View Full Version : Hypothetical idea for balance (3.5)

2009-09-11, 09:25 AM
I've been toying with this idea for a bit, and I'm just throwing it out there.

What if you used a variant system where enemies are worth a set amount of xp regardless of your level (like, a CR 1/2 is worth 100xp, period). Each class levels up after a set amount of experience is reached, but these set amounts are different for each class.

I'm thinking something where all classes are the same until about 5th or so, then the weaker classes start leveling up faster (fighter, monk) and the redonkulous classes start leveling up slower (wizard, druid). The other classes each vary somewhere between the extremes. Each class can only go to level 20, after which you have to multiclass or take a prestige class.

So in this system, a party that is effectively level 13 might have a Fighter 16, a rogue 13, a cleric 11, and a wizard 10, where they are each CR 13 themselves. (the numbers in this example are just examples, I don't know what the scale would really be).

Ideally, we would need to find which class is consistent in terms of its own level being equal to its CR (like a rogue 7 being CR 7) and go from there.

So, in a given adventure at, say, 12th level, is a 16th level fighter good enough for 12th level? How about an 8th level druid? Are they close to equal?

The only glaring problem with this is wealth by level. In the paragraph above, we would have to assume that both the 16th level fighter and the 8th level druid would both have 12th level gear. This whole idea is hard enough to balance, now you have to screw with WBL too.

Anyways, thoughts?

Yuki Akuma
2009-09-11, 09:30 AM

Second edition levelling mechanics?

...Well it worked ("worked") for 2e. You could give it a shot.

2009-09-11, 09:33 AM

Second edition levelling mechanics?

Yeah, pretty much.

2009-09-11, 09:38 AM
Multi-classing kills this idea before peeing on it.

2009-09-11, 09:50 AM
Same effect but more work required, would be to draw out the abilities of more powerful classes over more levels.
Make the non-epic levels 30 instead of 20. Classes like Fighter, Barbarian, and Rogue can simply be extended by 10 levels, and those as Druid, Cleric, and Wizard get the same amount of spells at the end, but gain new spells slower.

Though that still doesn't fix that spellcasters have loads of abilities to customize every day, and melee classes can hit enemies really hard.

2009-09-11, 09:51 AM
Multi-classing kills this idea before peeing on it.

Not really. If you keep the classes separate for experience, while also tracking a total (in the way 2E did), multiclassing stays relatively the same.

For example, in 2E, a 1st level Fighter/Mage needed 1000 experience to become a 2nd level Fighter, and 1500 experience to become a 2nd level Mage, and all of their experience gains were split between the two classes. And they can only gain a (combined) amount of experience up to their respective levels.

To do this in 3.5, you just track the two seperately. Let's take the experience totals from 2E into this:

You have a character that starts as a 1st level Fighter, and one who starts as a 1st level Wizard.
Fighter: 0/1000
Wizard: 0/1500

The Fighter hits Character Level 2 at 1000 experience, and the Wizard hits character level 2 at 1500 experience. Now the Wizard multiclasses to Fighter for a level, and the Fighter multiclasses to Wizard for a level. Their totals now look like this:

