a recipe for a good game system.

TheAbstruseOne
2012-06-06, 05:45 PM
So you are saying 20th level characters are walking, human shaped bottles of epic? I fail to see a problem with all of those except the helicopter thing, because in that case how fast you fall isn't a factor of your skill, but of gravity, so no matter how skilled you are you can't fall fast enough to pull that off.
None of those were from 20th level characters. The D&D examples are 5th level or lower (with the running-up-a-wall-and-slaughtering-several-people one being 1st level), and the rest are from Shadowrun or World of Darkness and yes, in every case there was nothing supernatural or otherworldly going on except that one character happened to be a troll with some cyberware to boost his Body attribute and another had some cyberware to increase speed and reflexes. The WoD examples didn't use any disciplines or magic. They all just involved some insanely unlikely dice rolls that were covered completely under the rules as written for that system. And trust me, especially when it comes to Shadowrun, I've read all the rules and understand them very well.


No, the DM is there to give you a setting, move the plot along, tell you what happened in the rest of the world, cover the sections that the rules intentionally leave vague, because a reasonable group could agree with a reasonable DM's interpretations. There are things however, where a reasonable group could still heavily disagree, because they do not have the appropriate experience to make proper judgements. (examples include anything involving magic, converting stats to abilities, any profession none of them are familiar with, just how epic should a lvl 20 character be, etc.)

Rulebooks exist to cover these situations. If a rulebook doesn't cover situations the players expect it to cover, it is a poorly written rulebook. Naturally, this means that the same rulebook might be poorly written for some groups, and excellent for others. The point of the commoner vs lvl 20 fighter argument is that, it indicates the system going too far to the rule 0 side of the Rule 0 vs Specific Rules scale for some people. Too far to either side of the scale is bad really, it is just different people draw the ideal point at different places. That doesn't mean anyone is wrong. Even if they disagree with you.

Am I making sense here? Does this explain why some of us take issue with the game's "only roll when reasonable" attitude?
See, here's the problem. You're stating as fact something that is simply your opinion. It's your opinion that a DM's job is solely to be Basil Exposition for a the story and to not make any rules decisions ever but simply enforce the rules as written. Honestly, I'm not even sure if that's a roleplaying game anymore. If that's the type of game you want to play, there are dozens of well-designed MMORPGs or video games which would suit your tastes perfectly.

There is never going to be a tabletop RPG rules system that allows for both freedom of action and that also requires no judgement calls from a DM/GM/etc. They will be required at some point in time to determine that the rules as written do not correctly cover the current situation and create a new ruling because it is simply impossible to create a system in which every single situation that may come up will be covered.

Petey7
2012-06-06, 06:09 PM
See, here's the problem. You're stating as fact something that is simply your opinion. It's your opinion that a DM's job is solely to be Basil Exposition for a the story and to not make any rules decisions ever but simply enforce the rules as written. Honestly, I'm not even sure if that's a roleplaying game anymore. If that's the type of game you want to play, there are dozens of well-designed MMORPGs or video games which would suit your tastes perfectly.

There is never going to be a tabletop RPG rules system that allows for both freedom of action and that also requires no judgement calls from a DM/GM/etc. They will be required at some point in time to determine that the rules as written do not correctly cover the current situation and create a new ruling because it is simply impossible to create a system in which every single situation that may come up will be covered.

Not to mention that sometimes the rules seem to contradict themselves (especially in 3.5), and it is necessary for someone to decide which rule to follow.

I'll agree that two different reasonable people can disagree on how something should be ruled, but I don't see how 5e is much different than 3.5 in that respect. Yes, it does encourage DMs to make gut decisions more, but you still had to make gut decisions in every other version of the game. Just like with any other version, the DM gets to make the call, and if someone disagrees, they should bring it up later, like after the session for example.

Seerow
2012-06-06, 06:10 PM
Under 3.x/PF, it is impossible for a 1st level character to kill a 20th level character no matter what the circumstances. My 1st level character could have the 20th level character tied up, conscious, and bare-ass naked. I could then walk up with a knife and attempt to kill the 20th level character by slitting his/her/its throat...and there is absolutely no iteration of the rules in which that action is physically possible. The 20th level character is going to have far too many HP and far too high of a Constitution save against a coup de gras for me to be able to kill him/her/it, and even if successful the best I get is an automatic critical hit.


Not sure this is strictly true. Especially if the 20th level character really is naked. Coup De Gras fort save scales really quickly thanks to it being DC 10+damage dealt with you getting a free crit.

Level 1 Fighter with 18 strength and a Scythe is going to deal 8d4+24 damage, average of 44 damage. I can't see a naked unbuffed level 20 character surviving a DC54 fort save. Even most equipped and buffed characters probably can't manage that.

Though yes, with a dagger where it's instead 2d4+8 damage, your odds are significantly lower. You're looking at a DC 23 on average, so all level 20 characters at least have a chance of surviving. Though I don't think any can do it automatically unbuffed and naked (best case scenario is probably someone with a good fort save and 20 con, for a +17, so you still kill him by slitting his throat about 25% of the time.




...and that was a kind of random tangent. Back to the discussion I guess.

demigodus
2012-06-06, 06:14 PM
See, here's the problem. You're stating as fact something that is simply your opinion. It's your opinion that a DM's job is solely to be Basil Exposition for a the story and to not make any rules decisions ever but simply enforce the rules as written. Honestly, I'm not even sure if that's a roleplaying game anymore. If that's the type of game you want to play, there are dozens of well-designed MMORPGs or video games which would suit your tastes perfectly.

There is never going to be a tabletop RPG rules system that allows for both freedom of action and that also requires no judgement calls from a DM/GM/etc. They will be required at some point in time to determine that the rules as written do not correctly cover the current situation and create a new ruling because it is simply impossible to create a system in which every single situation that may come up will be covered.

let's see some of the things I said in the section you quoted

No, the DM is there to give you a setting, move the plot along, tell you what happened in the rest of the world, cover the sections that the rules intentionally leave vague...


...it indicates the system going too far to the rule 0 side of the Rule 0 vs Specific Rules scale for some people. Too far to either side of the scale is bad really...

Yes, I don't believe it is the DM's job to rewrite the rules midgame (although that really wasn't something being addressed here, but I will give you that lucky guess). However, no rule interpretations ever? How the heck did you read that from my post after I said that dead opposite of that?

Stubbazubba
2012-06-06, 10:57 PM
If the problem with the "reasonableness" rule is too sticky with physical examples, use an illusion to trick a dragon with a huge WIS modifier.

The level 1 Illusionist has a total of +5 to this illusion, for whatever reason, and the Dragon has a +15 to his save against illusions, however that is finally calculated. Is that unreasonable? How on earth are we to know how Illusion magic or Dragon's Wisdom is supposed to work? By the numbers and the dice, there's a decent chance that the Illusionist can get lucky and trick the Dragon, but then again, that's the exact same chance the commoner had to out-wrestle the level 20 Fighter. So where's the cut-off for when something's unreasonable? Is it always the same, based on the disparity in modifiers? Or will it change based on how easily the situation is imagined? In this case, since it's a magical illusion against a magical creature's insight, basically no DM would be able to rule it out with an appeal to what seems realistic, and most would probably allow the roll, not realizing that, strictly-speaking, the odds are just the same here as in that instance when you laughed at the level 1 Wizard who wanted to try his darndest to out-wrestle the level 20 Fighter NPC the first session. If it's left up to the DM, then magic-users will get a lot more leniency because "it's magic" is a great explanation for plausibility in a fantasy setting. Mundanes, on the other hand, are only likely to be attempting tasks that the DM can actually imagine, and thus reject. So this Rule 0 nonsense applied to anything is just one more slap in the face for mundanes. This is why DM fiat is not a fair or good game mechanic, because it will be applied differently to different players at the same table without the DM even realizing it, let alone those who actively abuse it.

That's besides the problem that Seerow pointed out, that magic-users will be getting more and more reality-warping options hard-wired in this D&D Lite Edition, while mundanes have to play Mother May I with the DM.

NichG
2012-06-06, 11:13 PM
Level is a distraction in this discussion. There's no reason that a Lv20 character should necessarily be better at anything than a Lv1 character except that which their level directly represents. You could have the aged, experienced veteran fighter whose strength was 18 back in the day but has come down to 12 from age penalties lose an arm-wrestling match with the young, inexperienced warrior with incredibly brute strength (still has his 18), and have that not only be a possibility but actually be the logical outcome.

The point of separating the Ability checks from level-based stuff is to make level progression not necessarily imply all-around power. The real thing that remains to be seen is whether that idea will be violated by other mechanics, such as casters getting spells that let them all be Str 18 when they want to be.

Knaight
2012-06-06, 11:29 PM
And did 3.5 not allow for the same situations. I remember the first time I played, very first day, as a halfling rouge with a strength of 6, I successfully grabbled the parties fighter (I started off as a minor antagonist who was supposed to join the party after the first fight we had) who had a strength of 18, and held him down for 5 rounds. Also, in 3.5 a nat 20 is an automatic hit and a nat 1 one an automtic failure on attack rolls, so technically one can shoot the moon with a crossbow 5% of the time, and a seasoned warrior will fail to hit the side of a barn that they are standing two feet from 5% of the time.

The presence of flaws in another game does not negate the fact that said flaws are a problem in the game being discussed.

Menteith
2012-06-06, 11:33 PM
Level is a distraction in this discussion. There's no reason that a Lv20 character should necessarily be better at anything than a Lv1 character except that which their level directly represents. You could have the aged, experienced veteran fighter whose strength was 18 back in the day but has come down to 12 from age penalties lose an arm-wrestling match with the young, inexperienced warrior with incredibly brute strength (still has his 18), and have that not only be a possibility but actually be the logical outcome.

The point of separating the Ability checks from level-based stuff is to make level progression not necessarily imply all-around power. The real thing that remains to be seen is whether that idea will be violated by other mechanics, such as casters getting spells that let them all be Str 18 when they want to be.

Age and level aren't linked. Sure, it's entirely possible that a level capped melee class will no other options other than attacking will be weaker than what they were back at level one. But that's....well, that's terrible. Level 20 characters are at minimum the most powerful example of humanity. They are the strongest and smartest. There is no reason why a maximum level Fighter should ever lose to a minimum level Fighter. Again, this is subjective, and I actually feel that level 20 should be far more than just a strong human.

Knaight
2012-06-06, 11:41 PM
Age and level aren't linked. Sure, it's entirely possible that a level capped melee class will no other options other than attacking will be weaker than what they were back at level one. But that's....well, that's terrible. Level 20 characters are at minimum the most powerful example of humanity. They are the strongest and smartest. There is no reason why a maximum level Fighter should ever lose to a minimum level Fighter. Again, this is subjective, and I actually feel that level 20 should be far more than just a strong human.

The point being made was that things don't need to go up in level, unless they are specifically connected to the character concept. The fighter is no more magically competent, this particular one is no stronger (but probably far more skilled), one who doesn't really do much in the way of archery might still be terrible at range, so on and so forth. The idea is the removal of omni competence, where a level 20 is only better than a level 1 at the specific things they have chosen to be better at.

This could work, but it requires WotC to be fairly competent. WotC's competence is likely to be a sticking point regardless.

Menteith
2012-06-06, 11:51 PM
The point being made was that things don't need to go up in level, unless they are specifically connected to the character concept. The fighter is no more magically competent, this particular one is no stronger (but probably far more skilled), one who doesn't really do much in the way of archery might still be terrible at range, so on and so forth. The idea is the removal of omni competence, where a level 20 is only better than a level 1 at the specific things they have chosen to be better at.

This could work, but it requires WotC to be fairly competent. WotC's competence is likely to be a sticking point regardless.

As that's already how it works in previous editions, I don't have a problem with the theory. A Fighter in 3.5 is probably less competent at stealth than a first level Rogue. But when you're being shown up at something pretty core to your class (physical strength on a class that is limited to purely melee attacks and physical damage, which is governed by strength in 5E) by someone who is so far beneath you they'll need to hit your body for an hour to kill you, then I have a problem with how the scaling works.

Knaight
2012-06-06, 11:57 PM
As that's already how it works in previous editions, I don't have a problem with the theory. A Fighter in 3.5 is probably less competent at stealth than a first level Rogue. But when you're being shown up at something pretty core to your class (physical strength on a class that is limited to purely melee attacks and physical damage, which is governed by strength in 5E) by someone who is so far beneath you they'll need to hit your body for an hour to kill you, then I have a problem with how the scaling works.

This gets into the question of how key to the class it really is. Sure, the higher level fighter may have a lower strength score - that doesn't mean that they won't effortlessly win a fight. Maybe they're far faster. Maybe they're just far better with their weapon, and can block everything coming and get past most defense. So on and so forth. Now, if the type of fighter is specifically supposed to be a strong one who can wrestle giant beasts, and they start losing? Then we have a problem, because anyone who's throwing trolls around really shouldn't be losing at arm wrestling to a 1st level character. However, strength really doesn't need to be core to all fighters, so it works out.

Menteith
2012-06-07, 12:22 AM
This gets into the question of how key to the class it really is. Sure, the higher level fighter may have a lower strength score - that doesn't mean that they won't effortlessly win a fight. Maybe they're far faster. Maybe they're just far better with their weapon, and can block everything coming and get past most defense. So on and so forth. Now, if the type of fighter is specifically supposed to be a strong one who can wrestle giant beasts, and they start losing? Then we have a problem, because anyone who's throwing trolls around really shouldn't be losing at arm wrestling to a 1st level character. However, strength really doesn't need to be core to all fighters, so it works out.

Except in the weird examples we've been looking at, the Fighter has had the maximum allowable Strength from what we can gather, and there's still a good chance they'll lose contested checks with random shmucks.

NichG
2012-06-07, 12:40 AM
Except in the weird examples we've been looking at, the Fighter has had the maximum allowable Strength from what we can gather, and there's still a good chance they'll lose contested checks with random shmucks.

Thats a variance issue with a d20 versus a ~8-20 stat range corresponding to a modifier range of about 6. That means that the die has a range three times larger than the biggest range between characters you'd expect to see. But saying its a Fighter isn't really relevant - the same should be true of a Wizard with an 18 Str, or a Commoner with an 18 Str or anyone with an 18 Str.

Another way to put it is that its a fundamental assumption of the system that the game consists of a series of comparisons between somewhat matched entities. No contest that occurs will be so mismatched that numerically one side is guaranteed to win. This means that its not a system about diseased old man versus Hercules, its a system about a guy who can lift 200lbs against a guy who can lift 150lbs, where many ineffable factors could come into play in the form of a random element that is larger than the absolute level of ability.

holywhippet
2012-06-07, 12:54 AM
I don't know if it's been covered in the other threads (and no way I'm going to search through them all), but have they explained why the two sample clerics have a different number of orisons known? My guess is that it's either to do with the God being worshiped or to do with the differing wisdom bonus.

Menteith
2012-06-07, 01:10 AM
This means that its not a system about diseased old man versus Hercules, its a system about a guy who can lift 200lbs against a guy who can lift 150lbs, where many ineffable factors could come into play in the form of a random element that is larger than the absolute level of ability.

Why would I pay money for a system that incapable of modeling realistic behavior outside of specialized situations? I can guarantee that large stat disparities will come up when monsters are brought into it, even if you're ok with just ignoring low strength instances. Assume that the Str Ratio between a Fighter and a Dragon (or whatever monster) is similar to that of a Fighter and a 1 Str commoner - why is this situation not well covered by the rules?

This is all applicable to most contested checks. Here are a few absurd situations that can come up with scaling this low;

- The world's best detective is going to be unable to realize that the crumb-covered child is lying about eating cookies if the dice want to be weird.

- Someone who's never see a rope before will tie up Houdini inescapably.

- The expert tracker will be unable to follow the footprints of a drunk orc in soft mud.

All of these can come up with the current contested system....and what's more, they'll come up a statisically significant amount of the time.

Faerieheart
2012-06-07, 01:47 AM
Joe the strong as a bull 1st level commoner that's been pulling plows and doing manual labor all his life has say 16 strength. He comes into contact with James the 20th level fighter who's has 16 strength as well. They are laughing and joking in a bar and decide to see who's stronger.

Now the 20th level fighter has been slaying dragons and killing trolls with a sword of fire, but his strength is the same as Joe's. He certainly doesn't have 20 levels of arm wrestling, and even if he was a skilled arm wrestler only a certain amount of skill applies to what is mostly a straight strength check.

Even in 3.5 that's all it would be. Arm wrestling isn't a grapple check, it's an opposed strength check. In 3.5 the commoner and fighter have an equal chance. Same here.

The two get in a sword fight however, or an actual grapple where skill really applies then it becomes more than a opposed strength check.

Why should the level 20 fighter automatically win an arm wrestling match with a 1st level commoner? It's strength vs. strength.

NichG
2012-06-07, 02:02 AM
Why would I pay money for a system that incapable of modeling realistic behavior outside of specialized situations?

D&D 3.5 doesn't model accidental death due to falling off of horses. It doesn't model the behavior of weather, or how long a certain quantity of fuel can burn, or the number of fragments produce when an object is destroyed, or any number of other behaviors. No system can or should try to model everything. Universal models are either complex or wrong. A better approach is local approximations of the desired behavior that are stitched together where one approximation begins to fail and another takes over.


I can guarantee that large stat disparities will come up when monsters are brought into it, even if you're ok with just ignoring low strength instances.

Not all systems support the idea that it is ever possible for a human to win an arm-wrestling match with a 60ft tall Titan. There are many campaigns where it is reasonable to say 'this particular act is not something that a human can succeed at'.

That said, this system can handle a 60ft tall Titan arm-wrestling with an exceptionally powerful mortal. It just depends where you cap the variances.

Basically, the way I interpret it is: the opposed d20 roll is a model for a subset of possible interactions. For any advantage greater than, say, a spread of 10 points, it is a bad model. The system explicitly recognizes this and says 'if one side has overwhelming advantage, they simply win'.

A concrete mechanic that says this explicitly would be something like:

- Compare the modifiers of side A and side B. If the difference is 10 or greater, the side with the advantage always succeeds. Otherwise, have each side roll 1d20 and add their modifier and compare.

However, this already means that everything has to be statted on the same scale, which is difficult.



Assume that the Str Ratio between a Fighter and a Dragon (or whatever monster) is similar to that of a Fighter and a 1 Str commoner - why is this situation not well covered by the rules?


The rules in this case say 'the monster wins, no roll'.



This is all applicable to most contested checks. Here are a few absurd situations that can come up with scaling this low;

- The world's best detective is going to be unable to realize that the crumb-covered child is lying about eating cookies if the dice want to be weird.

- Someone who's never see a rope before will tie up Houdini inescapably.

- The expert tracker will be unable to follow the footprints of a drunk orc in soft mud.

All of these can come up with the current contested system....and what's more, they'll come up a statisically significant amount of the time.

All of these are consequences of the variance being much larger than the maximum possible advantage. This suggests to me that, again, the system is not trying to describe any of these situations. The system is instead only using the dice to model cases where the two contesting parties are nearly equal. Any significant inequality means that another approximation takes over - the better one wins.

Conundrum
2012-06-07, 05:00 AM
As that's already how it works in previous editions, I don't have a problem with the theory. A Fighter in 3.5 is probably less competent at stealth than a first level Rogue. But when you're being shown up at something pretty core to your class (physical strength on a class that is limited to purely melee attacks and physical damage, which is governed by strength in 5E) by someone who is so far beneath you they'll need to hit your body for an hour to kill you, then I have a problem with how the scaling works.

That's not how it works in previous editions. Because of the 1/2 level scaling, a warrior gets better at Arcana just by levelling up. How does that make sense? Is he reading the Wizard's arcane tomes before bed each night?


Another way to put it is that its a fundamental assumption of the system that the game consists of a series of comparisons between somewhat matched entities. No contest that occurs will be so mismatched that numerically one side is guaranteed to win. This means that its not a system about diseased old man versus Hercules, its a system about a guy who can lift 200lbs against a guy who can lift 150lbs, where many ineffable factors could come into play in the form of a random element that is larger than the absolute level of ability.

^ this.

blackseven
2012-06-07, 05:30 AM
bolded: doesn't really have much to do with the argument at hand. Deciding that people that want the rules to cover more then you do want them to cover every possible situation, is called a strawman argument. It is great for making the person you are arguing with look ridiculous. It is horrible for convincing your opponent of anything.

People are arguing that a STR 1 cripple possibly beating a STR 16 adventurer at arm wrestling is some sort of absolute repudiation of the system. I would argue that, while it does show the math is wonky in extreme situations, the very situation presented is OUTSIDE THE SCOPE of what the D&D rules are expected to cover.

This harkens back to people complaining that it was impossible to run a verisimilitudinous economy based on 3.5 RAW. Sure, there were things that maybe should have been patched (given values for commoner's wage vs. common tools, for instance), but citing it as definitive proof of "failure" is, well, intellectually dishonest.

The basic problem with the math is the D20. The D20 is horrible for D&D, but it simultaneously DEFINES D&D.


That's not how it works in previous editions. Because of the 1/2 level scaling, a warrior gets better at Arcana just by levelling up. How does that make sense? Is he reading the Wizard's arcane tomes before bed each night?

This was ONLY an issue in 4th. No other edition featured the automatic level scale. (I don't mean this as an edition war statement. I LIKE fourth, and having some form of Auto scale was a good idea, I think.)

Tehnar
2012-06-07, 07:41 AM
If the problem with the "reasonableness" rule is too sticky with physical examples, use an illusion to trick a dragon with a huge WIS modifier.

The level 1 Illusionist has a total of +5 to this illusion, for whatever reason, and the Dragon has a +15 to his save against illusions, however that is finally calculated. Is that unreasonable? How on earth are we to know how Illusion magic or Dragon's Wisdom is supposed to work? By the numbers and the dice, there's a decent chance that the Illusionist can get lucky and trick the Dragon, but then again, that's the exact same chance the commoner had to out-wrestle the level 20 Fighter. So where's the cut-off for when something's unreasonable? Is it always the same, based on the disparity in modifiers? Or will it change based on how easily the situation is imagined? In this case, since it's a magical illusion against a magical creature's insight, basically no DM would be able to rule it out with an appeal to what seems realistic, and most would probably allow the roll, not realizing that, strictly-speaking, the odds are just the same here as in that instance when you laughed at the level 1 Wizard who wanted to try his darndest to out-wrestle the level 20 Fighter NPC the first session. If it's left up to the DM, then magic-users will get a lot more leniency because "it's magic" is a great explanation for plausibility in a fantasy setting. Mundanes, on the other hand, are only likely to be attempting tasks that the DM can actually imagine, and thus reject. So this Rule 0 nonsense applied to anything is just one more slap in the face for mundanes. This is why DM fiat is not a fair or good game mechanic, because it will be applied differently to different players at the same table without the DM even realizing it, let alone those who actively abuse it.

Well said.


Thats a variance issue with a d20 versus a ~8-20 stat range corresponding to a modifier range of about 6. That means that the die has a range three times larger than the biggest range between characters you'd expect to see. But saying its a Fighter isn't really relevant - the same should be true of a Wizard with an 18 Str, or a Commoner with an 18 Str or anyone with an 18 Str.

Another way to put it is that its a fundamental assumption of the system that the game consists of a series of comparisons between somewhat matched entities. No contest that occurs will be so mismatched that numerically one side is guaranteed to win. This means that its not a system about diseased old man versus Hercules, its a system about a guy who can lift 200lbs against a guy who can lift 150lbs, where many ineffable factors could come into play in the form of a random element that is larger than the absolute level of ability.

The problem is that you don't know what entities are matched. Sure for the STR example most people would agree that a cripple can't beat a ogre, but what about other situations? Like in Stubbazubba post, where do you draw the line? What difference in abilities will make one autosucceed, and why didn't the designers put it in the first place?


D&D 3.5 doesn't model accidental death due to falling off of horses. It doesn't model the behavior of weather, or how long a certain quantity of fuel can burn, or the number of fragments produce when an object is destroyed, or any number of other behaviors. No system can or should try to model everything. Universal models are either complex or wrong. A better approach is local approximations of the desired behavior that are stitched together where one approximation begins to fail and another takes over.

Actually in the real world universal models are quite simple. To model any nonrelativistic, non quantum interaction between any two or more objects all you need to start with are Newtons laws. That is not to say that for specific cases you need approximations to even have a chance of calculating things, but the basic "rules" or "laws" of the system are simple.

In my game, I am looking for a system that can model common things adventurers will encounter, with a good set of guidelines from which you can extrapolate for those situations that are not common.



Basically, the way I interpret it is: the opposed d20 roll is a model for a subset of possible interactions. For any advantage greater than, say, a spread of 10 points, it is a bad model. The system explicitly recognizes this and says 'if one side has overwhelming advantage, they simply win'.

A concrete mechanic that says this explicitly would be something like:

- Compare the modifiers of side A and side B. If the difference is 10 or greater, the side with the advantage always succeeds. Otherwise, have each side roll 1d20 and add their modifier and compare.

However, this already means that everything has to be statted on the same scale, which is difficult.

The rules in this case say 'the monster wins, no roll'.

All of these are consequences of the variance being much larger than the maximum possible advantage. This suggests to me that, again, the system is not trying to describe any of these situations. The system is instead only using the dice to model cases where the two contesting parties are nearly equal. Any significant inequality means that another approximation takes over - the better one wins.

You made a houserule there about, but such a thing is not currently in the system (if it will ever be). Apart from the fact that it is not a good houserule, it is a stopgag solution to the problem that the designers should have already noticed and provided a solution of their own. I mean when I buy a product I expect it to work, not that I have to fix it all the time.


People are arguing that a STR 1 cripple possibly beating a STR 16 adventurer at arm wrestling is some sort of absolute repudiation of the system. I would argue that, while it does show the math is wonky in extreme situations, the very situation presented is OUTSIDE THE SCOPE of what the D&D rules are expected to cover.

The basic problem with the math is the D20. The D20 is horrible for D&D, but it simultaneously DEFINES D&D.


The d20 is a problem when the difference between the best possible modifier and the worst possible modifier the PCs can achieve is 6 (or 9 in the case of skills). If it is possible to increase that disparity to 20+ then some of those problems go away.

The problem only exists when the modifiers creatures can have to the d20 roll are small compared to each other and to the span of values on the d20.

When that mechanic is the core resolution method of 5E we have a big ****ing problem.

Conundrum
2012-06-07, 08:47 AM
Out of interest, do you think it's a problem that someone with a +17 modifier to a skill would lose to someone with no modifier to the skill 10% of the time? That's hardly an insignificant chance, and yet the modifier difference is still massive.

Aidan305
2012-06-07, 08:53 AM
People are arguing that a STR 1 cripple possibly beating a STR 16 adventurer at arm wrestling is some sort of absolute repudiation of the system. I would argue that, while it does show the math is wonky in extreme situations, the very situation presented is OUTSIDE THE SCOPE of what the D&D rules are expected to cover.
Isn't that covered in the DM's guide under "Don't bother calling for checks"?

Tehnar
2012-06-07, 09:13 AM
Out of interest, do you think it's a problem that someone with a +17 modifier to a skill would lose to someone with no modifier to the skill 10% of the time? That's hardly an insignificant chance, and yet the modifier difference is still massive.

In a direct contests, the guy with a +17 modifier loses or ties with the no modifier guy 1.5% of the time, strictly losing 0.75% of the time. That seems acceptable to me, as a artifact of the model.

I don't see where you got your 10% figure.

huttj509
2012-06-07, 09:42 AM
That's not how it works in previous editions. Because of the 1/2 level scaling, a warrior gets better at Arcana just by levelling up. How does that make sense? Is he reading the Wizard's arcane tomes before bed each night?

I think it was intended to reflect "he's been around the block a few times. Seen things Joe Guard has never dreamed of (and killed them). Been around hundreds of spells and magic effects, and those are just the ones that were levied against him. He's picked up a few things about how magic works."

I've never studied plumbing, but I've picked up a few things over the years. I bet you have too. Not necessarily anything involved or complex, but even "oh, the chain came off the toilet handle" is more than you knew in the past. How'd you learn it? Probably through experience, watching someone else fix it, and such, as opposed to sitting down and specifically studying the workings and mechanics of toilets.

Menteith
2012-06-07, 09:54 AM
And I'm not asking for across the board scaling. I think it's an interesting idea, and wouldn't be adverse to it, but it's not something that I'm used to (as it's specific to 4E). I'm talking about scaling within the areas your character is trying to specialize in. A Ranger should be able to track two halflings across an incredibly disturbed battlefield better than a drunk who wandered over and did a better job. Tyrion Lannister should be better at Diplomacy than Gregor Clegane, regardless of how they roll. Batman can hold his breath longer than random dudes off the street. Except that's not true inherently true under this system - and if you're comfortable with the system abnormally modeling things a large amount of the time, great. But I'm not really happy about having to house rule a system for common situations.

Stubbazubba
2012-06-07, 11:27 AM
It's as if they half went with this idea (http://www.wizards.com/DnD/Article.aspx?x=dnd/4ll/20110816) but didn't really finish it and then didn't tell anyone about it, either. Which is the worst of both worlds. The system described there last August would achieve what they're going for with this "reasonableness" thing easier and more transparently. The lines are clearly drawn, you can look at one item on your character sheet to know if you can, in fact, attempt something, and we have a hard-wired way of knowing when it applies. Now, there are also a lot of problems with that system, but ambiguity and putting the responsibility to make the game make sense on the DM are not among them.

Now, the other side of this issue is people's approach to the game. By the RAW, D&D has included very, very superhuman characters as you enter higher levels, in pretty much every edition. Nevertheless, many people don't realize that this is the case, and are under the impression that the rules are actually there to model fantasy action movies where characters push the limits of human capability, but certainly never break them outright, much less re-write whatever other rules of physics they want. The same system cannot work for both playstyles.

However, D&D Next can and ought to have rules for capping at a certain power level, whereafter attack bonus and AC cease to scale, spells, damage and HP continue to scale at increasingly lower rates, and what keeps progressing normally is just your Feats. Or something along those lines. Then define levels 1-6 as heroic, 7-12 as legendary, and 13+ as epic, and let DMs define what kind of game they are starting, i.e. "D&D (Level 1 to Legendary) LFM, inquire within." In the book, make it clear that the stories you can tell in these different tiers are pretty different, so you need to decide which feel you want, and you need to be clear with your players from the very beginning about it. As others have mentioned, both the numbers and the number of options need to increase significantly between each tier.

This, combined with a more transparently bounded skill system, will fix the issues we're going back and forth about here. I would regain so much respect for WotC if they did something like this.

kyoryu
2012-06-07, 12:31 PM
Tyrion Lannister should be better at Diplomacy than Gregor Clegane, regardless of how they roll.

I'm just going to deal with this example.

Is Tyrion "better" at Diplomacy? Sure. Does that mean he'll get a better result in *all situations*?

No.

Certainly, he will get a better result the vast majority of the time. But will he really get a better result when dealing with mercenaries? Maybe, maybe not. Is it possible that he's had an off day, and says something offensive? Of course. Is it possible that Gregor just happens to know some background on someone that Tyrion doesn't, allowing his to be more effective at building a friendship? Of course. Is it possible that Gregor has a common friend with the Diplomacy target that Tyrion doesn't? Yup. Heck, maybe the person being talked to is a woman that's attracted to Gregor, or someone just really impressed with his size.

Could we model all of that with situational modifiers? Of course! But that seems like a lot of hassle, and I don't think the results would be all that different than randomness. So I'll go with the simpler solution.

Of course, there's a big requirement for this kind of thing to work. And that's whether dice are rolled pre- or post- description. For many people, playing D&D is about describing what you do, and then rolling the dice to determine what happens afterwards.

For this kind of abstraction to work, you need to use a pre-description kind of thing - that is, you state your general intent, roll the dice, and then narrate what happened to achieve that result. It's a bit of an inversion, but I usually like the feel of it better, anyway.


It's as if they half went with this idea (http://www.wizards.com/DnD/Article.aspx?x=dnd/4ll/20110816) but didn't really finish it and then didn't tell anyone about it, either. Which is the worst of both worlds. The system described there last August would achieve what they're going for with this "reasonableness" thing easier and more transparently. The lines are clearly drawn, you can look at one item on your character sheet to know if you can, in fact, attempt something, and we have a hard-wired way of knowing when it applies. Now, there are also a lot of problems with that system, but ambiguity and putting the responsibility to make the game make sense on the DM are not among them.

I like the general gist of this system.

Menteith
2012-06-07, 12:46 PM
Certainly, he will get a better result the vast majority of the time. But will he really get a better result when dealing with mercenaries? Maybe, maybe not. Is it possible that he's had an off day, and says something offensive? Of course. Is it possible that Gregor just happens to know some background on someone that Tyrion doesn't, allowing his to be more effective at building a friendship? Of course. Is it possible that Gregor has a common friend with the Diplomacy target that Tyrion doesn't? Yup. Heck, maybe the person being talked to is a woman that's attracted to Gregor, or someone just really impressed with his size.

Could we model all of that with situational modifiers? Of course! But that seems like a lot of hassle, and I don't think the results would be all that different than randomness. So I'll go with the simpler solution.

Except changing reality around in order to accommodate bizarre dice rolls seems terrible to me. If you want to set up a situation where Gregor has a social advantage (say, dealing with his own group of men), then he should have a bonus on a roll. But by this system, Gregor - a socially inept psychopath who seems barely able to form coherent sentences - can walk up and negotiate a treaty more effectively than an intelligent and urbane negotiator who lives and dies by his ability to convince people. I don't see how giving a(possibly cumulative) +2/-2 or giving advantage/disadvantage to characters is harder than this; it's not a hassle, it's incredibly simple to do.

What if Gregor is making a Diplomacy check in a situation where he shouldn't have any benefits, and still rolls really well? Hell, what if Tyrion is in a situation with advantage, and still rolls poorly against Gregor? It breaks the story to have random results that don't make sense occur, and the huge variance in dice will ensure that it occurs on a pretty frequent basis. The simpler solution is to allow characters to actually benefit from what they've specialized in, and make sure that their specialization is meaningful, not to reshape reality to fix absurd results.

Craft (Cheese)
2012-06-07, 01:06 PM
It's as if they half went with this idea (http://www.wizards.com/DnD/Article.aspx?x=dnd/4ll/20110816) but didn't really finish it and then didn't tell anyone about it, either. Which is the worst of both worlds. The system described there last August would achieve what they're going for with this "reasonableness" thing easier and more transparently. The lines are clearly drawn, you can look at one item on your character sheet to know if you can, in fact, attempt something, and we have a hard-wired way of knowing when it applies. Now, there are also a lot of problems with that system, but ambiguity and putting the responsibility to make the game make sense on the DM are not among them.


Nah. That'll never work. It's not "iconic" enough.

Knaight
2012-06-07, 01:14 PM
Thats a variance issue with a d20 versus a ~8-20 stat range corresponding to a modifier range of about 6. That means that the die has a range three times larger than the biggest range between characters you'd expect to see. But saying its a Fighter isn't really relevant - the same should be true of a Wizard with an 18 Str, or a Commoner with an 18 Str or anyone with an 18 Str.

That's part of it. The other problem is that the d20 is a completely flat distribution. With an exploding die system, the die range is technically infinitely larger (the variance and standard deviation are not, however), but because the probability of any individual result drops precipitously per explosions necessary to get it, that isn't a problem.

kyoryu
2012-06-07, 01:27 PM
That's part of it. The other problem is that the d20 is a completely flat distribution. With an exploding die system, the die range is technically infinitely larger (the variance and standard deviation are not, however), but because the probability of any individual result drops precipitously per explosions necessary to get it, that isn't a problem.

A bell curve dice system (like GURPS' 3d6) has much of the same effect.


Except changing reality around in order to accommodate bizarre dice rolls seems terrible to me.


Yes, it's a different model, and different models often seem terrible when they're first encountered.

The point is you're not "changing" reality. You're just not modeling it all. When you get a result that says Gregor did (anomalously) well, then you explain it with an advantage that he had. The advantage, in fiction, was there *all along*, it just wasn't modeled. It's a very narrativist approach, of course, and that doesn't appeal to all people.

Even without retconning relationships, people get along with certain types well. Maybe the noblewoman that actually likes Gregor does so because he reminds her of a younger brother, or an old paramour, or something along those lines. Since we probably don't model every fact of every NPC, there's usually enough run to safely come up with an explanation for the "anomalous" behavior.

The system you seem to be looking for is one where higher skill = *always* does better, and I just fundamentally disagree with that. I suspect we won't meet in the middle on this, though. FWIW, I'd also be okay with *less* variance than what 5e appears to be proposing. I'm not necessarily saying that the amount of variance they're using is *just right*. I'm trying to explain how it could work, if anything.

And again, the fact that these anomalies mostly show up in edge case scenarios (extreme gaps in skill) dealing in areas other than the core simulation of D&D (combat), I'm generally okay with more weird results in those areas, since they're less likely to have a meaningful impact on the game.

Seerow
2012-06-07, 01:30 PM
That's part of it. The other problem is that the d20 is a completely flat distribution. With an exploding die system, the die range is technically infinitely larger (the variance and standard deviation are not, however), but because the probability of any individual result drops precipitously per explosions necessary to get it, that isn't a problem.

Right. Even without exploding dice, just having more dice makes for a weighted average that makes smaller bonuses worth more.

For example, if using 3d6 instead of 1d20, for the -5 vs +5, the chance of the -5's chance of winning is MUCH lower. To the point where it probably is statistically insignificant. (I don't feel like calculating the exact odds, someone else is welcome to if they want).

Similarly, something like Shadowrun's system allows smaller bonuses to be more meaningful, because you're rolling so many dice you're weighted towards the average. Sure silly things can happen, but they're such a small fraction of the time, it's practically irrelevant (Like I can only remember 1 critical glitch happening in 2 year long shadowrun campaign). The difference between a guy with +4 (rolling 4 dice), and a guy with a +10 (rolling 10 dice), is WAY more pronounced than rolling a d20+4 vs d20+10.




Of course, D&D is stuck with the d20 mechanic. I really wouldn't want to change that. But it does mean that bonuses need to be higher for any sort of real divergence in capability to make sense. I mean just for reference, in 5e an 'immortal' feat, the absolute hardest checks in the game, supposedly attempted only by Gods, is a DC27. If you get a +7 to a skill, you can succeed on a 20.

This means a character at level 1 with an 18 in a stat and skill training can succeed at a godlike task 1 attempt in 20. If the character is a rogue, anything below a 10 becomes a 10, so his minimum roll is a 17, so the DM can't even penalize him for failing by more than 10.

So now you have to decide: Do you want gods to be weak and not able to do awesome things, or do you want level 1 characters doing godlike things 5% of the time?

TheAbstruseOne
2012-06-07, 01:46 PM
I don't know if it's been covered in the other threads (and no way I'm going to search through them all), but have they explained why the two sample clerics have a different number of orisons known? My guess is that it's either to do with the God being worshiped or to do with the differing wisdom bonus.

Your question seems to have gotten buried in the ongoing debate about math and abstraction vs. realism, so I figured I'd address it since I stopped caring about the other debate :p

They've stated in blog posts and Q&As that the two clerics were presented to show the range of customization within a single class, particularly when it comes to clerics and domains. Pelor has typically been the god of good, healing, and light so that is reflected in the character. Moradin has typically been the god of law, war, and...ale? I don't know, it's the dwarf god.

Anyway, they wanted to show that the cleric class is going to respect the two typical depictions of clerics in the game. On one hand, you have the "priest" type cleric, who stands in the back and heals/buffs everyone while shooting spells. The "Healic" and "Laser Cleric" type builds from 4e (which also existed in previous editions as well, but I'm not as familiar with them personally). On the other hand, you have the run-up-in-their-face, leading from the front with heavy armor and a massive mace "warpriest" type cleric.

And judging from the various forums, they've failed at that because everyone's focused on how "boring" the fighter is because all it can do is attack and that's it (even though they're showing multiple build types of characters is possible in the game) and because everyone including myself keeps calling the Moradin cleric a "Paladin", even though they've stated that the paladin will be a distinct class separate from the cleric.

So now I'm going to go into my personal opinion. I feel that the paladin should be a cleric that multiclasses or gets a theme. With only a few exceptions, pretty much every class can be recreated thematically by multiclassing or adding themes to one of the four core classes we've seen. Fighter/Rogue = Assassin. Cleric/Rogue = Avenger. Rogue/Wizard = mystic-style Ninja. Fighter/Wizard = Spellsword. Cleric/Wizard = Sage, Druid, or Bard. It doesn't even need to be a multiclass, either. Put a roguish theme on a fighter or a fightery theme on a rogue and you can get different styles of assassin, for example.

I really think this is what they should do as it's something that fits with modern game design while still respecting the tradition of the game. Start with simple core rules and add complexity on top of that. It also allows for a lot more "fine tuning" of characters that prevents every single 1st level whatever from feeling exactly the same in play. An assassin is an assassin is an assassin, but a fighter with stealth abilities feels completely different from a rogue with better combat stats. One is more thuggish and while good with ambushes doesn't need to resort to them, while the other is more finesse-based and relies more on those sort of ambush tactics in combat because he/she can't take a hit as well.

NichG
2012-06-07, 01:46 PM
Actually in the real world universal models are quite simple. To model any nonrelativistic, non quantum interaction between any two or more objects all you need to start with are Newtons laws. That is not to say that for specific cases you need approximations to even have a chance of calculating things, but the basic "rules" or "laws" of the system are simple.


This may just be me being pedantic, but writing down 'Newton's Laws' isn't a complete description of any problem. You're sort of missing the problem part: the degrees of freedom, the initial conditions, the interactions (which derive from QM applied to the material microstructures). In fact, its far worse: macroscopic objects interact via material properties that are quite non-trivial. Non-Hamiltonian terms like friction, inelasticity.

So putting the details missing aside, the initial conditions/index of degrees of freedom is the problem. This is the 'complexity' I'm referring to. Just try inputting the contents of your room into a computer. Even just simple 3D models. It'll take hours, and you'll probably miss things.

In practice, usable models extract out the salient degrees of freedom and discard the rest as an approximation. That approximation has a regime of validity - describing an object as a rigid body works well for determining how it'll pile up if you drop it, but doesn't do too well in describing what'd happen if that same object were shot at pointblank range with a shotgun. That finite regime of validity is what you pay for having something simple enough to think about without a computer handling all the details.

Menteith
2012-06-07, 01:53 PM
Since we probably don't model every fact of every NPC, there's usually enough run to safely come up with an explanation for the "anomalous" behavior.

Usually being the key word here. I see where you're coming from, but (at least, with how my group and I play) there are certain NPCs and settings that are indeed very well defined. For some interactions, this might work, but other times it wouldn't work too well. If a setting is as fleshed out as the world of SoI&F, both with regard to player backgrounds and NPCs, this mechanic would be jarring.


The system you seem to be looking for is one where higher skill = *always* does better, and I just fundamentally disagree with that. I suspect we won't meet in the middle on this, though.

I don't want a system where a higher skill always wins. I want a system where a larger skill always matters - this could work with a "Degrees of Failure" system instead of a binary pass/fail (an expert swimmer will almost never be at risk of drowning while swimming a river - instead, they simply make less progress than normal), or a system that limits what certain skill levels can attempt (similar to the variant skill system Stubbazubba linked), or by using normalized dice (3d6 vs. 1d20) which makes the smaller bonuses of 5E more relevant (and decreases the chance of a wild result). I'm fine with it happening once in a blue moon, but a greater than 10% chance is baffling.

Seerow
2012-06-07, 03:13 PM
Wow double posts more than an hour apart. That's new.

Craft (Cheese)
2012-06-07, 03:50 PM
because everyone including myself keeps calling the Moradin cleric a "Paladin", even though they've stated that the paladin will be a distinct class separate from the cleric.

Nonsense! The Moradin cleric has spells that actually matter, and lacks the magic horsie.

Stubbazubba
2012-06-07, 04:25 PM
I suppose one somewhat easy fix is to say that if your modifier is X or more lower than the other guy's or the DC, you have Disadvantage. That will prevent the vast majority of "unreasonable" but mechanically feasible things from actually happening almost 100% of the time. This could work in reverse, as well; if your modifier is X higher than the other guy's or the DC, you have Advantage. Although, with opposed rolls, your kind of giving both an advantage and a dis for the same thing, which I don't particularly like, but it would use the innovation they have come up with to deal with this huge hole in the rules without putting all of the responsibility for fixing it on the DM.

kyoryu
2012-06-07, 04:33 PM
Usually being the key word here. I see where you're coming from, but (at least, with how my group and I play) there are certain NPCs and settings that are indeed very well defined. For some interactions, this might work, but other times it wouldn't work too well. If a setting is as fleshed out as the world of SoI&F, both with regard to player backgrounds and NPCs, this mechanic would be jarring.

Right. And if Joffrey is trying to Diplomacize Arya, it Just Don't Work. No roll needed. "Dude, you killed her father, killed her friend, killed her sister's wolf, and forced her wolf to run off. She hates you. She's not going to go out with you."

It seems like a lot of the disconnect or difference I'm seeing with some people is the difference between the rules being the "physics" of the world, and the rules being used to determine the result when the result is not clear. The latter is a pretty typical old-school view of gaming, while the former is much less so.

Basically, the former view says "hey, if it's impossible for something to happen, the rules should say so." The latter says "if it's impossible for something to happen, then why do you need rules?" They're both viable, and both equally subject to DM fiat (if we achieve impossibility for the Arya/Joff scenario by adding disadvantages, then those disads come from somewhere - namely the DM. Even if they come from a list, the DM decides which do or do not apply).

Seerow
2012-06-07, 04:39 PM
Right. And if Joffrey is trying to Diplomacize Arya, it Just Don't Work. No roll needed. "Dude, you killed her father, killed her friend, killed her sister's wolf, and forced her wolf to run off. She hates you. She's not going to go out with you."

It seems like a lot of the disconnect or difference I'm seeing with some people is the difference between the rules being the "physics" of the world, and the rules being used to determine the result when the result is not clear. The latter is a pretty typical old-school view of gaming, while the former is much less so.

Basically, the former view says "hey, if it's impossible for something to happen, the rules should say so." The latter says "if it's impossible for something to happen, then why do you need rules?" They're both viable, and both equally subject to DM fiat (if we achieve impossibility for the Arya/Joff scenario by adding disadvantages, then those disads come from somewhere - namely the DM. Even if they come from a list, the DM decides which do or do not apply).

The problem with this as has been mentioned several times earlier in the topic is 'what's reasonable' is going to vary from person to person, even if you have the same assumptions about power level (ie "We're playing Game of Thrones, not Wheel of Time").

The game system is supposed to be what tells you what is reasonable and what is not within the world. That's the whole point, the rules give you a baseline idea of what is possible within the world, and gives players a common ground to work with when cooperating.

TheAbstruseOne
2012-06-07, 04:44 PM
Wow double posts more than an hour apart. That's new.

Been having browser problems today. Woke up, checked the thread, wrote my reply, went to do other stuff (reddit, webcomics, etc.), came back to the tab and there was a 503 error, and refreshed the page. Thus, double post more than an hour apart. I deleted the second one though.


Nonsense! The Moradin cleric has spells that actually matter, and lacks the magic horsie.

In a debate on ENWorld about that right now. My argument for independent class over class + theme is when the class itself can support multiple themes while still retaining the "feel" of the original class. A fighter with an archery-based theme and nature background still feels like a subset of fighter. Would a paladin with an archery-based theme still "feel" like a paladin?

Conundrum
2012-06-07, 04:46 PM
In a direct contests, the guy with a +17 modifier loses or ties with the no modifier guy 1.5% of the time, strictly losing 0.75% of the time. That seems acceptable to me, as a artifact of the model.

I don't see where you got your 10% figure.

Me neither, to be honest. Some kind of awful brain-fart, I guess. Sorry :smalleek:

kyoryu
2012-06-07, 05:01 PM
The problem with this as has been mentioned several times earlier in the topic is 'what's reasonable' is going to vary from person to person, even if you have the same assumptions about power level (ie "We're playing Game of Thrones, not Wheel of Time").

The game system is supposed to be what tells you what is reasonable and what is not within the world. That's the whole point, the rules give you a baseline idea of what is possible within the world, and gives players a common ground to work with when cooperating.

Yes. And in my post, I basically pointed out that my solution to this is "it's the DM's say." This may seem arbitrary, but since the application of the rules is really up to the DM anyway, I don't see it as adding any arbitrariness that doesn't already exist.

At the base level, let's say there's a rule that says "disposition towards to Diplomacizer can yield between a +30 and -30 modifier to the roll." The DM still gets to make that decision, and is within the rules.

We can refine that further, with a list of possible positives and negatives. But the DM is still going to decide which among that list apply, and which don't.

So in a way, adding rules doesn't really change the fundamental fact that it's the DM deciding how likely something is to happen. It's just adding to the complexity.

I suspect you highly disagree with this idea. I'd also make a guess that you're very anti-rule-zero. I don't mean this as an insult, just that I'm starting to get a feeling for what you want out of games, and I'd bet that rule zero is pretty antithetical to it.

Menteith
2012-06-07, 05:08 PM
Basically, the former view says "hey, if it's impossible for something to happen, the rules should say so." The latter says "if it's impossible for something to happen, then why do you need rules?"

Because in a universe like D&D where the impossible can happen on a daily basis, it should be clear what is and what is not possible. With magic and other reality warping powers, I don't have a real life basis to determine what is and what isn't reasonable. For example, what happen if, say, Joffery had Glibness on him while trying to negotiate with Arya? Arguably, even though she has every reason to hate him, the magic empowering him could let him convince her. But it's still not unreasonable for someone to say that regardless of how convincing he is, he can't do it. In situations that I can't use real life experiences to cover for, the rules need to be clear.

EDIT

I'm against overt Rule 0. Players should have the ability to make changes in a dynamic environment, and our group prefers a clearly defined set of rules. When we make a house-rule, we'll decide as a group, and record the rule for future reference. Whenever a DM is forced midsession into saying "no, that's not how it works" in order to make a campaign work, something's gone wrong. Rarely, it's the only solution, but it always leaves a bad taste in my mouth when I have to do it.

Seerow
2012-06-07, 05:14 PM
I actually don't have a problem with rule 0. Houserules are something that come up all the time. I'm an active homebrewer (though I haven't really posted much in that area recently), and enjoy seeing things tinkered with and changed, without rule 0 that's not possible. Similarly, with my regular group the main rule of the game is, try to stick with the rules, but if we're not sure of what it is, don't waste time looking it up, let the DM decide, and look it up after the game so we can remember it later. I'd qualify that as rule 0.

The point is, the DM changing things, or making rulings, isn't something I'm opposed to. What I am opposed to, is that the game requires this of the DM. I don't agree with any system that demands a DM to have to make up rules for literally everything as you go. That puts a lot of pressure on new DMs who don't have experience, and discourages people away from taking that role. I don't agree with the idea that you can't have any real baseline assumptions about what is and isn't possible without asking the DM. The DM might change something with rule 0, but that would be a intentional change of the game's assumptions, rather than the game not having any assumptions at all. There is a very big distinction.

russdm
2012-06-07, 05:36 PM
I know Mearls is pushing it hard, because if people buy into it it makes his job way easier, but why do we expect the DM to do so much? I mean it's this exact sort of attitude that keeps people away from DMing. So much responsibility gets placed on the DM that most players are afraid to take it. I loved 4e because it made the DM's job so much easier, people around my area who had previously refused to DM were running games, and doing so easily. Returning to the olden days where rules are archaic and the DM needs to decipher everything puts the onus for everything back onto him, and that's something most people simply don't want to deal with.

That was part of my issue with DMing 3.5, but also more so because of the group i am playing with. 3 power gamers in it out of 5 players. Its really hard to make encounters that challenged that didn't go overboard on giving them stuff like xp and treasure as a result. Yeah, I screwed up severely both times that I DMed. And screwed up the first time I DMed, since that group had two of those power gamers. They have really put me off DMing as a result.

Chance is something that happens though. The level 1 dude might be able to win an arm-wrestle with the level 20 fighter. the chance exists that it might randomly happen. Chances happen in real life, so why wouldn't they exist in the game?

Both Backgrounds and Themes seem to me to be ways of personalizing your character. Also, i think the main issue with the moradin cleric and the pelor cleric is that people just don't get it. Let me example what i mean before people start flinging stuff at me.

First off: So while there is one cleric class, there are different ways to play that cleric. So far we have seen two differents kinds of clerics. Clerics get to worships deities and can 'Specialize' in parts of the deities' portfolio. A good with domains of good, law, protection and sun will have clerics that pick up one or two of those domains and each cleric will be different from the others.

Moradin Cleric: Others have said that this cleric is very paladiny, well that makes sense because paladins are warrior/fighter clerics. Paladins have usually been clerics that have strong martial sides. Also this cleric seems to be following the war aspect of moradin. How would this cleric be different if he followed the aspect of moradin relating to dwarves or ale? Wouldn't he have different spells/abilities/what-have-you to reflect that?

Pelor Cleric: This cleric seems to be very like the monks of europe that lived in monasteries and spent alot of time in study. A cleric like that would be different from a lay-cleric or a cleric that dedicated themselves to fighting undead. This cleric seems to be a fairly scholary sort or alteast spent time pursuing unique aspects.

The main thing though is that both kinds of clerics have shown up in the game before this. Why is it now a problem?

Linear Warriors/Quadratic Wizards has always been a problem for the game designers to solve and they have usually failed in my humble opinion. The problems stems alot from how the methodalogy is used. That causes problems that start small and grow large to create a massive divide.

The fighter(warrior) learns his abilities through practice (no matter how he learned origionally) and tests them out frequently on monsters. He will improve his skills quite a bit and learn new things, but all of this is still bound on one major caveat: His physical ability

The fighter's body can be only pushed so far and can only take so much. The fighter's body can only learn so much muscle memory before reaching a cap on limits. The Hulk might be special, but he still has limitions. He can only jump so high up despite his great strength, he can only punch so hard despite his great strength. likewise the fighter reaches a point where he simply cannot do more. He can't jump higher, he can't fight any better because he is still mortal. He will reach the pinnacles of his abilities and stop there, trying to surpass those limits but be unable to do so because his body in its state is unable to surpass them.

Wizards and other spellcasters are completely different. While they practice using magic and doing things, they are bound by different rules entirely. Their physical self rarely impedes them and they can use stronger magic easily as they level up. The wizard becomes able to surpass the physical limitations of the fighter at a lower cost and far quicker. After that point, they simply increase in ability until they can rewrite reality or nearly so. They also stay mortal during the whole time. Its less hard to develop magical ability than physical ability. Many athletes have dedicated the bulk of their lives or their entire lives to develop their skills and it usually requires that kind of dedication for them to have reache that point. Thats not the case with wizards. They can pursue magic to a point then do something else, but they will be stronger than an athlete that spent their life pursuing something.

Another major point is that the fighter had to suffer alot to get to where they are; the same with athletes. They suffered blood, sweat, and tears in their journey. Wizards don't have to doing any of that. They don't have to suffer anything to get to being extremely powerful other than to risk dying.

Something who is better at something should be able to succed more. Thats the result of being better at it.

Rule 0 is something that should be used when required. It shouldn't be needed in order to play/run the game. The game system should be compact and functional enough to allow for that. the game system also needs to be customizable for play. The game system needs to a foundation that the players/DMs build onto or take away from as they like. The system needs to work across the board because it is made to simulate/demonstrate 'real life'. The characters inhabit a world that has laws to it and the game system needs to be those laws. The laws on earth show up in the background and are rarely noticed. The law of gravity, the law of oxygen required for humans to breathe, the law that your body holds its shape and doesn't float apart or mutate every second where-in your arm is suddenly longer than your leg and then shorter than your nose. Do we notice those rules all the time or in a few individualized instances?

The game system is designed for versimilitude. Having to houserule large portions that should have been there already breaks that versimilitude because the laws the game world is based on are in flux or have problems. If the law says that you need oxygen to breathe, then you should need oxygen to breathe and have it available unless you have to make it available. If the law says you need oxygen to breathe, but there is no source of oxygen available, then you should be unable to breathe.

The presence of magic changes the very nature of existence. If it appeared in/on earth, it would completely change the way we live our lives. We have to assume something like that happened in teh game world. Magic has changed how the characters live and that needs to be reflected. It makes magic almost like oxygen for us and it should be treated as such.

Another major point is that sometimes the impossible happens. somehow it does.

If happen to be strong enough, then you should be able to do stuff anyway. If you happened to learn math when younger and learned how to add/subtract, then you should be able to do so without problem. You may need to recall what you learned or remember it, but you shouldn't be able to fail to add/subtract. You might not peform it correctly, but you shouldn't forget how to do it and have to relearn how to every time you need to. Also if you learned how, you shouldn't have to roll a dice that says because you failed on the roll that you don't know what they even happen to be. That can happen if you fail knowledge checks, even if you have already encountered a monster before.

You don't need to think on how to breathe naturally, but you do need to for situations like going into space or underwater, because those are places where it happens to be concern.

(I think i am rambling at this point. What was the topical discussion again?)

TheAbstruseOne
2012-06-07, 08:00 PM
(I think i am rambling at this point. What was the topical discussion again?)
I honestly thing the two of you are the only ones who know at this point...

New topic! What classes from previous editions do you think work better as stand-alone classes and which work better as themes or multiclass builds?

holywhippet
2012-06-07, 08:05 PM
In a debate on ENWorld about that right now. My argument for independent class over class + theme is when the class itself can support multiple themes while still retaining the "feel" of the original class. A fighter with an archery-based theme and nature background still feels like a subset of fighter. Would a paladin with an archery-based theme still "feel" like a paladin?

I have thought on occasion it might be interesting if they went back to something in the original D&D which was to require pre-requisite class levels for classes. I've never seen it myself (have ordered the reprint for when it comes out) but I gather to take levels in bard you needed to have levels in certain other classes first (fighter, thief and druid maybe?).

It would be interesting to have a system where either all characters start out with fairly simple classes and have to work on certain class levels or something similar in order to progress towards certain roles. Final Fantasy Tactics had something akin to what I mean, except there wasn't really much synergy or logic to class pre-requisites.

Treblain
2012-06-07, 08:14 PM
Why are we saying that the 5th edition rules allow for this whole variance problem? They DON'T, because they say to not use the rolls when inappropriate. The rule to use common sense instead of allowing unnecessary variance due to dice-rolling is also part of the rules.

The dice-roll system has a problem, and the rules address it. You can dislike how it's being addressed, but just because it isn't a mathematical solution for the problem doesn't let you ignore it and continue complaining.

Level and DC scaling is a problem that makes skills less meaningful in 4e. Lack of DC scaling in comparison to bonus increase is a problem that makes skills less meaningful in 3e. Both of those are being avoided, and without removing the d20. I'd call that a win.

Craft (Cheese)
2012-06-07, 08:26 PM
I honestly thing the two of you are the only ones who know at this point...

New topic! What classes from previous editions do you think work better as stand-alone classes and which work better as themes or multiclass builds?

Alright, you'll probably be able to see a fair bit of my heavy pro-wizard bias here (I just love the fluff behind a character who uses vast knowledge of the inner workings of the world to engineer solutions to things. I just naturally gravitate toward them even games where they suck). But here it goes:

Assassin
Barbarian
Fighter
Monk
Ranger
Rogue
Paladin
Warlord

To me, all of these classes are mergeable with each other in concept (of being just "Fighter + some extra twists"), assuming we're talking about the Assassin as nonmagical (which is what I believe it was at first; the 3.5 version added spells AFAIK). The magical 3.5 Assassin can probably be handled by just making slight changes to the wizard.

Druid
Cleric
Wizard

I got one problem with these 3: Except for Wild Shape, they're basically identical aside from the different spell lists. Make Wild Shape into a set of feats, give the wizard cure spells and just call it a day.

Sorcerer
Warlock

...Yeah. Sorry guys, but Spontaneous/At-will Casting is not different enough to justify a whole new class. 4E sorta made a worthwhile difference between them by making these two damage-based while the Wizard is control/utility based, but I'm all for making a smaller number of broader classes instead of a bazillion tightly focused classes like 4E went with.

Barbarian
Druid
Ranger

I'm separating these guys out because they all share a thing in common of being the "close to nature guy." (That is, if you consider the aspect of the Barbarian as the tribal warrior, rather than the other aspect of being a fighter who works off of instinct and natural talent rather than years of training.) For somewhat personal reasons that would be inappropriate for this discussion, I've always absolutely despised these type of fantasy tree-hugger characters and I wouldn't shed one tear if these three classes (and elves) were removed from 5E entirely.

Menteith
2012-06-07, 08:27 PM
Why are we saying that the 5th edition rules allow for this whole variance problem? They DON'T, because they say to not use the rolls when inappropriate. The rule to use common sense instead of allowing unnecessary variance due to dice-rolling is also part of the rules.

As has been said before, many supernatural/magical interactions have no basis in reality, and what's reasonable could vary drastically between players. Look at how widely different illusions behave in 3.5, or what qualifies as "against a subject's nature" with regard to Charm Person. Completely reasonable, intelligent people can have a wide range of opinions on how these spells behave, and be completely justified in their opinions. The easiest solution to these problems is to have a consistently applied rule - and I'd rather that they ship the game with that rule in place, instead of ask me to fill in areas.

Seerow
2012-06-07, 08:43 PM
I honestly thing the two of you are the only ones who know at this point...

New topic! What classes from previous editions do you think work better as stand-alone classes and which work better as themes or multiclass builds?

Hrm...

Honestly this depends on how big of a role classes play vs Themes and Backgrounds. Like right now the classes we can see from 5e vary so much in how much the class influences them it's impossible to tell what the intent is in how important a class is.

I mean, it is theoretically entirely possible to make an entire game just based around the core 4: Fighter/Cleric/Mage/Thief.

I might even argue for knocking that down one more, making Thief into a Theme geared towards Fighters with a sneaky/underhanded bent, and leave you with just Fighter/Cleric/Wizard. Throw in Psion because Psionics has become a pretty big thing in D&D as well.

From there, just about any class from any edition can be recreated from some combination of these. Druid? Nature Domain cleric. Ranger? Fighter with a Wilderness theme, maybe with a splash of cleric levels. Paladin? Cleric of Moradin already pretty much looks just like one, but multi-classing with Fighter could work just as well. Sorcerer and Warlock could easily be Wizard themes. Barbarian could be a Fighter theme. And so on.

So the question is really where to draw the line? Does having sneak attack and more skills justify the rogue being a separate class, rather than just making the Fighter more skilled, so he can fulfill the whole badass mundane role himself? Does a character having his magic innately rather than through study warrant its own class?

It's a tough choice. There's always been a lot of overlap between classes. I mean, it's even possible to drop below those core 4, I've been toying with the idea of dismantling the Cleric class, and turning Domains into features that anybody can pick up via feats or bonding magic item slots (a la ToM). In such a system you could get away with just the Fighter and Wizard in theory.




Now after all that ranting, going through the core books of 3e and 4e, and picking classes I would definitely want for a base class in a core game of 3.5 I'd say:

Martial:
-Warlord
-Barbarian
-Rogue

Arcane:
-Wizard
-Bard
-Warlock

Divine:
-Cleric
-Druid
-Warlock


Reasoning for this class set up:
-Warlord is the skilled tactitian. A versatile martial class with an emphasis on leadership and tactical abilities. Has access to most martial powers. Basically the Wizard of the Martial Characters. For those who viewed Fighter as a tactically focused character, that concept has been subsumed by this class.

-Barbarian is the raging berserker. For those who have traditionally viewed Fighter as the heavy hitter, that concept is subsumed by the Barbarian. Barbarian has fewer powers relative to the Warlord, but much better self-buffing ability, passively gaining great mobility, damage, and defenses while raging.

-Rogue is the sneaky bastard. His abilities are predatory in nature, locking down and devastating the opponents he singles out. His defenses are highly reactionary. He has low defenses, but many escape style powers that can be activated as a reaction. As an example, a rogue who gets targetted for an attack may be able to resist that attack with a hide check, which if successful negates the attack and hides the rogue until the end of his next turn.

-Wizard is the same old Wizard we've always had. Possible change includes more distinctive specializations. Possibly something like Psionics specializations where only specialists get access to the highest tier of abilities. So if you want Shapechange, you're going to be a transmuter, which means you're losing out on a lot of the best defensive buffs, summons, and other stuff.

-Sorcerer Like in 3.5, this class gets the power of the Wizard, but less variety, and more longevity. Has access to powerful at will blasting, with access to some Wizard spells to supplement its utility (likely still using spontaneous casting mechanic).

-Bard This class represents the jack of all trades who uses music to make magic. The class would remain close to how it was in 3.5, but with some of the better non-core toys moved into the core class, and likely with Inspire Damage automatically doing something to Dragonfire Inspiration (ie +d6 damage instead of +hit/damage).

-Cleric Loses heavy armor and its medium BAB, and becomes a real back line caster like the Wizard. The Cleric gets the most domains of any divine class. These domains give access to special abilities that align with that domain. Most clerics should be similar in flexibility to a specialist wizard.

-Warlock This one is kind of weird for a lot of people to consider as Divine, but the fluff of the Warlock is that he has made a pact with a greater power for his abilities. That to me screams divine. It could come from a god (good or evil), or a Demon King, but he gains his power from essentially divine beings. The Warlock picks a pact which gives him access to a couple of domains, and a bunch of passive and at will powers as he levels.

-Druid The druid gains only the Nature Domain, and as such has a very limited source of divine powers relative to the other divine classes, but in exchange gets to keep many of the features he has in 3.5, such as the animal companion and wild shape.



Out of those, the most questionable one was Druid, as I really think that should be doable with Cleric or Warlock easily, but I wanted to keep things symetrical (3 groups of 3), and none of the other core (or even non-core) divine classes fit the bill. In fact, the lack of diversity in divine classes, and the diversity available within the Cleric class, is a large part of why I consider getting rid of divine classes altogether.

The other classes all fit a pretty solid pattern. You have the Flexible/Complex class (Warlord/Wizard/Cleric), the Simple/Strong class (Barbarian/Sorcerer/Warlock), and the jack of all trades class (Rogue/Bard). I just couldn't think of anything divine that fits that same sort of role.




And wow I typed a lot here and am rambling quite a bit, so I'm gonna go ahead and cut it short there.

demigodus
2012-06-07, 08:46 PM
Why are we saying that the 5th edition rules allow for this whole variance problem? They DON'T, because they say to not use the rolls when inappropriate. The rule to use common sense instead of allowing unnecessary variance due to dice-rolling is also part of the rules.

common sense is also known as going with your gut instincts. The only time people can agree on something with common sense, is when they have some common basis upon which to make their decisions. Like, referencing something similar going on in real life.

This is really freaking hard to do in a game called Dungeons and Dragons where Magic makes up a very large portion of the setting and its rules.


The dice-roll system has a problem, and the rules address it. You can dislike how it's being addressed, but just because it isn't a mathematical solution for the problem doesn't let you ignore it and continue complaining.

Not complaining about the rules not addressing it. Complaining about the rules not fixing it. Big difference. We ARE allowed to disagree with how the rules address it, or dislike how they fail to fix the problem. And no, "if there is a problem, it is the DM's job to rule 0 a fix" is NOT a fix to the rules. In fact, here the way the rules address the problem is an entire another problem.


Level and DC scaling is a problem that makes skills less meaningful in 4e. Lack of DC scaling in comparison to bonus increase is a problem that makes skills less meaningful in 3e. Both of those are being avoided, and without removing the d20. I'd call that a win.

So, DCs increasing with the bonuses is a problem? DCs not increasing with bonuses is a problem? Hence 5e's system where bonuses don't increase, therefore we can't in anyway claim DCs correlate with it, solves the situation?

Is that really your argument? Please tell me I'm horribly misreading your point and clarify what you mean...

Seerow
2012-06-07, 08:49 PM
Why are we saying that the 5th edition rules allow for this whole variance problem? They DON'T, because they say to not use the rolls when inappropriate. The rule to use common sense instead of allowing unnecessary variance due to dice-rolling is also part of the rules.

The dice-roll system has a problem, and the rules address it. You can dislike how it's being addressed, but just because it isn't a mathematical solution for the problem doesn't let you ignore it and continue complaining.

Level and DC scaling is a problem that makes skills less meaningful in 4e. Lack of DC scaling in comparison to bonus increase is a problem that makes skills less meaningful in 3e. Both of those are being avoided, and without removing the d20. I'd call that a win.

This was addressed upthread, but in the DM Guidelines, it specifies a roll is inconsequential enough to not roll it when the DC is 5 lower than your attribute. In an opposed contest there is no DC so it's never inconsequential enough to not roll.

Anything beyond that is DM fiat, and has nothing to do with the system.



\
Barbarian
Druid
Ranger

I'm separating these guys out because they all share a thing in common of being the "close to nature guy." (That is, if you consider the aspect of the Barbarian as the tribal warrior, rather than the other aspect of being a fighter who works off of instinct and natural talent rather than years of training.) For somewhat personal reasons that would be inappropriate for this discussion, I've always absolutely despised these type of fantasy tree-hugger characters and I wouldn't shed one tear if these three classes (and elves) were removed from 5E entirely.

Meh, for the Barbarian I'd just rename him to "Berserker" if you want to get rid of the tribal flavor. I mean very little in the class lends itself towards nature in the same way that Ranger and Druid does, and "Guy who is really tough and busts things" I think is arguably a viable class concept.

Craft (Cheese)
2012-06-07, 08:54 PM
Meh, for the Barbarian I'd just rename him to "Berserker" if you want to get rid of the tribal flavor. I mean very little in the class lends itself towards nature in the same way that Ranger and Druid does, and "Guy who is really tough and busts things" I think is arguably a viable class concept.

Fair enough, though I still think this Beserker is better handled as Fighter + Rage-based theme. But many people were very upset when the Barbarian was missing from the 4E PHB, so we know WotC's not going to do that again. Ah well.

Also, I can't believe I completely forgot about the Bard!

demigodus
2012-06-07, 09:06 PM
This was addressed upthread, but in the DM Guidelines, it specifies a roll is inconsequential enough to not roll it when the DC is 5 lower than your attribute. In an opposed contest there is no DC so it's never inconsequential enough to not roll.

Anything beyond that is DM fiat, and has nothing to do with the system.

Ah, thank you for that clarification.

Still, now we have the amusing result where, if your Str is 20 (+5 mod), you auto-succeed on anything DC 15 or lower. DC 16? You have a 50% failure chance. Skills could change that probably, but I found that idea amusing that 55% success rate = so trivial you always succeed.

That guideline would work if your stat mod was rewritten to be your stat - 10 (without the divide by 2 part that 3.5/4e has)

TheAbstruseOne
2012-06-07, 09:13 PM
Warlord is hands down the class I do not want to see in Next. The entire function of the class is subsumed by the cleric and it only really made sense in 4e due to the power source/role dynamic.

Same thing for paladins. There's two threads on ENWorld right now where I just cannot get someone to give me a good reason for the paladin as a stand-alone class. The paladin is a holy knight in shining armor. You take away the "holy", "knight", or "armor" and you no longer have a paladin. You have something else entirely. That means there's just not enough difference in "style" of paladin to justify a class. Also, pretty much every D&D campaign world (save Dark Sun, Dragonlance during certain points, Ravenloft depending on who you ask, and probably a few of the others I never looked at) is a polytheistic world. The archetype of the paladin is firmly rooted as the idea of the Christian Soldier. What would a paladin of Melora look like? And wouldn't a cleric of Bahamut or Kord look a lot like a paladin?

Assassins are a class I've always loved...but I'll be the first to admit they'll work far better as a theme or multiclassing than they ever would as a stand-alone class. If I could figure out how to pull the pregens apart, I could probably make an assassin. Fighter with the thief theme = brutish combat-oriented assassin ready to fight toe-to-toe. Rogue with the slayer theme = combat pragmatist who strikes from the shadows and takes every advantage to drop his opponent as soon as possible so he doesn't have to fight toe-to-toe. Hell, rogue with the arcane dabbler theme and you'd get something pretty close to the "mystic ninja" style assassin.

Rangers I'm on the fence about. I can see them working very well as a class + theme, but they'd still be missing something. An archery or two-weapon theme and woodland background on a fighter would get you something that feels like a ranger, but it'd be missing that mystical aspect...it'd only work if fighters are allowed two themes so they could take the combat theme and then the healer/herbalist theme from the Pelor cleric.

Barbarians? Give fighters two themes and tack one onto the Slayer that gives him a rage ability. That simple.

Sorcerers might work as a theme, but would require a bit of finessing to shoehorn in using the existing rules. Using the Next rules, you'd have to have a wizard's spell list that casts like a cleric does. Not sure if a theme will be able to re-write a class like that, but I don't think that's enough to warrant and entire new class.

Bards = Wizard class + healer theme + performer background. Done.

Avengers...oh I love avengers...but really a cleric with a roguish theme or a rogue with a clericy theme would do all the work.

I'm not going to talk about druids because I can't stand the damn things. Never played one, never had one in one of my games by luck alone, the fluff doesn't interest me, and the class abilities range from useless to completely and utterly broken depending on edition.

And with that, I can't really think of any other classes off the top of my head. Everything else can easily and obviously be covered by a theme (Warlock? Wizard with a theme. Swashbuckler? Fighter with a theme. Invoker? Cleric with a theme.)

Seerow
2012-06-07, 09:14 PM
Ah, thank you for that clarification.

Still, now we have the amusing result where, if your Str is 20 (+5 mod), you auto-succeed on anything DC 15 or lower. DC 16? You have a 50% failure chance. Skills could change that probably, but I found that idea amusing that 55% success rate = so trivial you always succeed.

That guideline would work if your stat mod was rewritten to be your stat - 10 (without the divide by 2 part that 3.5/4e has)

Yes, it likely would. But characters with a +10 mod on their roll just from attribute are already too high to fit into WotC's 'bounded accuracy' system.

Seerow
2012-06-07, 09:18 PM
Warlord is hands down the class I do not want to see in Next. The entire function of the class is subsumed by the cleric and it only really made sense in 4e due to the power source/role dynamic.

This, combined with the later stuff in the post reads to me as "I want a Cleric as the healer in the group, nobody else should be healing it just doesn't make sense".

Which really sucks for any group who wants to, y'know, survive, without having a cleric or access to magic items like wands of CLW/LV.

Seriously, the Warlord was far and away one of the most popular additions to 4e, I know a lot of people who absolutely despise 4e who still liked the class. Scrapping it simply because you don't like the thought of a healer who isn't fueled by gods isn't a good idea.

TheAbstruseOne
2012-06-07, 09:34 PM
This, combined with the later stuff in the post reads to me as "I want a Cleric as the healer in the group, nobody else should be healing it just doesn't make sense".

Which really sucks for any group who wants to, y'know, survive, without having a cleric or access to magic items like wands of CLW/LV.

Seriously, the Warlord was far and away one of the most popular additions to 4e, I know a lot of people who absolutely despise 4e who still liked the class. Scrapping it simply because you don't like the thought of a healer who isn't fueled by gods isn't a good idea.
I like 4e and I like the class. I just don't like the class outside of 4e. Outside 4e, it's just a reskinned cleric. The cleric says, "The light of Pelor heals you!" The warlord says, "Get up, it's only a flesh wound!" Then they each heal 1d8 + Wis mod damage and their turn is over.

And saying that I don't like the warlord because I think only clerics should heal is like saying I don't like alchemists because I think only wizards should get to throw around fireballs. Not true. I just don't think that there's enough difference between what the cleric and the warlord do mechanically to justify being a separate class. I'm also championing the idea of combat healing outside the cleric on various forums as well, either by using a healer's kit during combat or taking a second wind like action that allows you to spend hit dice (at a penalty of some sort...like paying 2 for 1 or rolling them with Disadvantage).

And I don't like the alchemist because I don't like the idea of Crazy Harry from the Muppet Show playing in my D&D game.

I could be won over by the warlord if they managed to do it right, though. If they can find a way to make the warlord feel mechanically unique and still tie into the rest of the game world thematically, I'd be thrilled. I'm not sure if they'll be able to pull it off though.

Side note: What's LV? Wracking my brain and I can't figure it out...

Seerow
2012-06-07, 09:38 PM
LV would be Lesser Vigor. Editing in response to the rest in a minute.



Edit: Okay, the thing about the warlord is the warlord's schtick isn't just healing. He's just capable of doing that well. Which is good, because having a guy around who can make you get better is good. And yes, having a different class for someone to be able to heal without having to be religious is a good idea. I mean in 3.5 there were work arounds (frequent refluffings of Clerics/Favored Souls/Crusaders at low level, and just packing wands and someone having UMD at high levels), but not needing to refluff is nice.

What's more, the Warlord fills a niche beyond just healing. He is the martial commander. Your tactical strike leader. The guy who may not be as personally badass as everyone else, but that's okay because he can direct everyone else to be that much better. Think Captain America in the Avengers, he's not the most powerful character in the group by a long shot, but he earns his keep by being the leader, providing direction and helping others be more effective. That is the role that the Warlord fills.

I can hear the argument for that having significant overlap with the Fighter (and indeed, with my list I took out the Fighter and dispersed him between the Warlord and the Barbarian), but because of overlap with the Cleric? No.




Honestly though, judging by how many classes you specified you would remove, it seems you fall closer to the "only 3 classes with lots of specializations" from the beginning of my class rambling post. So maybe this whole point is moot. However it is important to note in a system where you have so few classes, each class needs to be able to represent a large variety of archtypes and roles, which means if you have Fighter/Cleric/Wizard, then you can't say "Only clerics can heal", you should be able to get an approximation of the Warlord through use of options of the Fighter class.

russdm
2012-06-07, 09:48 PM
My two gold, er copper...

Barbarians have always been weird. They happen to be rather strange. So far, from my playing experience, Barbarians have been basically beserkers. They rage to fight better in combat, and they can't read or write. They rarely have any of their nature side come up.

Bards make sense as a separate class mainly because you need to have a class for elan. other than that, i can't see much of the appeal. "I can sing to make my allies hit better, and i am a wannabe rogue. Oh, i also have this nifty ability to find stuff out, which is nearly the same as a skill. Oh, i can cast magic through music."

Druids: I think i like the concept, but i think the implentation went a little crazy. The stuff is decent, but the fluff is kinda meh. They do get animal companions, plus being able to have a bunch of spell-casting. The combination of Natural spell with wildshape made them broken, but its somewhat understandable. What they could have done was to make some spells not castable in wildshape with natural spell anyway or maybe make spells a little harder to cast? I don't know.

Paladins are falling more towards clerics on the scale of clerics to fighters. A theme/build for clerics allowing them to pick some stuff to make the cleric play like a paladin makes sense.

Rangers act more like a druid/fighter mix that is limited in what it can done. Maybe if having the druid class, then the ranger could be a build for it?

I won't cover the 4e classes because i don't recall enough to discuss them.

Sorcerors and wizards were two sides of the same coin. They were basically opposites, but nearly all presentation of wizards in the books (dragonlance, Forgotten Realms) treated them as being sorcerers. I think maybe if they had something to make them unique it could work, otherwise it should be a subset of wizard.

I like the idea of having an Assassin class, mainly because it sounds cool. It could have some specialized stuff, plus a mix of fighter/rogue stuff. Being a theme could work also.

I think alot of classes currently available could be made into themes for classes. I think that the classes should be more unique, in that they fill a certain niche and don't have as much overlap. Being a fighter should be something that matters from first to twentieth level and you should be able to do more than 'make sandwiches for the wizard' to quote the idea of what a high level fighter could do according to one of my power gamers.

I think 4e's attitude towards classes was a step in the right direction, but it went overboard. They didn't do the best job of balance but then trying to balance fighters agaisnt wizards is nearly impossible.

Would it be too much to ask for the rules to be made in such as way as to prevent absurd Munchkinery? No more Pun-Puns? Ways to make it fun to DM without having to worry about power gamers or bizarre encounter design?

Please???

Also, you (this is directed at no one) remind me of my bird.

Craft (Cheese)
2012-06-07, 09:50 PM
I could be won over by the warlord if they managed to do it right, though. If they can find a way to make the warlord feel mechanically unique and still tie into the rest of the game world thematically, I'd be thrilled. I'm not sure if they'll be able to pull it off though.

Let it work how it does in 4E: A class that specializes in letting other characters take extra out-of-turn actions. It's good enough, and separate enough from what the cleric does, that it'd definitely see play.

(Personally though I'd just give the wizard spells to do that instead of making a whole nother class)

Kerrin
2012-06-07, 09:51 PM
I honestly thing the two of you are the only ones who know at this point...

New topic! What classes from previous editions do you think work better as stand-alone classes and which work better as themes or multiclass builds?

Two chassis: mundane and magical

5e is supposed to be built on modular bones so they should prove it by building all the classes from these two chassis by adding modules, themes, kits, whatever.

I like modular, reusable building blocks that can be combined invarious ways. I think it could be done.

TheAbstruseOne
2012-06-07, 10:03 PM
Honestly though, judging by how many classes you specified you would remove, it seems you fall closer to the "only 3 classes with lots of specializations" from the beginning of my class rambling post. So maybe this whole point is moot. However it is important to note in a system where you have so few classes, each class needs to be able to represent a large variety of archtypes and roles, which means if you have Fighter/Cleric/Wizard, then you can't say "Only clerics can heal", you should be able to get an approximation of the Warlord through use of options of the Fighter class.
Except I specifically said I don't think that only clerics should heal. I want there to be tactical options for combat healing that don't involve the cleric. I want to be able to have a fighter, a thief, and a wizard go into a dungeon and not have it be an automatic death sentence. I don't want to have run a game where one player ends up saying "Fine, I guess I'll be the cleric this time..."

I just want each class to feel unique and cover enough ground that there's multiple build options.


Let it work how it does in 4E: A class that specializes in letting other characters take extra out-of-turn actions. It's good enough, and separate enough from what the cleric does, that it'd definitely see play.
Like that. See, that's a mechanic I can get behind. It's unique enough that it doesn't fit neatly into any of the other classes, but it's got a lot of room for variations in different builds. I still wouldn't want it as one of the "core" classes in the PHB, but I'd love to see something like that in a sourcebook.

TheAbstruseOne
2012-06-07, 10:09 PM
Two chassis: mundane and magical

5e is supposed to be built on modular bones so they should prove it by building all the classes from these two chassis by adding modules, themes, kits, whatever.

I like modular, reusable building blocks that can be combined invarious ways. I think it could be done.

That might be going a bit too far in that direction. You abstract that far out and you're starting to get away from one of those "iconic D&D" mechanics and getting into the sort of bland, generic system that d20 Modern was when it first came out. You're also back to the power source/role system from 4e which means they'll have to create a class that meets every combination of the two for the OCD types and then you get weirdo classes no one ever plays like Runepriest or those off-the-wall psionic ones.

And for the record, I think it should be Fighter/Cleric/Wizard/Rogue if you're only using "core" classes. Fighter = attack, Cleric = divine magic, Wizard = arcane magic, Rogue = skills.

Draz74
2012-06-08, 01:01 AM
If they do a "Warlord" class right, I actually think the class it will overlap with dangerously isn't the Fighter or the Cleric -- it's the Bard. The Bard's the one who's always had the schtick of "I radiate some kind of aura (not a spell) that makes my allies more effective."

I actually wouldn't mind if they basically just made the Bard a Theme/Background combination based on the Warlord class. But I don't think they'll do that, since Bard is more "iconic."

Seerow
2012-06-08, 01:08 AM
If they do a "Warlord" class right, I actually think the class it will overlap with dangerously isn't the Fighter or the Cleric -- it's the Bard. The Bard's the one who's always had the schtick of "I radiate some kind of aura (not a spell) that makes my allies more effective."

I actually wouldn't mind if they basically just made the Bard a Theme/Background combination based on the Warlord class. But I don't think they'll do that, since Bard is more "iconic."


Well like I mentioned, anything more than Fightin' Man, Holy Man, and Magic Man is going to have some overlap in themes and roles. The question of where to draw the line is highly subjective (as evidenced by the arguments over the last page!)

While Warlord and Bard, may share similar roles within the party, I think they're distinct enough flavor-wise; and that can be extrapolated mechanically to remain separate. And clerics... well they overlap with everyone, in much the same way that a wizard overlaps with everyone, except clerics get told up front they're supposed to, because who their god is is supposed to change their role. (You may not have gotten the idea yet that I'm not particularly fond of the cleric class...)

Craft (Cheese)
2012-06-08, 01:29 AM
If they do a "Warlord" class right, I actually think the class it will overlap with dangerously isn't the Fighter or the Cleric -- it's the Bard. The Bard's the one who's always had the schtick of "I radiate some kind of aura (not a spell) that makes my allies more effective."

I actually wouldn't mind if they basically just made the Bard a Theme/Background combination based on the Warlord class. But I don't think they'll do that, since Bard is more "iconic."

I actually think they could fix this by borrowing yet another idea from 4E: Bards are the ultimate dabblers and have the unique ability to multiclass as much as they want without penalty.

My idea was that a Bard can basically gestalt themselves with another class whenever they take a Bard level, but only for the first 1-3 levels of that class. So a 2nd level Bard can be Gestalted with 2 levels of fighter, having all of their 2nd level bard abilities (whatever those are) in addition to the Fighter's better BAB, damage, and Fighter's Surge ability. Then at level 3, they can gestalt in a level of Cleric to get orisons and more 1st-level spells (assuming the 5e bard will have spells like the 3e bard did).

Problem with this is that it makes a 2nd level Bard strictly better than a 2nd-level Fighter, but I'm sure there's some way to work that out. Maybe don't allow gestalting to start until level 3? Either way, it'd probably make the Bard the most fun core class to just mess around with the various build types, just like the 4E bard was.

blackseven
2012-06-08, 04:58 AM
Again, I don't find the argument against "reasonable" to be compelling. Yes, "reasonable" is subjective, but it's not as wildly subjective as people seem to think. If we can't agree that a STR 16 person beating a STR 1 person in an arm wrestling match without a roll is "reasonable", I'm pretty sure I'm not going to want to play with that person.

Yes, the rules SHOULD provide more guidelines to establish a rough benchmark of what is reasonable, especially in regard to opposed check. NO, the list will NOT be exhaustive, nor will it be "objective" in the same sense that arithmetic is objective. There WILL be DM judgement calls, and players WILL disagree with them at times. You can and should lower the occurrence of this within the rules, but ultimately it becomes a question of minutiae. What's a good threshold for "DM calls over player objections?" I think about 5% is acceptable for me. Others seem to put the threshold at something like 0.01%.

The idea that rules have to be airtight is simply untenable. You are going to have to trust a DM to make calls. The rules cannot, ultimately, be detailed and thorough enough to eliminate this.

This is not a "binary" choice. We will all have a favored ranged within the continuum of "discretion" versus "explicit."


That might be going a bit too far in that direction. You abstract that far out and you're starting to get away from one of those "iconic D&D" mechanics and getting into the sort of bland, generic system that d20 Modern was when it first came out.

It's not right for D&D, but I happen to think that the "bland/generic" D20 modern base classes were brilliant. It better reflects the people you would find in world it was trying to emulate (2000+ AD Earth). The Advanced Classes allowed you specialization and strong flavor if you wanted, but you could easily just be some sort of everyman thrust into a situation.


And for the record, I think it should be Fighter/Cleric/Wizard/Rogue if you're only using "core" classes. Fighter = attack, Cleric = divine magic, Wizard = arcane magic, Rogue = skills.


Well like I mentioned, anything more than Fightin' Man, Holy Man, and Magic Man is going to have some overlap in themes and roles.

Although Seerow is right that the very easelier iteration of D&D only had those 3 classes, I think that, by now, the four (Fighter, Cleric, Mage, Thief) are *irreducibly* iconic.

Clawhound
2012-06-08, 08:51 AM
If they can keep the core classes simple, and keep the "special widgets" that each class gets modular enough, then I'm all for as many iconic classes as possible.

Clawhound
2012-06-08, 09:10 AM
As you can't create a perfect rules system, then you will certainly wind up with corner cases. Your tolerance for those corner cases greatly influences your system. The fewer the corner cases that you want, the more rules that you need, but more rules requires more testing and produces even more places where the rules can run wrong.

3.X certainly proved that enough rules still leads to weird results. So more rules, by themselves, are not guaranteed to be better, even if they are more "realistic." Look at the long thread elsewhere on these boards with "common sense" interpretations for the 3.X rules. In many cases, all the extra rules only masked the fact that behind them still sat weird results.

You can simplify the rules, which greatly reduces the number of possible interactions, and reduce breakage, but that also codify the existence of those absurdities. Unexpected breakage is reduced at the cost of unrealism.

That's it. Those are the tensions that exist in any design. Nothing will get rid of that tension, for there is no perfect game.

The only real question is what flaws are you willing to accept.

Kerrin
2012-06-08, 09:47 AM
That might be going a bit too far in that direction. You abstract that far out and you're starting to get away from one of those "iconic D&D" mechanics and getting into the sort of bland, generic system that d20 Modern was when it first came out. You're also back to the power source/role system from 4e which means they'll have to create a class that meets every combination of the two for the OCD types and then you get weirdo classes no one ever plays like Runepriest or those off-the-wall psionic ones.

And for the record, I think it should be Fighter/Cleric/Wizard/Rogue if you're only using "core" classes. Fighter = attack, Cleric = divine magic, Wizard = arcane magic, Rogue = skills.
To clarify a bit, they could take this approach internally within the design team to produce the iconic four classes: fighter, cleric, mage, thief ... and put those in the players handbook. Then they could use the same process to produce the other classes for the players handbook, for example barbarian, monk, bard, etc.

Then they include some set of themes, kits, whateve that players can use to customize their personal flavor of the classes, for example the warrior cleric, the mystic assassin, etc.

Seerow
2012-06-08, 09:48 AM
Again, I don't find the argument against "reasonable" to be compelling. Yes, "reasonable" is subjective, but it's not as wildly subjective as people seem to think. If we can't agree that a STR 16 person beating a STR 1 person in an arm wrestling match without a roll is "reasonable", I'm pretty sure I'm not going to want to play with that person.

Okay, and when it's a str 10 (+0) person against a str 18 guy with training (+7), is it still reasonable? What about when we leave the realm of physical behind and it's a 16 int Wizard's illusion against someone with 6 wis? Or a Bard's bluff against a low wis guard? Are these also reasonable?

And if you say yes to all of these examples, and any case where the modifiers are 7-8 points apart is in fact different enough for it to be reasonable for an auto success, why do the rules not support that? Because the rules as written tell me that with that 8 point difference, the better character is running with a 20% chance of failure. Why is it up to the DM and group to decide "no that doesn't make sense" rather than the rules telling me up front where the line of "This is easy enough to ignore" is drawn.



The idea that rules have to be airtight is simply untenable. You are going to have to trust a DM to make calls. The rules cannot, ultimately, be detailed and thorough enough to eliminate this.

The problem is we're not even looking at really weird corner cases. We're looking at numbers that are explicitly attainable taking actions that are very basic, and generating results that make no sense. I mean I don't care if the rules don't cover every possible situation, but when the rules tell me the strongest most skilled man in the world will fail to break down an average dungeon door about 40% of the time, and the weak frail wizard will actually succeed at it 10% of the time, that is a resolution system that makes no sense and should never have been allowed to be printed by a designer with half a lick of sense.





Although Seerow is right that the very easelier iteration of D&D only had those 3 classes, I think that, by now, the four (Fighter, Cleric, Mage, Thief) are *irreducibly* iconic.

I disagree. The design space between the Fighter and the Rogue traditionally have a lot of overlap, as the two main Mundane classes. Note how with the three specified, each one has a different power source, and will clearly represent different concepts. On the other hand, the only real reason to separate the Fighter and Rogue is if you believe that the Fighter can't do dex based combat (something I disagree with), that the Fighter can't take advantage of underhanded tactics (something I disagree with), or that the Fighter shouldn't be allowed to be skilled (something I heavily disagree with).

The two CAN be represented as separate classes, but ultimately the Fighter is capable of filling the rogue's role in and out of combat, and in a minimalistic game it should. Mainly because making the rogue separate and making something like skill usage a Rogue primary ability gimps the Fighter out of being able to be skilled... leaving him with basically no way to interract with the world out of combat (see: 3.5)

TheAbstruseOne
2012-06-08, 09:49 AM
If they do a "Warlord" class right, I actually think the class it will overlap with dangerously isn't the Fighter or the Cleric -- it's the Bard. The Bard's the one who's always had the schtick of "I radiate some kind of aura (not a spell) that makes my allies more effective."

I actually wouldn't mind if they basically just made the Bard a Theme/Background combination based on the Warlord class. But I don't think they'll do that, since Bard is more "iconic."

The bard's overlap is going to be with the rogue. They both tend to be the jack-of-all-trades type character. If a warlord has an aura, it's going to have a different feeling. The warlord, if done right (and I have been convinced it can be), would be best served as support - sacrificing his own actions to give the rest of the party extra actions or buffs. His flavor will be pointing out weaknesses in enemy lines and things like that which make other characters more effective. Some builds will be more hands-on, while others will be the "lazy warlord" build. They could do something like "Make an attack with disadvantage. If it hits, an ally adjacent to the creature can make an attack with advantage." Or they spend their action to let others attack on their turn. Or they spend action to give their allies advantage next turn. Any healing they do should be built around the current system, though. Instead of the cleric's CLW/HW, they should allow characters to spend hit dice while still in combat. Fits with the feel of the warlord being a non-magical healer and leader.

A bard, on the other hand, should focus on the other side of combat - debuffing enemies. Using an illusion spell to distract enemies giving them disadvantage on attacks. Using a performance aura that as a bit of range to it (20ft or 30ft) that gives +1 to hit or +1 damage or something like that. Immobilize and trip up and distract enemies. 4e was one of the only editions to make bards an actual playable useful class that made sense, and they should build on that success. Give the bard a definable role. Where they're going to overlap is with rogues and illusionists, though. They should have a lot of out-of-combat usefulness as that sort of jack of all trades, especially in social situations (which can step on a lot of rogue builds).

Both of these classes, though, should be in a sourcebook. Their usefulness is solely as support classes. They're never going to be on the front lines of a fight and it's even more true with the warlord that they're rarely going to take front and center in a story (bards at least can dominate social encounters). They're going to be best in large groups, with 6 or 8 players like the old school games used to be sometimes. When you've got more players than your one cleric can keep up with, when there's a lot of tactical options, when your wizard's a blaster build rather than a controller build or the one controller can't keep up with the hordes. Somewhere that one character frequently giving out advantage and disadvantage isn't going to completely break the game, but at the same time is going to be able to do it in critical spots to turn the tide of a battle.

prufock
2012-06-08, 09:50 AM
I just told everyone I'd be running an extra session for playtest on Memorial Day afternoon and anyone who wanted to show up could.

I'd love it if it were this simple, but my gaming group is only 4 people including me. Scheduling is a problem:
- One player works until around 7 every weeknight. We can't play late on weeknights because I work at 830 in the morning and another player works 8 or sometimes 6 am shifts.
- One player has an odd-scheduled job which includes some nights on the weekends, and is doing a course or two at university. Some weekends we don't get together at all, including this coming weekend.

I know we could have a bigger group, but we are picky over who joins us and we prefer to keep the group small (6 would be my absolute max number of players). We've had others join us before, but often our playstyles don't match, personalities don't match, it creates even MORE scheduling conflicts, etc. So we're sticking with 4 for now, unless a good option comes along.

That said, NEXT weekend, we are going to try to do 2 sessions if we can make it work: one for a game of Mutants and Masterminds, one for the Next playtest!

Fatebreaker
2012-06-08, 12:23 PM
For the discussion over mechanics and modeling scenarios, let's back up a step and re-establish some basics:

Mechanics influence play.


Mechanics resolve conflicts within the narrative of the game.


Good mechanics model system intentions.


System intentions should be clearly expressed.
D&D uses a d20 as its most basic method of conflict resolution. Roll a number, beat another number. Because it's a twenty-sided die, it allows for 5% increments with regards towards failure or success. If it was a d100, it would allow for 1% increments, and if it was a d10, it would allow for 10% increments. So, because the system uses a d20, anything the system intends to be reliably (even if infrequently) modeled by the mechanics is something with a 5% or greater chance of success.

As a result, anything which the mechanics claim has a 5% or greater chance of success is something which becomes a legitimate option within the scope of the system.

If the mechanics allow for the reliable repetition of outcomes which are contrary to the intentions of the system, then the mechanics are flawed.

Now, sometimes we'll find weird examples where the rules don't quite interact properly. Hopefully these will be the true "corner cases" at the extremes of the system. But that's not the case here. The values are extreme, yes, but the mechanic itself is not -- from a game design point of view, the mechanic and it's functionality are very much central to the system as a whole!

The real issue is that you are selectively rejecting actions which are mechanically identical but thematically different within the core conflict resolution mechanism of the game.

Take the problem we've been examining, and strip all the fluff from it. It looks like this:


(Attribute) 1 vs. (Attribute) 20 = (d20-4) vs. (d20+5) = 11% or greater chance for (Attribute) 1 to succeed.

This holds true for all (Attribute) values which produce a modifier difference of 9.

The problem here is that the core conflict resolution mechanism of the game claims that this scenario is well worth rolling for -- it has more than a 10% chance of success for the weaker party! Whether it's the cripple out-wrestling Conan or a brain-damaged troll outwitting the Wizened Sage is irrelevant -- your group will find some iteration of this math problem which is fluffed as worth rolling for and some iteration of this math problem which is fluffed as not rolling for. When you start axing some of these scenarios as "unreasonable" and not worth rolling for, you're basically saying that the most fundamental aspect of the game does not work.

It's a playtest. Finding the flaws now and pointing them out in a clear, rational manner is one of the primary points of the exercise. Having a core mechanic return results so ridiculous that the "solution" is to hand-wave them away is indeed a flaw, because it undermines some of the most basic assumptions about the game.

Rule Zero should be an option. The DM and the players should always be able to tailor their game to fit their table. But Rule Zero should be the means by which they customize the game, not the means by which they fix basic mechanical flaws within the core of the system.

Or, put another way, you can Rule Zero and houserule and "be reasonable" with both good systems and bad systems. But wouldn't you rather have a better system to start with than a bad one?

Scots Dragon
2012-06-08, 01:12 PM
One thought in that area I had, though it'd require some careful rebalancing, would be to have every number have its own adjustment, with 10 as a +0, 11 as a +1, 12 as a +2, and each ascending number adding a larger figure. 9 would be a -1, 8 would be a -2, etc.

The end result would be that 18 is a +8 and a 1 is a -9, a distance of 17 rather than 9. It's not a perfect result, but it does reduce the percentage somewhat drastically, and another way of handling it on top of that is to use bell curve rolls (2d10 or 3d8) in place of the sacred cow of d20 rolls.

While providing a similar range of possible results, it has a much lower chance of insane results that make the game impossible to work with. A maximum or minimum roll in either case is a 1-in-100 or 1-in-216 chance, depending on the specific dice used.

Fatebreaker
2012-06-08, 02:01 PM
One thought in that area I had, though it'd require some careful rebalancing, would be to have every number have its own adjustment, with 10 as a +0, 11 as a +1, 12 as a +2, and each ascending number adding a larger figure. 9 would be a -1, 8 would be a -2, etc.

The end result would be that 18 is a +8 and a 1 is a -9, a distance of 17 rather than 9. It's not a perfect result, but it does reduce the percentage somewhat drastically, and another way of handling it on top of that is to use bell curve rolls (2d10 or 3d8) in place of the sacred cow of d20 rolls.

Either of these (alternate bonuses or alternate dice) would be interesting, especially swapping out the d20 for something else. I also liked the suggestion to incorporate Advantage/Disadvantage is you had a modifier some number higher than the DC/your opponents modifier, since that plays into their new Advantage/Disadvantage system, and makes success or failure in extreme cases more unlikely without skewing the math on more equally contested rolls.

For fun, it might be interesting to run the playtest scenario while tweaking the basic parameters of the Attribute Check system. What happens when you use 2d10? 3d8? 4d6? What happens when you scale modifiers in a different sort of way? What happens when you use Advantage/Disadvantage when you have a modifier of +4? +5? Where is the sweet spot, or is there not a sweet spot at all?

Fun questions. I'd be curious to see the answers.

Clawhound
2012-06-08, 02:06 PM
There are a number of different kind of challenges in this game, each with its own measure of randomness/variability/uncertainty. How do you measure that uncertainty? How do you model it?

One way is bonuses. One side gets larger bonuses than the other, then a die is rolled accounting for uncertainty.

One way is situational bonuses. The side that get advantage or disadvantage skews the numbers either favorably or disfavorably.

One way is repetition. We don't measure combat in one hit. Combats take multiple rounds. One lucky roll may influence the combat, but mostly it the static bonuses that influence combat.

One way is verisimilitude. There is no way that sneezing hard hurts you opponent, no matter how well you roll.

So, let's taking bowling as an example. In one round of bowling, a complete amateur can outbowl a seasoned professional just by getting lucky. In no way does the seasoned professional look bad, because bowling is not measured in single throws. All bowlers know that there is variability in the game. That's why bowling measures ten sets. In those ten sets, the likelihood of the far worse player winning is infinitesimal.

Combat is a place where wild whims of luck can work in your favor, or against it. Even a seasoned warrior can make a mistake and get hit by a nobody. However, over the course of the battle, the seasoned warrior's hit points, equipment, and skill will overcome the amateur.

Single rolls are best for short actions. Do you hit? Do you open the door? Did you pick the lock? There's a yes/no answer for this.

Multi-rolls are best against results where randomness must be tempered with skill, such as combat, challenges, diplomacy, and group knowledge rolls. In these situations, luck may help you, but it's not substitute for competence or ability.

Maybe you need to combine systems. The ladder system in most sport combines luck (single games) with endurance (winning many games in a row). This gives the contest uncertainty but still rewards excellence.

What does this mean for D&D? It means that the DMs choice of conflict resolution greatly influences how correctly or incorrectly the conflict resolves. There's no game design way to avoid that.

NichG
2012-06-08, 02:13 PM
Another solution, though I think its one that doesn't feel very D&D, is to have a 'Stat Rank' thing in addition to the numbers. Only characters/creatures with the same Stat Rank in the given statistic of comparison need to roll against eachother. An advantage or disadvantage is Stat Rank creates an auto-succeed/auto-fail situation.

So, if Fighters should just be Strong compared to everyone else, they would get Strength Rank 2 at a certain level, whereas normal humans have Strength Rank 1 (and small animals like cats have Strength Rank 0). That way you can decouple fluffed strength with out-of-control numerical scaling needed to represent those numbers in an opposed contest with the d20 as the core resolution mechanic (since it'd mean that someone who can auto-defeat a human is also probably dealing an extra 30 points of damage per attack, which is undesirable).

Lets say then from this point of view that humanoids default to Rank 1 in all stats. Smaller animals have Str Rank 0, Int Rank 0, Wis Rank 2 (if we take Wis to be perception and assume that animals all have a better sense of smell). Ogres, trolls, and giants have Str Rank 2. Dragons, Kraken, etc have Str Rank 3. Illithids might have Int Rank 2. And so on.

A bad disease might lower your Stat Rank instead of just lowering your stat (therefore modeling the diseased commoner example). Infants and children might just have Str Rank 0, so no they can't beat the trained soldier at a Str contest.

Also, if Fighters are the only people to get stuff like Str Rank 2, it does help make them be (slightly) more quadratic, in that there's a point where impossible feats of strength open up to them. Though it doesn't nearly fix the versatility problem.

Stubbazubba
2012-06-08, 02:14 PM
Again, I don't find the argument against "reasonable" to be compelling. Yes, "reasonable" is subjective, but it's not as wildly subjective as people seem to think. If we can't agree that a STR 16 person beating a STR 1 person in an arm wrestling match without a roll is "reasonable", I'm pretty sure I'm not going to want to play with that person.

Then, as I've already suggested, change it to a non-STR or even any physical-based scenario. Let's say the guard has a 10 in his WIS (human average) and the level 1 Bard has a 20 in CHA (human limit). Now, the Bard is trying to bluff his way past the guard. Going on sheer reasonableness, the guard might catch a lucky break and see through the Bard (the Bard's only level 1, and these sorts of challenges are supposed to be, well, challenging), so you call for a roll. In fact, that is only a 5-point difference (d20+0 vs. d20+5) on that roll! That means there's over a 1 in 4 chance that the dice will favor the guard. Clearly this is not a foregone conclusion by any means.

But now let's change the fluff; that same Bard is now level 20, and trying to bluff his way into Olympus, and a god is standing guard outside the gate. This god's WIS is 30 (god limit), vs. the Bard's 20 CHA (human limit). At this point, you're dealing with a supernatural creature with the power to re-write physics with divine magic, who has been guarding this gate from infernal threats for millennia; there's no way some human Bard, even the best of human Bards, who had to actually roll against that not-a-Rhodes-Scholar guard, and his ability to Bluff hasn't increased since, is going to just turn-a-phrase and convince him he's actually a god in disguise. Reasonableness demands that no roll be allowed, you fail automatically. But, you guessed it, this is only a 5-pt. difference on the actual roll, and the Bard, in reality, has a 26.25% chance of succeeding, and that's not a foregone conclusion at all.

So within the same class, between levels, the rule is wont to be applied inconsistently.

But let's change it up again; a level 20 Fighter (STR 20, +2 on grapples) is locked in epic combat with a dragon (STR 30), who grapples him to the ground with a single paw. The Fighter says he wants to try and pry one or two of the the claws off of him so he can escape. The DM laughs at him and dismisses it; no mortal can just lift a giant's claw off of himself. As a matter of fact, assuming the absence of size bonuses (which are absent from the playtest docs), there was only a difference of 3 on that roll! The Fighter had almost a 40% chance of success! But it was unreasonable because the DM considered the image it would produce unreasonable.

But then, level 20 Bard is trying to trick the same dragon into leaving his lair, and he has a CHA of 20, but a +2 to bluff checks when he casts a certain Illusion spell. The dragon has a WIS of 32. The Bard says he wants to cast an illusion and bluff the dragon out. The DM, not realizing that, mechanically, the Bard's odds are less than the Fighter's, looks at it and says, "Ooh, that's clever. Go for it." And the Bard succeeds!

Man, wasn't that heroic? Isn't everyone having fun? But guess which class just got hosed because it's not magic, and the rules are very prone to be applied inconsistently, based on what's "reasonable"? So the rule isn't likely to be consistently applied between different classes of the same level, either. And we're talking very small differences in actual modifier, here. This kind of thing could come up every session possibly multiple times, ergo it is not a .01% deal. This is like 1/3 of the game is completely up to the DM's arbitration instead of using any mechanics whatsoever.


The idea that rules have to be airtight is simply untenable. You are going to have to trust a DM to make calls.

And I totally agree. But the DM making a call should not be the first rule for all skill checks. That's not Rule 0 anymore, that's Rule 1. The mechanics are there to make the vast majority of things fair and workable, and fun; the DM's job is to arbitrate in corner cases, which should come up maybe, maybe once a session, at the very most, but this? This is no longer a corner case; trying to grapple a dragon isn't a corner case, nor is bluffing a guard, be he mundane or celestial. This is what your class does, and the DM is encouraged to deny you your own schtick if it seems fishy when he pictures it. Which will be different for every DM. It throws a huge portion of the game out, and replaces it with DM fiat instead. If that's the case, then I don't need to pay WotC or anyone else $150 to ask another human being if my character can auto-succeed at this, 'cause hey, he's like Hercules and that guy's a cripple.

When Rule 0 is the only way the rules actually work, you're paying for vaporware. Rule 0 is the only rule you need, then. Rule 0 is and ought to remain a last resort, when mechanics have failed for whatever rare scenario, not built into the mechanics to remove all responsibility from the designers. This is what's left (http://www.xkcd.org/971/).

EDIT: Put another way: The mechanics of the game form a bridge across a wide chasm. It's imperfect, but 95/100 times you can get across safely; Rule 0 is a team of firefighters with a giant trampoline on the ground far below, ready to catch you on the off-chance that you fall through. If, however, the bridge is missing 1/4 or 1/3 of the chasm's length, then Rule 0 is forced to become part of the bridge itself, and do a lot more heavy lifting than is appropriate. That's not good for the game, but it is explicitly what Next has done; built an incomplete mechanic, and then stapled on Rule 0 to make it complete. They've hard-coded the Oberoni fallacy into the rules themselves, which I think is a harbringer of the last days, IIRC from the last time I read the Book of Revelations.

So either the d20 is a completely incompatible RNG for the numbers they're using, or the skill system, if nothing else, needs to be overhauled and auto-successes/failures determined in some consistent, objective way, at least by RAW.

kyoryu
2012-06-08, 02:59 PM
This is what your class does, and the DM is encouraged to deny you your own schtick if it seems fishy when he pictures it. Which will be different for every DM. It throws a huge portion of the game out, and replaces it with DM fiat instead. If that's the case, then I don't need to pay WotC or anyone else $150 to ask another human being if my character can auto-succeed at this, 'cause hey, he's like Hercules and that guy's a cripple.

And I believe this is the crux of the argument, the role of the DM in the game.

I'm a bit more on the old-school side in this, I guess. I don't worry about bad DMs, because in the end, bad DMs will make bad games no matter what. If the DM assumes the world knows what the PCs are up to, have unlimited resources, and magically can think of everything the PCs come up, then it will be a pretty terrible game no matter what. The DM screwing me on a couple of skill checks is pretty minor compared to that.

Essentially, I view the primary interaction as players and DM. Others view the primary interaction as the players and DM interacting through the rules. They're both valid approaches.

Seerow
2012-06-08, 03:06 PM
And I believe this is the crux of the argument, the role of the DM in the game.

I'm a bit more on the old-school side in this, I guess. I don't worry about bad DMs, because in the end, bad DMs will make bad games no matter what. If the DM assumes the world knows what the PCs are up to, have unlimited resources, and magically can think of everything the PCs come up, then it will be a pretty terrible game no matter what. The DM screwing me on a couple of skill checks is pretty minor compared to that.

See I disagree. A DM can be mediocre but still run a fun game with a well-defined system, but when put into a situation where he has to make ALL of the decisions, the game is terrible. Most DMs are not game designers and don't want to be. For DMs who do want to be game designers, mother may I and open ended rules are great. For anyone else, they lead to a bunch of meaningless arguments and bad feelings that could have been avoided with a better ruleset.

Stubbazubba
2012-06-08, 03:21 PM
I'm a bit more on the old-school side in this, I guess. I don't worry about bad DMs,

My examples weren't even about bad DM's. You're right, a bad DM can never be made good by rules, and that's not what I'm getting at. I'm saying it's unreasonable to think that even a good DM can simply adjudicate that much and be consistent and fair to everyone. That is, in my opinion, what rules are for. To help good DMs be great ones, not give them a hundred chances of accidentally shafting a player every single session.


Essentially, I view the primary interaction as players and DM. Others view the primary interaction as the players and DM interacting through the rules. They're both valid approaches.

Perhaps I'm taking you too literally here, but doesn't the former not require rules at all? Freeform RP is kind of cool; it moves fast, it's very improv-based, and thus its way more flexible than anything else. I've tried it and really enjoyed it, and I didn't put down a dime to do so. I don't think D&D has ever been that game, though. The rules are there to provide the level of structure implied in the latter approach. At least that's my take on things. In your opinion, what role do rules serve in the former, more old-school approach, if not to regulate DM-player interaction?

kyoryu
2012-06-08, 03:35 PM
See I disagree. A DM can be mediocre but still run a fun game with a well-defined system, but when put into a situation where he has to make ALL of the decisions, the game is terrible. Most DMs are not game designers and don't want to be. For DMs who do want to be game designers, mother may I and open ended rules are great. For anyone else, they lead to a bunch of meaningless arguments and bad feelings that could have been avoided with a better ruleset.

Your terminology shows your bias.

Any interesting game I've been in has too many situations that can't be codified into rules. Okay, you just punched a guard. How do the others react? What happens when the survivors (if any) report this? What happens if any bystanders report this? I just can't imagine any scenario where these things are all covered by rules.

The impact of scenarios like this upon the overall play experience is so much higher than simply determining where the specific margin is between "possible but unlikely" and "impossible" that I don't see it as a worthwhile cause of concern. Mostly because I probably won't be attempting "impossible" things anyway, as it's a waste of time, and NPCs I run generally won't be doing that anyway.


My examples weren't even about bad DM's. You're right, a bad DM can never be made good by rules, and that's not what I'm getting at. I'm saying it's unreasonable to think that even a good DM can simply adjudicate that much and be consistent and fair to everyone. That is, in my opinion, what rules are for. To help good DMs be great ones, not give them a hundred chances of accidentally shafting a player every single session.

Perhaps I'm taking you too literally here, but doesn't the former not require rules at all? Freeform RP is kind of cool; it moves fast, it's very improv-based, and thus its way more flexible than anything else. I've tried it and really enjoyed it, and I didn't put down a dime to do so. I don't think D&D has ever been that game, though. The rules are there to provide the level of structure implied in the latter approach. At least that's my take on things. In your opinion, what role do rules serve in the former, more old-school approach, if not to regulate DM-player interaction?

At the extreme, sure. In reality, the game mostly plays by the rules, with the DM having the ability to override them as appropriate. Which is, yes, determined by the DM. See above for why that doesn't bother me too much.

I don't think the case of "disallow a contested roll when the result is pretty clear" is that much of an area - again, this is only going to come up in severely unbalanced scenarios, and even then one would assume that in the vast majority of these decisions, the players and DM would agree. There's only a handful of edge-case scenarios where there might be disagreement, and even then, it's probably not going to be that big of a deal.

Maybe some of that comes from playing sports, where refs make bad calls, and you just get on with it.

Fatebreaker
2012-06-08, 03:37 PM
Another solution, though I think its one that doesn't feel very D&D, is to have a 'Stat Rank' thing in addition to the numbers. Only characters/creatures with the same Stat Rank in the given statistic of comparison need to roll against eachother. An advantage or disadvantage is Stat Rank creates an auto-succeed/auto-fail situation.

You're right. That doesn't feel very D&D. But I'd be very curious to see how it would work out, because it neatly helps to sidestep some of the weird math.

That huge dragon pinning the Fighter to the ground? Yeah, he's just too big to move. Raw strength can only do so much against the combined powers of mass and leverage. It also creates an alternative means to acquiring power, and options are a good thing.

You could also disconnect your Stat Rank from your Stat Score if you want the Stat Rank to fill in for a bunch of fiddly bonuses. Instead of dwarves getting a +x against poison or +y to stamina-related checks, they get a higher Constitution Rank. A Con 14 elf and a Con 14 dwarf both die when you stab them in the face, but the ConR 1 elf has less stamina, is more vulnerable to poisons and disease, and so on than the ConR 2 dwarf.

Again, not very D&D, but... that doesn't mean it's a bad idea. I'd be curious to see this in action.


And I believe this is the crux of the argument, the role of the DM in the game.

-snip-

Essentially, I view the primary interaction as players and DM. Others view the primary interaction as the players and DM interacting through the rules. They're both valid approaches.

They are, in fact, both valid approaches, and I've had fun with both. However, for the amount of money we're talking about, the rules need to be good rules.

If the rules are not good rules, then under either approach, buying them is bad. If you're a player+DM group, you didn't need them, and now your baseline for when you break out Rule Zero is lower, which means more work on the group just to break even. If you're a player+DM+rules group, you did need those rules, and now the bad rules are actively impeding you. In either case, why would I pay money for rules that make it harder for me to have fun?

I am not buying my friends. I am buying the rules. I expect my investment to facilitate, not hinder, my fun.

Stubbazubba
2012-06-08, 03:59 PM
At the extreme, sure. In reality, the game mostly plays by the rules, with the DM having the ability to override them as appropriate. Which is, yes, determined by the DM. See above for why that doesn't bother me too much.

I'm sincerely curious to hear your thoughts on what the rules are for in an old-school game, though.


I don't think the case of "disallow a contested roll when the result is pretty clear" is that much of an area - again, this is only going to come up in severely unbalanced scenarios, and even then one would assume that in the vast majority of these decisions, the players and DM would agree. There's only a handful of edge-case scenarios where there might be disagreement, and even then, it's probably not going to be that big of a deal.

Well, um, I just wrote, like, a page of not-edge-case scenarios where it could come up in almost any session. And is it OK to be unfair to the Fighter so long as he agrees to it?


Maybe some of that comes from playing sports, where refs make bad calls, and you just get on with it.

And some games are won or lost based on one ref's call, not to mention how many are heavily affected by an apparently biased ref. I mean, I've played sports, too, and I got on with it, but not without some level of resentment towards the ref. Why does that make it OK for an RPG?

I don't mean to sound rude, but so far all you've done is dismiss actual arguments with a wave of your hand and something resembling an anecdote. It's not that I want to argue, but I want to know what the thinking is behind your preferred approach. If it's nothing more than, "Well, it's not a huge deal," then that still begs the question of why not use (theoretical) rules which would make it an even less big deal and take so much of that onus off the DM? If there's more to it than that, please explain.

In other news, I also like the idea of non-arithmetic Ranks for either skills or stats, but it would need to be done pretty carefully.

Seerow
2012-06-08, 04:14 PM
On an unrelated note, fun thing about D&D Next that was noticed elsewhere while looking at the effects of strength:

In 5e, a 10str character can carry 100lbs unencumbered. If we compare this against the light load of characters in 3.5e, a 10str 5e character is directly equivalent to a 18str 3.5e character! The heavy load in 5e for 10str caps out at 200, the same as a 15str character in 3.5e.


But then, when we move up to a 20str character, he can only carry 200 unencumbered, or 400 fully encumbered. The max encumbrance lines up with what a 20str 3.5e character gets, while the unencumbered lines up with a 23str character.


Basically, in 5e characters got stronger. How much stronger depends on what load amount we're looking at, encumbered or unencumbered, but it's higher across the board. The most annoying thing though is that it is completely linear. Combine that with low attribute caps, and tough guys are weaker than they were before, both in the absolute cap sense, and in the relative to the average joe sense.

In 3.5, my 34 strength character can carry around 2800lbs as a heavy load. In 5e, I won't ever be able to carry more than 400lbs. In 3.5e, a 20str character could carry 4x more than a 10str character, in 5e he can carry only twice as much.

In mythology, Beowulf is as strong as any 30 normal men. in 3.5e this meant being able to effortlessly carry 30x33 = 990 lbs, which translates to 35 strength, which isn't unreasonable for a high level character. In 5e, to be as strong as 30 men requires being able to effortlessly carry 3000lbs, which requires a strength score of 300, literally 10x what Gods get capped at.

tl;dr: Wizards don't even need Fighters as pack mules anymore, and encumbrance scaling means Beowulf is as strong as any 10 gods combined.

Person_Man
2012-06-08, 05:10 PM
So I've made a couple of posts over on the WotC D&D Next Feedback forums, and I have to say that I think WotC is screwed.

First, it seems like the tenor of conversation over there is just terrible. I see a lot of people being personally berated and insulted, myself included. Maybe I've just been sheltered by the Playground for too long, but some people over there seem to be real jerks, and the mods never seem to do anything other then move threads to the appropriate forum and delete WotC intellectual property (which makes me think there clearly is not going to be an SRD).

Second, there seems to be a highly fractured fan base. People generally want it to be more like their favorite edition (or something entirely new). And since each editions are dramatically different from each other, no one is particularly happy. Some people explicitly want balanced combat, classes, and attributes, other people explicitly hate those things because they feel that it detracts from roleplaying, simulation, world building, their preferred fluff, etc. And D&D Next doesn't deliver on either.

Third, every mechanic has a highly vocal set of detractors, and none of the mechanics seem particularly inventive or interesting.


Has anyone seen any sort of consensus on D&D Next?

Seerow
2012-06-08, 05:20 PM
So I've made a couple of posts over on the WotC D&D Next Feedback forums, and I have to say that I think WotC is screwed.

First, it seems like the tenor of conversation over there is just terrible. I see a lot of people being personally berated and insulted, myself included. Maybe I've just been sheltered by the Playground for too long, but some people over there seem to be real jerks, and the mods never seem to do anything other then move threads to the appropriate forum and delete WotC intellectual property (which makes me think there clearly is not going to be an SRD).

The WotC mods do delete stuff, they're just usually pretty slow about it, and as far as I can tell there's no warnings, and bans only occur for major things, so there's nothing to discourage people from doing the same thing over and over. (In fact I got a post deleted for calling out one guy for trolling and derailing a topic repeatedly)



Second, there seems to be a highly fractured fan base. People generally want it to be more like their favorite edition (or something entirely new). And since each editions are dramatically different from each other, no one is particularly happy. Some people explicitly want balanced combat, classes, and attributes, other people explicitly hate those things because they feel that it detracts from roleplaying, simulation, world building, their preferred fluff, etc. And D&D Next doesn't deliver on either.

Honestly, I see a lot of unanimity on stuff over on the WotC forums that makes me bang my head against the wall. Like the Bounded Accuracy thing? Feedback is overwhelmingly positive from people who haven't thought any of the ramifications through and/or people who want characters to be perpetually low level. Also it seems the largest segment of posters there are the type who thinks that "Improvise" is a valid class feature.


Third, every mechanic has a highly vocal set of detractors, and none of the mechanics seem particularly inventive or interesting.

Yeah just about the only mechanic that seems well received is Advantage/Disadvantage. I haven't seen anyone really complain they don't like that. And honestly when you think about it, that was just about the only real 'mechanic' that D&D next introduced. There's not a lot to love or hate, most people are loving or hating the 'feel' of the material released in the playtest.

TheAbstruseOne
2012-06-08, 05:32 PM
No more maths please. My brain is hurting...

The thing to keep in mind is that a bad DM is going to be a bad DM and a good DM is going to be a good DM. If you have a lot of rules rather than relying on DM rulings, it can mitigate the badness of the bad DM but it's also going to hamstring the good DM. If you have few rules but lots of rulings, the good DM is going to have room to shine but the bad DM is going to be worse. There's no right or wrong answer here, only compromise between how far you want the sliders to go. If you bring them until they meet all the way in the middle, you've got Skyrim or World of Warcraft - the DM is a computer that just does all the math and tells you what happened based on algorithms. It's not an RPG anymore. If you push them all the way out, you've got a group storytelling exercise rather than a roleplaying game.

As far as the tone of conversation, the WotC forums have always been, for lack of a better word, a bit of a troll haven. Right now it's the focus point of everyone who hates WotC regardless of what they think of 5e. There are a lot of people who are jumping to conclusions and making wild assumptions based off the playtest, then saying that Next is horrible and sucks and they'll never buy it because of it. I frankly gave up on trying to have conversations there because every single thread seems to get derailed.

They're also updating at an insanely fast rate, with probably 10-30 posts every hour on all four groups each 24/7. They'd have to have a small army of moderators to keep up. Poor Trevor sure as hell can't do it alone. Best they can do is shuffle everything around to make sure it's at least in the right place, check on the posts that get reported (like the guy that called me an autistic "derogatory term for homosexual which is also a synonym for 'cigarette' in the UK" because I didn't agree with...something, I can't even remember what anymore), and make sure their IP isn't getting violated.

As far as an OGL goes, it's way too early to tell. They are, however, having to enforce their IP strictly in the playtest sections because if they don't enforce it, they can lose it. Let one person get away with posting stats wholesale and you have someone else posting it, then someone else, then someone else putting it on their blog and making money off it through banner ads or possibly printing it out and distributing it. You can't copyright rules, only the way in which they're presented. So that's why the "don't distribute or share or post excerpts or anything like that" is part of the Playtest Agreement - so they have an avenue for legal recourse if some third party publisher decides to print their own version of Next. It's may also come as word-on-high from Hasbro for them to be able to even do the playtest - make sure it's strictly controlled. Otherwise, WotC just talked Hasbro into giving away something they could be making money on.

Craft (Cheese)
2012-06-08, 06:18 PM
As far as an OGL goes, it's way too early to tell. They are, however, having to enforce their IP strictly in the playtest sections because if they don't enforce it, they can lose it.

This is actually closer to literally true than you may think: There's precedent for courts denying a copyright holder because they didn't attempt to enforce their copyright strictly enough. (I'd have to ask my lawyer friends if you want specific cases.) This is why IP holders fight so hard to take down even seemingly harmless things.

TheAbstruseOne
2012-06-08, 06:26 PM
This is actually closer to literally true than you may think: There's precedent for courts denying a copyright holder because they didn't attempt to enforce their copyright strictly enough. (I'd have to ask my lawyer friends if you want specific cases.) This is why IP holders fight so hard to take down even seemingly harmless things.

Oh, I know. Also why McDonald's sues everyone who puts "Mc" in front of something. They forced the closure and bankruptcy of an elderly couple who owned McDonald's Bakery even though their last name was McDonald and the bakery had been in the family since the 1800s or 1900s, I can't remember the exact year but it was a good few decades before the first McDonald's Hamburger Stand opened. They've done the same to a McDonald's car dealership and a few other places as well.

It's also how the British government lost the trademark on the police phone box. When a big budget Fox TV movie was being made in 1996, the police department sued because they were using the image of their phone box. They wanted royalties. The court ruled that because they never defended the trademark in the previous 30+ years and the phone box was more associated with the subject of the film than the police department, they didn't have the trademark. That means that police departments in England are now paying royalties to the BBC because the TARDIS is a trademark and the police phone box it was based off of is not.

Oh, um, the movie in question was the crappy Doctor Who movie. Figured that part went without saying...

Flickerdart
2012-06-08, 07:10 PM
That means that police departments in England are now paying royalties to the BBC because the TARDIS is a trademark and the police phone box it was based off of is not.
Pretty sure the BBC is government owned, so the government is paying itself money?

Craft (Cheese)
2012-06-08, 07:15 PM
Pretty sure the BBC is government owned, so the government is paying itself money?

Governments are rarely cohesive, single actors. They're built of chambers and factions and departments and subdepartments that often work at cross-purposes from each other. Which is how you end up with crazy stuff like anti-tobacco laws and pro-tobacco subsidies being passed in the same bill, let alone relatively minor stuff like this particular case.


But we're starting to get really off topic now, so new topic: What does everyone want WotC to do with Wild Shape?

TheAbstruseOne
2012-06-08, 07:19 PM
But we're starting to get really off topic now, so new topic: What does everyone want WotC to do with Wild Shape?
Kill it with fire.

Sorry, I hate druids. Hate hate hate hate hate druids. Every edition and every incarnation. They're either underpowered enough to be useless or broken as hell with nothing in between. And the rules are always complicated as all hell to figure out. Want absolutely nothing whatsoever to do with them. Bad evil bad hisssssss!

Well, you did ask my opinion of what they should do, not what I expect them to do...

Tehnar
2012-06-08, 07:20 PM
The thing to keep in mind is that a bad DM is going to be a bad DM and a good DM is going to be a good DM. If you have a lot of rules rather than relying on DM rulings, it can mitigate the badness of the bad DM but it's also going to hamstring the good DM. If you have few rules but lots of rulings, the good DM is going to have room to shine but the bad DM is going to be worse. There's no right or wrong answer here, only compromise between how far you want the sliders to go. If you bring them until they meet all the way in the middle, you've got Skyrim or World of Warcraft - the DM is a computer that just does all the math and tells you what happened based on algorithms. It's not an RPG anymore. If you push them all the way out, you've got a group storytelling exercise rather than a roleplaying game.



I don't agree that a lot of rules will hamstring a good DM, because a DM is much more then making rulings. In fact I feel that a good set of rules will help a good DM because it will provide verisimilitude and he/she will have more time to develop a story, NPCs, encounters and so forth.

Seerow
2012-06-08, 07:20 PM
But we're starting to get really off topic now, so new topic: What does everyone want WotC to do with Wild Shape?

I don't mind wild shape if it's restricted to small animals. The druid turning into a house cat or a bird to avoid detection or get away is fine. It's a nifty utility.

Wild Shaping available as a combat application should not be available to any class that has other major abilities to go with it. I mean, you can have a class completely balanced around wildshape and the flexibility it offers both in and out of combat. If they bring back combat wildshape, it should be on a class that is dedicated to nothing but that.

Craft (Cheese)
2012-06-08, 07:27 PM
I don't mind wild shape if it's restricted to small animals. The druid turning into a house cat or a bird to avoid detection or get away is fine. It's a nifty utility.

Wild Shaping available as a combat application should not be available to any class that has other major abilities to go with it. I mean, you can have a class completely balanced around wildshape and the flexibility it offers both in and out of combat. If they bring back combat wildshape, it should be on a class that is dedicated to nothing but that.

I've actually had an idea bouncing around in my head for a while for a class based entirely around an improved form of wild shape, except what forms they can take is limited by their environment. Like, the forms available when they're in a forest is different from when they're in a swamp, jungle, or tundra. Add in a limited ability to be able to change their environment (like, say, suddenly flooding a forest with water so they gain access to swamp forms, or having trees shoot up through the floor to get forest powers) and I think there's enough to build a really interesting, unique class around it.

Seerow
2012-06-08, 07:34 PM
I've actually had an idea bouncing around in my head for a while for a class based entirely around an improved form of wild shape, except what forms they can take is limited by their environment. Like, the forms available when they're in a forest is different from when they're in a swamp, jungle, or tundra. Add in a limited ability to be able to change their environment (like, say, suddenly flooding a forest with water so they gain access to swamp forms, or having trees shoot up through the floor to get forest powers) and I think there's enough to build a really interesting, unique class around it.

That does sound interesting, but fairly complex. Especially for drawing the line for things like Jungle vs Forrest. Or would you make very broad terrain types (something like "Open" "Wooded" "Swampy" "Water" "Underground")? Because unless they are that broad, categorizing every potential form could get to be a real pain.

As an aside, for a pure wildshape class, I'd be more interested in being able to change themselves to adapt, rather than changing the environment. I mean imagine someone taking a tiger form and adding fins/gills for an aquatic environment, or picking up a lizard style tail for the tail sweep attack. At least I think being a shapeshifting chimeric would be pretty interesting.

Craft (Cheese)
2012-06-08, 07:40 PM
As an aside, for a pure wildshape class, I'd be more interested in being able to change themselves to adapt, rather than changing the environment. I mean imagine someone taking a tiger form and adding fins/gills for an aquatic environment, or picking up a lizard style tail for the tail sweep attack. At least I think being a shapeshifting chimeric would be pretty interesting.

That's also pretty cool, but I intended for the environment-changing mechanics to be useful for more than just letting you get new wildshape forms. Things like making a volcano pop out of the ground and cover an opposing army with pyroclastic flows.


I'd probably just implement both of our ideas though. High-powered? Yes, but still cool and fun, and a druid conception I'd actually want to play.

Fatebreaker
2012-06-08, 08:36 PM
No more maths please. My brain is hurting...

The math happens whether you think about it or not. Best to actively question whether it's actually working.


The thing to keep in mind is that a bad DM is going to be a bad DM and a good DM is going to be a good DM. If you have a lot of rules rather than relying on DM rulings, it can mitigate the badness of the bad DM but it's also going to hamstring the good DM. If you have few rules but lots of rulings, the good DM is going to have room to shine but the bad DM is going to be worse.

Well, "good DM" and "bad DM" are not absolute values. A good DM in one system can be a bad DM in another. This is especially true if the DM does not understand how the system actually functions, as opposed to how it claims to function.

Neither rules-light nor rules-heavy are inherently better systems. While we can debate the value of each, the supremacy of one over the other is not the point I've been putting forward (nor, I believe, are those whom I find myself agreeing with, such as Seerow or Stubbazubba). The real point is that whatever mechanics a system does use should work, and work in such a way that it actually models the intentions of the system. Preferably, they should also be elegant. That is to say, as streamlined as possible for ease of application and as clearly expressed as is effective. But functionality is the priority. A mechanic which puts a DM in the position of allowing blatantly ludicrous outcomes to occur with regularity or using Rule Zero with regularity is not what I would define as a functional rule, especially when it's a core mechanic for a "costs real money-dollars" system.


I've actually had an idea bouncing around in my head for a while for a class based entirely around an improved form of wild shape, except what forms they can take is limited by their environment. Like, the forms available when they're in a forest is different from when they're in a swamp, jungle, or tundra. Add in a limited ability to be able to change their environment (like, say, suddenly flooding a forest with water so they gain access to swamp forms, or having trees shoot up through the floor to get forest powers) and I think there's enough to build a really interesting, unique class around it.

I'd be curious to see a shapeshifter class who could also change the environment. Very high-end on the power level, but let's face it, that's what higher levels are for, eh?

I think it would actually work really well in an RPG themed around a Magic: the Gathering style of world, since that's a setting where the different environments are key to the game.

That said, I agree with Seerow on Wild Shape in general: either make that the core focus of the class, or make it limited to a utility rather than combat ability.

-----

Somewhere up-thread, there was a question regarding classes. Things like, "which classes should exist," and "which classes could be themes" and stuff like that.

To toss my two copper into that, I personally think that 4e had the right idea with the Defender/Leader/Controller/Striker vibe. Define the class by what role it needs to perform. Add fluff and abilities to match.

Want to play a divine class? There's a variant for that in each role. Divine leader? Cleric. Divine defender? Paladin. Divine Striker? Avenger. Divine controller? Invoker. Allow multiclassing, or have some sort of modular system that allows you to add on elements from another role.

Disclaimer: I'm not saying to lift the whole 4e system. I mean, sure, I liked it, but I can see legitimate reasons for people to disagree with it. I'm saying you have four roles, over which you add power sources, themes, mechanics, and flavor which suit your character concept. A divine defender should play differently than a martial defender, while still fulfilling the same basic "defender" class role.

Craft (Cheese)
2012-06-08, 08:57 PM
To toss my two copper into that, I personally think that 4e had the right idea with the Defender/Leader/Controller/Striker vibe. Define the class by what role it needs to perform. Add fluff and abilities to match.

Want to play a divine class? There's a variant for that in each role. Divine leader? Cleric. Divine defender? Paladin. Divine Striker? Avenger. Divine controller? Invoker. Allow multiclassing, or have some sort of modular system that allows you to add on elements from another role.

I still think that classes don't really make sense in the type of game 4E was trying to be. Really, the more I think about it, the more reasons I'm finding to question class systems in general.

TheAbstruseOne
2012-06-08, 08:58 PM
I've actually had an idea bouncing around in my head for a while for a class based entirely around an improved form of wild shape, except what forms they can take is limited by their environment. Like, the forms available when they're in a forest is different from when they're in a swamp, jungle, or tundra. Add in a limited ability to be able to change their environment (like, say, suddenly flooding a forest with water so they gain access to swamp forms, or having trees shoot up through the floor to get forest powers) and I think there's enough to build a really interesting, unique class around it.
If you want some inspiration, check out Shadowrun's spirit summoning rules for shamen (1st-3rd ed).

Basically a shaman can only summon a nature spirit of the location they're at, which becomes the spirit's "domain". So you want to summon a forest spirit, you have to summon it in the forest and the spirit is limited to the forest.

The fun comes in when you get two shamen in a fight. Say that same forest has a river in it and it's under the open sky on the side of a mountain. That's four different domains. And each shaman individually can decide which domain they're "in" when summoning a spirit regardless of what the other shaman chooses and without moving. So Shaman A summons a forest spirit. Shaman B summons a mountain spirit. Shaman A decides the forest spirit isn't going to work so summons a water spirit (and though not having moved an inch physically, he has declared that he has changed domains and is now in the water domain, thus no longer having control over the forest spirit).

It gets so much fun at times that one of the examples giving in one sourcebook is an argument over whether someone was in the front seat or back seat of a car because it determined which domain they were in. If they were in the back seat which was over the sidewalk, they were in the city domain. If they were in the front seat, they were in the living room of the house and thus in a hearth domain.

TheAbstruseOne
2012-06-08, 09:01 PM
I still think that classes don't really make sense in the type of game 4E was trying to be. Really, the more I think about it, the more reasons I'm finding to question class systems in general.
Classes are staying. That's not in question. Take classes out of the game and it doesn't feel like Dungeons & Dragons anymore. It's now just "Generic Fantasy D20". If I wanted that, there's a ton of OSR games out there.

However, you're right that 4e roles make no sense in Next. They're working to get back to allowing more fine-tune customization of what your character can do, which precludes assigning classes specific roles. You're a fighter? You're not pigeonholed as the tank because you can take the Slayer theme and be a damage-dealing machine.

Seerow
2012-06-08, 09:02 PM
I still think that classes don't really make sense in the type of game 4E was trying to be. Really, the more I think about it, the more reasons I'm finding to question class systems in general.

Class systems are nice because they provide discrete blocks of power, and package together abilities that in a point buy system players might ignore in favor of more power. For example in a class system you can be sure players will pick up some amount of non-combat utility through their class levels. In a point buy system, you always have that guy who decides it's better to have 400 points invested into "Swording that other guy in the face" and forgets things like "Talking to people" or "See things"

(Not that 3.5 is really better about that given the scarcity of skill points and the way they can be distributed, but ideally a class system ensure that any class has ability to take part to some degree in all areas of play)

Craft (Cheese)
2012-06-08, 09:07 PM
Oh, of course. They're not gonna ditch classes in D&D no matter what, something that I think really held them back in 4E's design. Still though, I can't help but wonder if there really are any legitimate areas where classes make for a better game design than some other system.

Other than being the system most people are already familiar with, of course. I don't think following conventions for their own sake is a legitimate design purpose.


The fun comes in when you get two shamen in a fight. Say that same forest has a river in it and it's under the open sky on the side of a mountain. That's four different domains. And each shaman individually can decide which domain they're "in" when summoning a spirit regardless of what the other shaman chooses and without moving. So Shaman A summons a forest spirit. Shaman B summons a mountain spirit. Shaman A decides the forest spirit isn't going to work so summons a water spirit (and though not having moved an inch physically, he has declared that he has changed domains and is now in the water domain, thus no longer having control over the forest spirit).

Interesting! Question: Can Shaman B shift to the Forest domain and gain control over the forest spirit that Shaman A summoned?



Class systems are nice because they provide discrete blocks of power

So they give the player... less choice? I can understand making the game less complicated is worth it sometimes, but I don't see how providing less choice to the player, all else remaining equal, is ever a good thing.

(Unless you meant something else and I'm just an idiot.)


and package together abilities that in a point buy system players might ignore in favor of more power. For example in a class system you can be sure players will pick up some amount of non-combat utility through their class levels. In a point buy system, you always have that guy who decides it's better to have 400 points invested into "Swording that other guy in the face" and forgets things like "Talking to people" or "See things"

Easy: Separate currencies for different skill groups. Combat skill points (stabbing people, spellcasting talent, concentration), Exploration skill points (Search, Spot, Swim, Disable Device), Social skill points (Bluff, Intimidate, Diplomacy, Forge Script). Maybe putting the Profession, Craft, and Perform skills in their own category, along with Knowledge skills in their own category.


Plus classes exacerbate a problem with character building in general: It rewards you for acting within the confines of a tight little box you build for yourself, instead of actually exploring and experimenting within the confines of the game system. Classes tend to make the box that much tighter and acting outside it impossible instead of just less profitable.

Fatebreaker
2012-06-08, 09:47 PM
I still think that classes don't really make sense in the type of game 4E was trying to be. Really, the more I think about it, the more reasons I'm finding to question class systems in general.

Honestly, I'm not a fan of a class system -- it gets real weird real fast, because it's easy to conjure up character concepts which don't "fit" into a single class. But if you're going to do it, then the key is to make sure that every class has a mechanical purpose, because roles are the entire reason you use a class system in the first place.


Classes are staying. That's not in question. Take classes out of the game and it doesn't feel like Dungeons & Dragons anymore. It's now just "Generic Fantasy D20". If I wanted that, there's a ton of OSR games out there.

However, you're right that 4e roles make no sense in Next. They're working to get back to allowing more fine-tune customization of what your character can do, which precludes assigning classes specific roles. You're a fighter? You're not pigeonholed as the tank because you can take the Slayer theme and be a damage-dealing machine.

Why don't the roles make sense? Roles are larger than classes -- many classes can fit within a role. Some classes straddle the line between roles. Other classes have a primary role and a secondary role. Even classless systems still have roles. But the basic functions of combat are fairly well-described (not defined, but described) by the Defender/Leader/Controller/Striker model.

"Protect allies." -- Defender

"Help allies." -- Leader

"Hinder enemies." -- Controller

"Murder time." -- Striker

That's a fairly clear starting point from which to begin character customization.

TheAbstruseOne
2012-06-08, 09:51 PM
Interesting! Question: Can Shaman B shift to the Forest domain and gain control over the forest spirit that Shaman A summoned?
Well, depends on what the spirit does when it's "released". If it was given a command first, it's going to stick around and follow it. If not, it's going back to the astral plane (or whatever metaplane it came from). If it stays around or if Shaman A is still in control over it, then yes, Shaman B can try to take it over. Just like Shaman A can switch to the mountain domain and try to take over that spirit.

And then the street samurai fires off a bazooka and blows the enemy shaman into chunky salsa. It's a really fun game. If you've never played, pick up Shadowrun 3rd Edition used off Amazon. There's also a very rich and detailed metaplot.

And to get back to topic, yes I would love a Shadowrun-style magic system in D&D but I don't think it's possible with the level-based advancement.

king.com
2012-06-09, 12:40 AM
Honestly, I'm not a fan of a class system -- it gets real weird real fast, because it's easy to conjure up character concepts which don't "fit" into a single class. But if you're going to do it, then the key is to make sure that every class has a mechanical purpose, because roles are the entire reason you use a class system in the first place.


Your going to run into character concepts that dont fit in a game period regardless of a class system or not. A class system offers an easy entry for new players to build a character without being slap in the face with a list of too many option and lets a system develop character concepts that exist in the games setting. The only time class systems fall apart are when they are applied to games which try to be generalists. D&D is an example of this as its goal is to try and allow all kinds of fantasy settings to coexist under the same rulesystems.

I find classless systems to be a massive pain to get into. Its like the feat system for D&D, a massive list of variables that freee a person into inaction or allows them to completely override how a system is normally supposed to operate (i.e. optimising above and beyond). It is TERRIBLE for a new player to try and look at no matter how great the system may be. Eclipse Phase is a great system but there is no way in hell I will EVER make a character in that system and therfore extremely unlikely to actually play anything more than the coupel of 1 shots i've been involved in.

I remember my first Shadowrun character (I wanted to play a talky spellcaster). Starting character rolling something like 19 dice for all talky skills. Things got dumb very quickly and as a result the most interest aspect of roleplaying games became irrelevent to me. I didnt know at the time what is a reasonble skill to be set at and what your supposed to spend improvement points on (given I was a diplomatic GOD at that point). Similiarly I played Runequest (i think it was) once and my characters highest skill was like a 15 and was unable to hit or kill anything (playing a combat focused kind of character) or achieve anything for the game.

Simple answer you might say is that you tell the players what a default skill is like or what you need to be set to in order to be competent. At that point your telling players what good and bad sets of numbers they need in order to fit a concept and at that point they're going for a class.



Plus classes exacerbate a problem with character building in general: It rewards you for acting within the confines of a tight little box you build for yourself, instead of actually exploring and experimenting within the confines of the game system. Classes tend to make the box that much tighter and acting outside it impossible instead of just less profitable.

Every experience of class vs class-based systems I've had speaks the opposite. Classless system which dont specificly state you have to split points into areas (which often results in such a massive pain of a system that its not worth doing - see Eclipse Phase) have you spending points in a field until you hit maximum possible and then you move into a different field, you WANT to be ahead of the curve skill wise so you keep pumpinping points into it until your perfect and cant fail, THEN you start falling into areas that your allies have covered either weakly or not at all.

Take a class based system like Dark Heresy or Rogue Trader (which I know I never shut up about but I still think they are the best examples of character creation systems I've ever touched). You pick your class and you have a table of skills and talents to choose from. Before you can advanced to the next table for your class you need to spend a certain amount of XP into that table. A table covers a wide array of skillsets. This FORCES you to break out of the mono-task mindset very quickly or else you cant buy many more skills (given how this system operates a single skill character is bored most of the time anyway).

Take the assassin, focuses on killing things and sneaking. In a classless system you would max out your skills with either guns or melee weapons or more specified if thats how the system works and maybe something like sneaking to add to that. Raise those to max then swap over. Dark Heresy requires you pick up a bunch of side skills, so you spend most of your points on weapon training and security , sneaking and also things like drive skills and underworld knowledge so you can throw these pieces of information out when the party needs it. Its not your speciality but your forced to make a rounded human being.

If this gamedidnt use a class system you would have an absolute mess given how the universe operates and how the game is supposed to go. The format for the game requires certain types of individuals to be in place. Your working for a brutal organistation of the Inquisition, if your concept is 'I want to be a Pacifist who goes around helping people' you dont exist in this game system. Creating a skillset that accepts that makes zero sense.

I understand the need for class-less systems in games that need to be generalist and cover more bases than theres space to make classes for but they are not inherently cleaner than a class system.



Easy: Separate currencies for different skill groups. Combat skill points (stabbing people, spellcasting talent, concentration), Exploration skill points (Search, Spot, Swim, Disable Device), Social skill points (Bluff, Intimidate, Diplomacy, Forge Script). Maybe putting the Profession, Craft, and Perform skills in their own category, along with Knowledge skills in their own category.


Aren't you then forcing people to make a certain type of character with certain skills and requirements that they can pick from and certain features and aspects of their character that need to be taken? What if I want to just play a guy who hits things or a guy who talks my way through everything? You've put me in a tight little box.

At this point I would like to point out this is the essense of what makes a class system. You have limited how your expereinced is developed based upon guidelines. The only difference between a class system and what you've described is that a class allocates those points differently depending on which class is chosen.

erikun
2012-06-09, 01:12 AM
You're right. That doesn't feel very D&D. But I'd be very curious to see how it would work out, because it neatly helps to sidestep some of the weird math.

That huge dragon pinning the Fighter to the ground? Yeah, he's just too big to move. Raw strength can only do so much against the combined powers of mass and leverage. It also creates an alternative means to acquiring power, and options are a good thing.

You could also disconnect your Stat Rank from your Stat Score if you want the Stat Rank to fill in for a bunch of fiddly bonuses. Instead of dwarves getting a +x against poison or +y to stamina-related checks, they get a higher Constitution Rank. A Con 14 elf and a Con 14 dwarf both die when you stab them in the face, but the ConR 1 elf has less stamina, is more vulnerable to poisons and disease, and so on than the ConR 2 dwarf.

Again, not very D&D, but... that doesn't mean it's a bad idea. I'd be curious to see this in action.
HeroQuest does this. Stats gain "ranks" (called Masteries) every time they increase over 20, and each Mastery bumps what you actually rolled up one grade against a lower challange.

For example, if you have 5w2 in a skill (5 ranks with 2 Masteries, basically a 45) against a 17 difficulty, and your roll is a failure, then that is bumped up twice - once to a success, and once to a critical success. If it is reversed - a 17 skill against a 5w2 difficulty - then a critical success would be knocked down twice, once to a basic success and once to a failure.

Masteries cancel out for simplicity; 5w2 vs. 17w2 wouldn't bump the success up or down in either direction, but just cancel out and behave like a 5 vs. 17 roll.


The thing to keep in mind is that a bad DM is going to be a bad DM and a good DM is going to be a good DM. If you have a lot of rules rather than relying on DM rulings, it can mitigate the badness of the bad DM but it's also going to hamstring the good DM. If you have few rules but lots of rulings, the good DM is going to have room to shine but the bad DM is going to be worse.
If I may jump on the conversation for a moment: I'd prefer rules that allow a new DM get setup easily and pointed in the right direction. I remember AD&D 2nd edition, and it was quite confusing to even figure out how to get started.

There isn't any ruling that will stop a bad DM from DMing badly, but I'd at least like a ruleset that allows a new DM to start the right way and doesn't get in the way of a good DM.


But we're starting to get really off topic now, so new topic: What does everyone want WotC to do with Wild Shape?
Go back to 2nd edition, where you basically had a single animal form and stuck with that. You want to turn into something for melee combat? Good, you're now a bear, I hope you enjoy buying some armor and someone to suit yourself up because you're not getting spellcasting or wilding clasps or even healing for Wildshaping.

Perhaps a bit more seriously, Wild Shape sounds like a fun idea... although perhaps not too much for a Druid. Maybe if we go back to AD&D, where Druids were basically witch-priests, it would fit far better. The D&D4 Barbarian "animal totem" rages were also neat, if a bit silly at times.
However, the current nature-Clerics who can also turn into animals and get a lot of benefits from it really aren't the best (or even most interesting) option available.


I've actually had an idea bouncing around in my head for a while for a class based entirely around an improved form of wild shape, except what forms they can take is limited by their environment. Like, the forms available when they're in a forest is different from when they're in a swamp, jungle, or tundra. Add in a limited ability to be able to change their environment (like, say, suddenly flooding a forest with water so they gain access to swamp forms, or having trees shoot up through the floor to get forest powers) and I think there's enough to build a really interesting, unique class around it.
Forcing the Druid to pick an "animal totem spirit" that they need to focus their abilities around could be interesting. For example, a snake-Druid could turn into a snake, speak with snakes, and becomes an expert at grappling and making poisons. A tiger-Druid would be good at stealth, stalking, and pounce-striking.

Actually, that might make a really interesting class if you could take away the spellcasting...


So they give the player... less choice? I can understand making the game less complicated is worth it sometimes, but I don't see how providing less choice to the player, all else remaining equal, is ever a good thing.

(Unless you meant something else and I'm just an idiot.)
They make it easy to quickly look through options and make choices. 10 classes vs 100 abilities may sound like a bad idea for the former, but when you want to put a game together and start playing quickly, those 10 classes are the much better option.

Craft (Cheese)
2012-06-09, 04:31 AM
Every experience of class vs class-based systems I've had speaks the opposite. Classless system which dont specificly state you have to split points into areas (which often results in such a massive pain of a system that its not worth doing - see Eclipse Phase) have you spending points in a field until you hit maximum possible and then you move into a different field, you WANT to be ahead of the curve skill wise so you keep pumpinping points into it until your perfect and cant fail, THEN you start falling into areas that your allies have covered either weakly or not at all.

Well, actually, when I say "classless" I don't automatically mean a GURPS-like point-buy system. I think there's a great big design space out there that needs exploring, and it's a lot wider than that.

Anyway, this problem really has more to do with specialization than with anything to do with point buy: When one character can do something to the entire party's benefit, it doesn't make sense to have two characters who can do the same trick. If there's as many useful tricks as there are party members, then the logical result is you have a small group of hyperspecialized one-trick ponies.

Really, when your design is based on a D&D-like approach of each player choosing one of a handful of character classes each having a (theoretically) focused specialty, this problem becomes harder to deal with, not easier.


Take a class based system like Dark Heresy or Rogue Trader (which I know I never shut up about but I still think they are the best examples of character creation systems I've ever touched). You pick your class and you have a table of skills and talents to choose from. Before you can advanced to the next table for your class you need to spend a certain amount of XP into that table. A table covers a wide array of skillsets. This FORCES you to break out of the mono-task mindset very quickly or else you cant buy many more skills (given how this system operates a single skill character is bored most of the time anyway).

It's certainly interesting (and now I want to take a look at Dark Heresy for ideas) but I don't see how this system couldn't be applied if you chucked the class system.


If this gamedidnt use a class system you would have an absolute mess given how the universe operates and how the game is supposed to go. The format for the game requires certain types of individuals to be in place. Your working for a brutal organistation of the Inquisition, if your concept is 'I want to be a Pacifist who goes around helping people' you dont exist in this game system. Creating a skillset that accepts that makes zero sense.

I understand the need for class-less systems in games that need to be generalist and cover more bases than theres space to make classes for but they are not inherently cleaner than a class system.

This isn't really what I mean: My problem with classes here isn't that they stop you from making pacifists in a combat-based game, but that they often stop (or severely hinder) you from making things that fit with the game for seemingly arbitrary reasons. Why is the best liar the guy who sings? Why is the most learned character the spellcaster? You can come up with reasons for these things but what if you want to make a character who defies the stereotypes? A booksmart fighter would not spell the end of fantasy, and yet D&D doesn't let me do this without seriously crippling my character.


Aren't you then forcing people to make a certain type of character with certain skills and requirements that they can pick from and certain features and aspects of their character that need to be taken? What if I want to just play a guy who hits things or a guy who talks my way through everything? You've put me in a tight little box.

The problem here is the system of "You're good at some things, and bad at other things, so good things will happen if you stick to what you're good at and don't attempt anything else unless you absolutely have to." Chucking classes doesn't come anywhere close to solving this problem, but I never said it would.

Scots Dragon
2012-06-09, 05:25 AM
What you describe there is a limitation of the classes as presented, not a limitation of the class based system. In fact, what you present there specifically is a limitation of 3.5e/4e, particularly skills and feats. It's perfectly possible to spend a few non-weapon proficiency slots as a fighter in 2nd edition in order to become something of a scholar, knowing about history or religion or otherwise, for instance.

It's also perfectly possible in that same system to make a mage whose background is centred around etiquette and wordplay as much as the bard, provided he has enough charisma to pull it off. It costs a bit more, but not in as crippling a manner as 3rd edition.


Making things 'outside of the box' is something that can be done quite easily within a class-based structure. But in the case of D&D, it is quite important to make sure not to lose the basic feel and structure that makes the game work to begin with.

Craft (Cheese)
2012-06-09, 06:07 AM
D&D 3.5 actually dodges most of my issues with class systems anyway thanks to its unusually flexible multiclassing system: Attribute score dependencies are the main limitation. I was just using it as a particular example because D&D's the only class-based tabletop system that I know reasonably well.

But we're getting off topic again. Who wants to see hybrids come back? I wouldn't mind seeing some sort of partial gestalt system exist in the core rules alongside normal 3.5-style multiclassing.

Clawhound
2012-06-09, 06:10 AM
As for DM's making decisions or using Rule 0 too often, what is too often?

I don't mean that as a philosophical question. How often is often? I'm happy with 80/20. That may be too frequent for you.

As in so many things D&D, we have different tolerances for running off the rails. I happen to enjoy running off the rails, so a high frequency of ad-hocing suits me. The same isn't true for the next person.

I am going to predict, based on the comments of the designers, that the "core" game will be a very loose game. The DM will be required to make frequent Rule 0 judgements. I don't think that any feedback will change this direction as their surveys over the last year indicated that the general player base preferred this direction before they chose it.

I also see no change in the basic d20 mechanic occurring, and so the game will keep all its flaws.

D&D also has a large house rule and homebrew community. By large, I mean staggeringly large. They are designing rules that feed that community. That means publishing the sparsest game possible that can still be called D&D.

This is all to remember that the designers are not trying to design the game "right". They are trying to publish a game that makes money by appealing to their fan base.

Craft (Cheese)
2012-06-09, 06:13 AM
This is all to remember that the designers are not trying to design the game "right". They are trying to publish a game that makes money by appealing to their fan base.

You'd think that good game design would be more appealing to the fan base than bad.

Yora
2012-06-09, 06:24 AM
A good game is one that appeals to the broadest group of people. Not one that is super-extremely beloved by a small group of people.

Craft (Cheese)
2012-06-09, 06:29 AM
A good game is one that appeals to the broadest group of people.

This just in: Farmville is among the greatest games ever made. Stop the presses!

Seriously though, this is maybe only a useful filter once you factor in time. A long time. A game that's still extremely popular a decade after it's released probably has something going it, but I don't think immediate popularity is a good measure of good game design.

Scots Dragon
2012-06-09, 07:04 AM
The terms 'underrated' and 'underappreciated' exist for a reason, after all. It is more than possible for games to simply not have been noticed, or have otherwise been overlooked. Either by lack of awareness or lack of distribution, for instance.

blackseven
2012-06-09, 07:39 AM
I disagree. The design space between the Fighter and the Rogue traditionally have a lot of overlap, as the two main Mundane classes.

...

The two CAN be represented as separate classes, but ultimately the Fighter is capable of filling the rogue's role in and out of combat, and in a minimalistic game it should. Mainly because making the rogue separate and making something like skill usage a Rogue primary ability gimps the Fighter out of being able to be skilled... leaving him with basically no way to interract with the world out of combat (see: 3.5)

I agree with you were we creating a system from scratch, but we are not.

If you say "Rogue/Thief" whatever is simply a variant of Fighter, you're likely to hear a lot of objections, no matter the strength of your mechanical arguments. The Rogue/Thief had been its own class category longer than most in this thread have been alive.

I'm not saying its ideal, but it's a design constraint if you use the D&D label. Ditching the Rogue as a category equal to Fight/Cleric/Mage is almost as crazy as ditching the D20 (and I HATE the D20.)

Again, someone else brought it up, but if you are going to roll "rogue" into "fighter," then why not roll "cleric" and "mage" together? (And Psion, for that matter.) You can abstract out "power source" to mere fluff, and just give the "magic user" a choice of several lists that include both traditionally cleric and wizard spells. There's no necessary reason to separate "divine" and "arcane" magic: it can be all "fluff." One PC gets it from study (wizards), another is born with it (sorcerer), a third gets it through prayer (clerics), a fourth gets through devotion to nature (druid), and the last gets it by the power of their mind (psion.) But mechanically they all use the same spells.

king.com
2012-06-09, 10:59 AM
Well, actually, when I say "classless" I don't automatically mean a GURPS-like point-buy system. I think there's a great big design space out there that needs exploring, and it's a lot wider than that.

Anyway, this problem really has more to do with specialization than with anything to do with point buy: When one character can do something to the entire party's benefit, it doesn't make sense to have two characters who can do the same trick. If there's as many useful tricks as there are party members, then the logical result is you have a small group of hyperspecialized one-trick ponies.

Really, when your design is based on a D&D-like approach of each player choosing one of a handful of character classes each having a (theoretically) focused specialty, this problem becomes harder to deal with, not easier.


My definition of a class system is one which requries you to take a pregenerate set of rules an restrictions/advantages which specify much of what you can and cant do for much of your characters life. Classless is the opposite.

I think your analysis of specialisation being the problem is largely incorrect. Your standard roles to put it simply are: the tank, dps, healer, skill-monkey and talky. There might be more game specific roles but those are essentially how things work. Most systems even cross over those roles like a sorcerer being both dps and talky. Theres a limited number of specialisations and even if you have 10,000 unique specialisations to go down, they dont matter. The game is dictated by what the party can do. If nobody is playing a character who is going to talk their way out of a situation, why would a GM ever put them in a talky environmnet? If an opportunity occurs where then CAN use it fine but it is pointless for a party where nobody can talk as there is nothing the party can do about it. Instead the players spend their xp on the field where they KNOW theyare going to use it, combat being the obvious but also in their specialisations.

Anyone who knows a little about roleplaying is going to understand that GM is not going to immediately stop the game because nobody picked up lockpicking so the magic door cant be opened and everyone sits around doing nothing (dont argue the example sure there are other ways around it but Im using it as an example where a GM wont set fail states because a party didnt specialise in the right way).

This means there is zero reason why a game would have a player freely choose skills which dont benefit his specialisation. A class system actually produces a reason to get to take a broader set of skills.



It's certainly interesting (and now I want to take a look at Dark Heresy for ideas) but I don't see how this system couldn't be applied if you chucked the class system.


In later versions of the system (Black Crusade) they dropped classes for levelling purposes and you only pick a background. The result is that 90% of skills are completely ignored and you end up with the character who 'stabs dudes REAL good' and 'the talky guy' and of course 'the smart guy'. Outside of the few startign skills you get, there is zero reason to take anything outside what makes you better.

Not to mention any new player is immediately overwhelmed by the 500 options available to them and given zero explanation as to what they should take and why or how they develop rather than "heres a bunch of xp, heres a bunch of tables....go". Its not fun.

Classes give new players direction. I think 3.5's character creation is terrible, I cant even will myself to make a character for it. Your talk to pick a race...fine that makes sense. Then some stats then some classes alrighty. Then a pick skills...okay it tells me whats for my class that makes sense. The choose feats. This massive full double spread page of feats. Nothing indicates where I should start or what I should do or what would be beneficial to me or anything. My first character was a Human sorcerer with improved unarmed strike and deflect arrows. My logic was that I didnt want to have to carry weapons and since I was in the back my biggest threat was being shot by an arrow. In some games this makes sense but until I play many games of D&D I never learn this is generally a bad idea.

The game cannot reasonably teach you its natural flow and logic until you play it and see how it runs. Then you have a better grasp on the system. For a new player there is no direction and you better hope that theres someone to explain how any of it works or your going to run into the same trap of incorrectly applying logic into a system that bases itself on a different set of logic. How can you except a person to understand and make a large often unwieldy selectino of skills and abilities without understand how these skils and abilities operate.



This isn't really what I mean: My problem with classes here isn't that they stop you from making pacifists in a combat-based game, but that they often stop (or severely hinder) you from making things that fit with the game for seemingly arbitrary reasons. Why is the best liar the guy who sings? Why is the most learned character the spellcaster? You can come up with reasons for these things but what if you want to make a character who defies the stereotypes? A booksmart fighter would not spell the end of fantasy, and yet D&D doesn't let me do this without seriously crippling my character.


Im not saying making a pacifist in a combat-base game. Dark Heresy is NOT a combat-based game. Im saying making a character who exists with ideas and a personality which contradicts everything that exists for a human being in that universe. A universe of fanatical religious dogma, and extreme violent Xenophobic culture and an overwhelming hatred for anything that is different. Somewhere in which a pacifist doesn't exist, survive long or get recruited into an organisation which actively investigations and cleanses anything that is a threat.

One aspect of a class system you dont seem to appreciate is that class system allow you to create a concept during the character creation process. Rogue Trader for exampel has a list of selections to make to determine how you got to your career path. Your first choice is a homeworld, simple. Then you choose things like a particular hardship you went through, your outlook, what drove you to seek space. These are AMAZING for creating a character. I dont need to have a clearly defined concept, I just pick what sounds cool and in the end I have not simply a set of numbers and figures but instead a complicated and fleshed out character that is in a particular career (term used for class) as a result of certain his past. For many players especially new players it is one of the most difficult things in the world to have a character concept beyond "its kinda like me but...". This lets you create that character. I understand if thats not appealing but to me thats the heart of character creation, to create a character. With every class-less system I've ever touched I've only felt as if Im left with a bunch of numbers on a sheet.

D&D, atleast to me was the game that once tried to emulate what one might call traditional fantasy (please dont argue with me on what traditional fantasy is, im just talking about what was popular culture of fantasy for a brief period in history and yes theres exceptions, yes people are intrepreting something wrong, I dont care). The idea that there was a smooth talking poet who played a lute to gain the affection of a barmaid, each song a lie of some adventure he was never on. The traditional knight fighting for truth justice and god. He was good, pure, incorruptable and a perfect realisation of the truest form of morality. Something that shaped into a D&D Paladin.

Your real problem seems to be that D&D's class system is bad, not class systems as a concept. I dont know what you've played but I have no arguement that D&D's class system is a mess. It offers almost none of the advantages of being a class system with a messy set of freedom from a classless system. Honestly its one of the biggest reasons I dont play D&D is because its not fun to make a character. I dont doubt theres a lot of flexiblity in 3.5 but I will likely never see it simply because it doesnt make anything I can grasp and understand.

I personally havn't looked too much at 4th Ed (and hence why edition wars threads are some of my favourite on the internet and im sorry I missed their peak ages ago), I dont know people who run it and took a quick look at the player handbook and immediately didnt want to mess with it given the big list of powers I was expected to examine, sort through and determine what was best for my character before I knew how anything worked. That seems far too much like work to make me want to spend time and money on a new game system. I just hope that D&D Next does something, ANYTHING to indicate what I should be taking and what feats a class should be related to or develop a system that lets a play mess with the game before making choices like that.

Fatebreaker
2012-06-09, 11:09 AM
Take a class based system like Dark Heresy or Rogue Trader...

The Dark Heresy line works because it adopts the best aspects of a class system and a classless system. It's not a class system in the way D&D is a class system. D&D has you cash in your XP for a level, and that level determines how your character improves. Oh, did your Fighter want more skill points instead of a better Fortitude save? Too bad! Dark Heresy "classes" change how much xp it costs to buy different abilities, so it guides you down a role, but it doesn't dictate how you improve your character.

Or, put another way, D&D's class system asks the player to make one big choice, then ride the consequences until their next level. Dark Heresy asks the player what kind of role they want to fulfill, and then lets them make regular tweaks to that role within the level.

Personally, I think it's a great compromise. I usually describe it to new players as a "half-class" system, or an "archetype" system.

Incidentally, L5R operates on a similar model. You pick your broad archetype, then customize heavily within that archetype. I love both systems.

Craft (Cheese), I heartily second king.com's advocacy for looking into Dark Heresy and it's attendant games. There are four games within the same world, each focused on a different aspect of that world and a different style of play. Dark Heresy is more gritty, gloomy, "low-level" investigation. Think sci-fi noir horror. Rogue Trader is more powerful, exploration and conquest and wild adventure... plus the horror. Deathwatch and Black Crusade are BADASS! (...plus horror). Definitely give 'em a whirl.


Eclipse Phase

DUDE! Eclipse Phase is awesome!

...once you get past character creation. Which is... rough. There's a couple excel programs (http://eclipsephase.com/resources) out there which make it much easier, and once you've got the hang of it, it's fairly smooth, but... yeah. Not a beginner-friendly character creator. If you want to see a good, friendly classless character creation system, the old West End Games d6 system was amazing. Fifteen minutes or less or your character's free!

Eclipse Phase is a good example of how a buggy game can still be fun. My crew just started up a campaign over Memorial Day weekend, and it's definitely worth overcoming the system flaws. It has a great community, too, and the developers actively listen to folks. I can't wait to see how they improve the system as time goes on. Very excited.


As for DM's making decisions or using Rule 0 too often, what is too often?

"Too often," for me, is when a DM consistently uses Rule Zero to inconsistently interpret a core mechanic of the game because it is poorly designed and does not accurately model systemic intentions.

I like Rule Zero. What I don't like is the group having to regularly choose between enduring the immersion-breaking effects of a poorly constructed rule or hoping that Rule Zero Roulette comes out in their favor.


As in so many things D&D, we have different tolerances for running off the rails. I happen to enjoy running off the rails, so a high frequency of ad-hocing suits me. The same isn't true for the next person.

Choosing to run off the rails is fine. Being run off the rails by faulty mechanics is not. One is born of choice. The other is born of frustration... and somebody not doing their job.

Johnny Sunshine
2012-06-09, 11:19 AM
Wizards and other spellcasters are completely different. ... Their physical self rarely impedes them and they can use stronger magic easily as they level up. ... Its less hard to develop magical ability than physical ability. ... Another major point is that the fighter had to suffer alot to get to where they are; the same with athletes. They suffered blood, sweat, and tears in their journey. Wizards don't have to doing any of that.

As someone in a very mental-oriented profession who had to suffer through 15+ years of sweat, tears, and metaphorical blood, I would disagree with the implication that increasing in mental power is guaranteed and effortless.

As far as wizards go, I'm sure they'd disgree as well. Just as a PC fighter is far and away a better fighter than the average warrior in the game world, so too is a PC wizard more powerful than the average magic-user.

I think that predominantly viewing the game through the eyes of uber-powerful PC adventurers as we do, our perspectives are skewed to take for granted that one can simply increase in power forever. But our characters can only do that becase they "cheat" by being the stars of the show.

Consider this: You claim a warrior's growth in power is limited by their body, because they are mortal. Well, so too is it with casters. They are limited to what spells they can comprehend depending on how high their associated ability score is.

Most people see it as a waste of time playing a wizard without an Int of 17 or 18, for example. Yet, there are likely many, many more NPC wizards running around with Int scores below 15, who struggle and sweat all their lives but just can't seem to increse in power anymore.

By the way, also consider a fighter with a Str of 14 or 15. Less desireable, but certainly not as great a career limitation as it would be for a wizard. The fighter can still learn lots of tricks and techniques and become an incredibly powerful warrior, while after a while the wizard will be having severe spellbook envy against his fellow mages.


They don't have to suffer anything to get to being extremely powerful other than to risk dying.

Um, yeah. No big deal there. :smalltongue:

king.com
2012-06-09, 11:38 AM
The Dark Heresy line works because it adopts the best aspects of a class system and a classless system. It's not a class system in the way D&D is a class system. D&D has you cash in your XP for a level, and that level determines how your character improves. Oh, did your Fighter want more skill points instead of a better Fortitude save? Too bad! Dark Heresy "classes" change how much xp it costs to buy different abilities, so it guides you down a role, but it doesn't dictate how you improve your character.

Or, put another way, D&D's class system asks the player to make one big choice, then ride the consequences until their next level. Dark Heresy asks the player what kind of role they want to fulfill, and then lets them make regular tweaks to that role within the level.


For me I dont see a difference, maybe my definition of Class system is simply broader but both games have you choosing a class which places large restrictions on what you can and cant do. I know the fundamental different on how level up works but the progression follows the same format. Did you want more more toughness? "1000 XP you picked a class that doesnt do that well." Something so far beyond reach its not worth ever grabbing. An assassin can NEVER use a heavy weapon.

Sure I can see the arguement of it being more open than D&D but its still based so heavily around a class principle. If you want to move from level 1 to 2 you have a list of 8-10 skills and talents you can pick from and nothing else. Its much the equivilant to where you allocate skill points and what new feat you pick on level up. The only major difference is that Dark Heresy would give you more skill points than you have class skills and if you pick one a couple of times you can pick it again so you end up dabbling in various areas.

I still stay thats a D&D implementation problem, not a class system problem. I really would love to get into D&D but as Im not in a situation where if I were to play it I would have to run it, I need a system I can get into and want to learn the ins and outs of. If I don't want to do that theres a big problem. I'm really enjoying what they have put out for D&D Next and I really hope they implement a character creation system that actually offers a strong and interesting process for developing a character.



DUDE! Eclipse Phase is awesome!

...once you get past character creation. Which is... rough. There's a couple excel programs (http://eclipsephase.com/resources) out there which make it much easier, and once you've got the hang of it, it's fairly smooth, but... yeah. Not a beginner-friendly character creator. If you want to see a good, friendly classless character creation system, the old West End Games d6 system was amazing. Fifteen minutes or less or your character's free!

Eclipse Phase is a good example of how a buggy game can still be fun. My crew just started up a campaign over Memorial Day weekend, and it's definitely worth overcoming the system flaws. It has a great community, too, and the developers actively listen to folks. I can't wait to see how they improve the system as time goes on. Very excited.


Yea I was using the excel program, it was a combined effore of 4 people all working together and we got a single character completed. After that everyone else just grabbed one out of the back fo a book. The game itself is great fun but geeeeez I've saved the one character made and I will never make a character again but simply present this same character sheet everytime someone runs an Eclipse Phase game.

Personally I found Shadowrun 4th Ed's character creation to be my favourite classless. It breaks everything up into clear categories to be implemented and when it hits skills it tells you. If you want to talk? Buy these skills. If you want to shoot? Buy these ones. Spell selection gets far more tricky but extensions to character creation systems I dont have as much of a problem working with. I just wish it had told be that 19 dice is world altering levels of diplomancy.

Johnny Sunshine
2012-06-09, 11:39 AM
So now you have to decide: Do you want gods to be weak and not able to do awesome things, or do you want level 1 characters doing godlike things 5% of the time?

PCs are the best and brightest in their worlds. By virtue of being the main characters in a fiction story, things tend to go their way much more often, they can increase in power far more easily than a common person could, they frequently seem to be "in the zone," et cetera. Being Level 1 just means they are relatively inexperienced.

To have an 18 in an ability score is an incredibly rare thing in the game world, but as players we blow it off as nice, but not exactly hard to obtain. To have +7 training in a skill is, again, a phenomenal thing.

And that DC 27? That's just the bare cusp of "godhood" difficulty, something that most gods wouldn't even expend true effort to accomplish.

So, let's take an example with some real-world parallels to illustrate things. Let's say the NSA, FBI and Pentagon all work together to make the most secure computer system ever seen, which controls everything in their offices. From their control centers, they are effectively "gods" of their computer system, able to accomplish anything within it as they wish.

On the other side, let's put a hacker genius. Give him the brain of an Einstein or Hawking (specifically in tune with hacking through; I acknowledge different sorts of genius), as well as the absolutely best equipment available.

Put the hacker through a few test runs and training mods to get him familiar with his gear and practice mentally approaching the problems he will tend to face - thus putting him roughly the equivalent of "Level 1" - and further, assume he's been through any and all possible training available with the best gurus in existence.

Am I okay with a rules system that lets this hacker do something relatively insignificant against this computer system - perhaps getting it to flash the lights on and off briefly in a Pentagon bathroom - 5% of the time he tries to hack the system? Yeah, seems okay to me.

Oracle_Hunter
2012-06-09, 11:49 AM
As for DM's making decisions or using Rule 0 too often, what is too often?
It isn't necessarily a matter of frequency, but of kind.

If a DM must use Rule Zero on common and/or reoccurring situations, then the system requires too much Rule Zero.

For example, if a D&D-style system had no or unusable rules regarding killing dragons and exploring dungeons, it would require too much Rule Zero. If the same system required Rule Zero to adjudicate a pie-eating contest, that would not be too much Rule Zero for the system.

Johnny Sunshine
2012-06-09, 11:57 AM
Still, now we have the amusing result where, if your Str is 20 (+5 mod), you auto-succeed on anything DC 15 or lower. DC 16? You have a 50% failure chance. Skills could change that probably, but I found that idea amusing that 55% success rate = so trivial you always succeed.

It isn't really effortless for a character of Str 20 to accomplish a DC 15 task, but rather from a gaming perspective it's somewhat unreasonable to risk derailing the adventure just because a die said so.

At any rate, I feel the "autosucceed" rule is worded to make it clear it's an optional general guideline, to be adjusted on-fly as the situation warrants, and also it would only apply when circumstances are excellent.

Even when using such a rule though, should a Str 20 character try even a measly DC 7 task I would still require a roll if failure would result in damage or another significant event (pretty much anything during combat, for example), and I believe the presentation and wording of the rules in the preview documents get this point across.

Seerow
2012-06-09, 12:03 PM
PCs are the best and brightest in their worlds. By virtue of being the main characters in a fiction story, things tend to go their way much more often, they can increase in power far more easily than a common person could, they frequently seem to be "in the zone," et cetera. Being Level 1 just means they are relatively inexperienced.

To have an 18 in an ability score is an incredibly rare thing in the game world, but as players we blow it off as nice, but not exactly hard to obtain. To have +7 training in a skill is, again, a phenomenal thing.


False. Even in OD&D, an 18 was a 1/216 thing. This means in the smallest village you can expect to have someone there with an 18 in at least one stat. For every 1300 people, you can expect someone with an 18 in any given attribute. In a large city, you can reasonably expect to have a dozen people with an 18 in each stat. (er I just realized that's unclear, I mean to have a dozen people with an 18 in Str, a dozen with an 18 in Int, a dozen with 18 in con, etc).

So no, 18s aren't that rare. PC's have them more frequently than the common man in more recent editions, but a commoner with an 18 in a stat has always been around with some degree of frequency. And it makes sense that characters will have training in areas that suit their attributes.


And that DC 27? That's just the bare cusp of "godhood" difficulty, something that most gods wouldn't even expend true effort to accomplish.

False. The playtest packet states that Immortal Tasks are so difficult only gods can accomplish them with any reliability. They are the highest DC in the game. There isn't "Gods do this effortlessly, and can do even more awesome things", these ARE the awesome things. And level 1 commoners with a high stat can pull it off. There is no way you can spin that as making the least bit of sense.

Johnny Sunshine
2012-06-09, 12:45 PM
False. Even in OD&D, an 18 was a 1/216 thing...
So no, 18s aren't that rare.

The way I interpreted it, an 18 in OD&D is a 1/216 thing for PC adventurers - the special, dynamic, paragon movers and shakers of the world - not every measely sentient being. So from a global perspective I do see it as quite rare.

What if we looked around the real world and used Intelligence as an example, with IQ 100 = Int 10 being average. Would you really expect to find 1/216 of people walking down the street have IQs of 180? This is an extreme example with bellcurves that don't match up statistically, I realize, but hopefully it gets my point across.

Hm. I just realized that with this argument, 1/216 heroic adventurers would have any given score at a 3. Well then, let me adjust my thinking:

The bellcurves for ability scores are more broad than those abilities are for people in the real world. In the real world, the abstract equivalent of an 18 in an ability score is outrageously rare. Although in D&D most scores congregate at the peak of the curve around 10-11 or so, in the real world the peak would be much higher, and the tail ends of the curve would involve much less people.

I suppose the main reason for doing this in the game is to create much more diversity and interesting characters. If D&D tried to perfectly replicate real life, pretty much everyone would be playing characters with nearly every score in the mid-range, which would be really dull.

Well, now I don't know what to think. Okay, so maybe it isn't that rare to have an 18. It's still nowhere near common though, and I feel my most important point still applies:



False. The playtest packet states that Immortal Tasks are so difficult only gods can accomplish them with any reliability. They are the highest DC in the game. There isn't "Gods do this effortlessly, and can do even more awesome things", these ARE the awesome things. And level 1 commoners with a high stat can pull it off. There is no way you can spin that as making the least bit of sense.

There is a big difference between "only gods can accomplish them with any realiablity" and saying that it's a difficult task for gods. It just means it's really difficult for a mortal. Also, I don't see anything that says or even implies that DC 27 is the highest possible, that the laws of the game universe physics or whatnot make higher DCs impossible. Rather, it seems to say that for the purposes of mortal adventurers encountering obstacles it doesn't make practical sense to use higher DCs within a gaming session.

This is clearly an area of the rules that could use some rewording or expanding, since you and I are developing such opposite interpretations of it.

blackseven
2012-06-09, 12:52 PM
False. Even in OD&D, an 18 was a 1/216 thing. This means in the smallest village you can expect to have someone there with an 18 in at least one stat. For every 1300 people, you can expect someone with an 18 in any given attribute. In a large city, you can reasonably expect to have a dozen people with an 18 in each stat. (er I just realized that's unclear, I mean to have a dozen people with an 18 in Str, a dozen with an 18 in Int, a dozen with 18 in con, etc).

Why do you assume that the character generation rules are meant to "accurately" model *everyone* in the game world?

kyoryu
2012-06-09, 01:03 PM
big snip

Totally agreed, and it's the main reason I don't like 3.x. By treating classes essentially as very coarse-grained skills, it ends up being a hybrid creature that gains the strength of neither.

It's also the main reason a lot of people *do* like 3.x. The options and possibilities of the system, as unbalanced as it is, creates a much higher emphasis on the character creation subgame. Combined with the emphasis on the preparation subgame, it creates a very unique playstyle for people whose main interest is in "outclevering" the encounters.

Seerow
2012-06-09, 01:05 PM
Why do you assume that the character generation rules are meant to "accurately" model *everyone* in the game world?

Because in OD&D adventurers weren't different. Everyone was just 3d6 take what you roll in that order. There's a reason why player characters these days get higher point buys and stuff like 4d6 drop lowest. It's because the 3d6 is an average person. But that generates average people with an 18 fairly regularly.

kyoryu
2012-06-09, 01:06 PM
The way I interpreted it, an 18 in OD&D is a 1/216 thing for PC adventurers - the special, dynamic, paragon movers and shakers of the world - not every measely sentient being. So from a global perspective I do see it as quite rare.

Not how I've ever seen it, given that 4d6 is used to generate above-average characters when dice rolling is used in the first place.

True, most folks with an 18 probably become movers and shakers, but that's becuase they're naturally gifted, not because they're some special caste of folks.

But, my preference for "adventurers as normal folks that do exceptional things" vs. "adventurers as exceptional folks" is well-documented in this thread.

Fatebreaker
2012-06-09, 01:15 PM
King.com & Dark Heresy Sidebar:

For me I dont see a difference, maybe my definition of Class system is simply broader but both games have you choosing a class which places large restrictions on what you can and cant do.

Well, for me at least, Dark Heresy & Co. straddle the line between class and classless systems. It's leaning towards the class side, but it occupies a design space distinct from that of D&D.

The key, to me, is the question of how much freedom you have over your character. Gaining a level (rank) in Dark Heresy carries no inherent bonuses. It just opens up new opportunities. Gaining a level in D&D automatically improves your base attack bonus, your saves, your hit points, your skills, and so forth at a rate determined by their class. They gain specific class abilities exactly at a specific point and nowhere else. Some casters have a greater degree of freedom, since wizards and clerics gain spell slots which they can trade out on a daily basis.

Dark Heresy is much more free in letting you determine what your character looks like. Character creation gives you starting abilities based on your backgrounds and your chosen class, but after that, it's in your hands. There, the class serves as a guide. Some things are feasible, and some things are just beyond the scope of your chosen role, but within that role, there's a larger degree of flexibility and choice.

At least, that's how I see it.


I still stay thats a D&D implementation problem, not a class system problem. I really would love to get into D&D but as Im not in a situation where if I were to play it I would have to run it, I need a system I can get into and want to learn the ins and outs of. If I don't want to do that theres a big problem. I'm really enjoying what they have put out for D&D Next and I really hope they implement a character creation system that actually offers a strong and interesting process for developing a character.

Well, I agree with you that there's a disconnect between the design goal of "all-inclusive fantasy system" and "stereotyped classes," which invalidates neither on their own, but they operate poorly in conjunction with one another. 3.x was a huge disappointment to me in this area. 4e's decision to clearly define each class within the role it was intended to fulfill was a good step in the right direction, as was its greater clarity in saying "this is the kind of world which this game operates in."

I'm very curious to see how 5e's modularity operates.


Personally I found Shadowrun 4th Ed's character creation to be my favourite classless. It breaks everything up into clear categories to be implemented and when it hits skills it tells you. If you want to talk? Buy these skills. If you want to shoot? Buy these ones. Spell selection gets far more tricky but extensions to character creation systems I dont have as much of a problem working with. I just wish it had told be that 19 dice is world altering levels of diplomancy.

The most important thing I found that Shadowrun doesn't tell you is how important it is to get Wired Reflexes (or equivalent). Going multiple times before the bad guys even realize a fight has broken out is brutal. Half our first Shadowrun group stumbled into Wired Reflexes, and the other half didn't. The other half did not enjoy the game nearly so much as those of us who lived in bullet time. This, by the way, was SR3, not SR4, which I have the books for but haven't played yet. Between SR3 and Eclipse Phase, I probably won't get a chance to.

Point is, I'm always a fan of the game communicating important concepts to you. That sort of thing strikes me as very mature, very honest, and very good game design.

"Adventures are/are not different!"

*shrug* I don't mind a game which tells me that players are different or better than non-players. Exalted and 4e both go about it in very different ways, but they're up front and honest about it. And I don't mind a game where players are exactly like everyone else, except for how they happen to have a real-life puppeteer. L5R does this rather well.

I do mind a game which... isn't clear on the subject. 3.x, for example, had PC classes cost the same XP as NPC classes, with no special requirements to pick a PC class. So why would anyone choose an NPC class? It just makes me picture a world full of muddy li'l tykes going "Golly gee, I sure hope I grow up to be a commoner!"

Scowling Dragon
2012-06-09, 01:51 PM
Is anybody else already sick to the bone of DND5e discussions?

TheAbstruseOne
2012-06-09, 02:02 PM
Is anybody else already sick to the bone of DND5e discussions?
Nope, just sick of listening to everyone go in circles and/or dissecting the math to insane degrees of minutia. Also, about ready to go on the various forums and just start pointing out all the logical fallacies people keep using over and over again. Reductio ad absurdum seems to be a very popular one, with straw man arguments being a solid second.

Draz74
2012-06-09, 03:22 PM
Also, about ready to go on the various forums and just start pointing out all the logical fallacies people keep using over and over again. Reductio ad absurdum seems to be a very popular one, with straw man arguments being a solid second.

Since when is ruductio ad absurdum a fallacy? It was taught to me as a valid law of logic in high school math class ...

Knaight
2012-06-09, 03:37 PM
Since when is ruductio ad absurdum a fallacy? It was taught to me as a valid law of logic in high school math class ...

It isn't a fallacy. (http://en.wikipedia.org/wiki/Reducio_ad_absurdum)

TheAbstruseOne
2012-06-09, 03:40 PM
Since when is ruductio ad absurdum a fallacy? It was taught to me as a valid law of logic in high school math class ...
Math and debate aren't the same thing. That's why you get spherical cows in a vacuum. Reductio ad absurdum in debate is when one side reduces an argument to its furthest extreme and then ridicules the result. So if I say that such-and-such shouldn't be a class, then someone else says "Well if that's not a class, then rogue shouldn't be a class and neither should fighter or wizard and then we just don't have any classes at all and that's just stupid!" It doesn't address the actual argument made. I have a better example, but it's political in nature and a hot-button topic right now and I don't want to derail the thread.

Fatebreaker
2012-06-09, 03:49 PM
Is anybody else already sick to the bone of DND5e discussions?

Clearly not, otherwise we wouldn't be talking. If you yourself are no longer interested, no one is making you participate. On the other hand, if you would like to change the course of the discussion to something more suiting to your interests, is there a subject you'd like to discuss?


Nope, just sick of listening to everyone go in circles and/or dissecting the math to insane degrees of minutia.

Like I said earlier, the math happens whether you believe it, are aware of it, or even care about it. And, whether you believe it, are aware of it, or even care about it, it will change your game.

For folks who care or are interested in game design, then opening up the mechanics and seeing what makes them tick, or how they function in different circumstances, or even finding where they break can all be a fascinating subject.

For folks who have a vested interest in spending their time and/or real money-dollars on a quality game, answering the question "How does it actually work?" is very relevant to their interests.

And for the folks at Wizards -- some of whom can and will lose their jobs based on the next year in the world of D&D -- these answers aren't just fascinating or relevant. It's a priority.

So, since it's a playtest, and since it's a fascinating time to watch a game in development, and since it's a chance to see whether a new product is worth investing in, and since it's a life-altering moment for real folks just like you or I, I hope you can understand why people do want to look at the math.

If you don't, that's fine. You can skip those bits. You can also, should you choose, read them and try to learn why they matter.

Draz74
2012-06-09, 03:51 PM
Math and debate aren't the same thing. That's why you get spherical cows in a vacuum. Reductio ad absurdum in debate is when one side reduces an argument to its furthest extreme and then ridicules the result. So if I say that such-and-such shouldn't be a class, then someone else says "Well if that's not a class, then rogue shouldn't be a class and neither should fighter or wizard and then we just don't have any classes at all and that's just stupid!" It doesn't address the actual argument made. I have a better example, but it's political in nature and a hot-button topic right now and I don't want to derail the thread.

Huh, I thought that was a strawman.

TheAbstruseOne
2012-06-09, 04:09 PM
Huh, I thought that was a strawman.
No, but they're related. A strawman is setting up a false argument to change the subject. Example: I say avengers shouldn't be a class. Someone else says that they shouldn't be in the game because they're from 4e and 4e is horrible because blah blah blah. That's a strawman argument because it's setting up a separate argument that has nothing to do with the original one. You're not saying anything as to the merits and drawbacks of having avengers in Next, you're simply railing against 4th Edition in an attempt to get me to defend 4e.

Basically, reductio ad absurdum stays on-topic but attempts to shift the debate within that topic to the point where the discussion has no value. "This leads to this which leads to this which is completely stupid!" Strawman attempts to shift the topic completely to something else that is more easily defeated. "This is the same as that and that is completely stupid!"

Another fun one that's a bit more rare due to forum rules (but keeps popping up in arguments against the developers) is argumentum ad hominim. This is where they attack the person rather than the argument. Example: Mearls said in this review of Keep on the Borderlands that he hates the module and the style of play, then in this interview was the first person to say "DM May I" and talked about how rulings over rules was stupid. Now he's saying the exact opposite. How can we believe anything he says if all he's doing is being a mouthpiece for WotC?" This is attacking Mearls himself for changing his mind (for the record, it was over a period of 7 years) rather than discussing the idea of rules vs. rulings. It's attempting to dismiss the argument by dismissing the person putting the idea forward.

The reason all these bug the hell out of me is that they add absolutely nothing to the discussion, and right now the discussion itself is incredibly important. We need to be discussing the ideas of what is a class and what isn't a class, the merits of rules vs. rulings, the idea of a simple and streamlined rules system and adding complexity through powers/abilities/feats/spells/modules/whatever, what the default assumptions for playstyles should be or if there should be assumptions, etc. etc.

Yora
2012-06-09, 04:18 PM
Is anybody else already sick to the bone of DND5e discussions?

On Enworld there are a couple of quite interesting and constructive ones.

Knaight
2012-06-09, 04:37 PM
Math and debate aren't the same thing. That's why you get spherical cows in a vacuum. Reductio ad absurdum in debate is when one side reduces an argument to its furthest extreme and then ridicules the result. So if I say that such-and-such shouldn't be a class, then someone else says "Well if that's not a class, then rogue shouldn't be a class and neither should fighter or wizard and then we just don't have any classes at all and that's just stupid!" It doesn't address the actual argument made. I have a better example, but it's political in nature and a hot-button topic right now and I don't want to derail the thread.

That isn't Reductio ad absurdum. That's a false dichotomy, wherein it presents the only options of a particular class existing, or there being no classes at all, when the option of a game consisting only of other classes exists. It is probably also a slippery slope argument in practice, but this does depend on the exact reasoning.

RedWarlock
2012-06-09, 04:45 PM
Is anybody else already sick to the bone of DND5e discussions?

You're asking this IN the D&D 5e discussion thread? Seriously?

TheAbstruseOne
2012-06-09, 04:49 PM
On Enworld there are a couple of quite interesting and constructive ones.
I'd really like an outside opinion on some of these damn threads I'm in on that site...feels like I'm banging my head against the wall and it'd be good to know that at least some good is coming out of it.

TheAbstruseOne
2012-06-09, 06:40 PM
New question for discussion: Now that they've done Keep on the Borderlands, what classic module would you like to see them produce for the next playtest packet (assuming there's a new module included)?

Scots Dragon
2012-06-09, 06:54 PM
Tomb of Horrors.

Because a module you can escape from with your character's life is simply not worth bothering with.

TheAbstruseOne
2012-06-09, 07:11 PM
Tomb of Horrors.

Because a module you can escape from with your character's life is simply not worth bothering with.
I'm working on Temple of Elemental Evil right now just to see how flexible the playtest material we have is. Having to do a lot of fudging and reskinning to make it work. For example, the spider in the moathouse looks surprisingly like a bugbear (it's a non-venomous giant spider).

But yeah...this thing's looking very deadly if you don't play old school style, with pack horses and meat shields-- err, I mean hirelings.

king.com
2012-06-09, 11:18 PM
Personally I want a scenario packet that isnt just a dungeon crawl, something that gives enough freedom for players to actually take advantage of the 'think therfore it is' skill system.

Fatebreaker
2012-06-10, 02:20 AM
New question for discussion: Now that they've done Keep on the Borderlands, what classic module would you like to see them produce for the next playtest packet (assuming there's a new module included)?

I'd actually like to see them not do a classic module.

Besides the fact that I'd like to see a module designed to fit the system in place, I really want to see what people think without the nostalgia goggles.

Something from late 3.x or from 4e would also be a nice way to say, "Hey, this really can play all editions!" But personally, I'd like to see a custom scenario, one designed to take advantage of what 5e actually brings to the table. I'm not sure what that scenario is off the top of my head... but I'd like to see what WotC thinks it is. I think that would be very telling.

Yora
2012-06-10, 06:18 AM
Sunless Citadel.

Also kind of a classic, but unlike Keep on the Borderland it's actually really good. It has a plot, NPCs to talk to, and features most of the situations that new players have to learn the rules for.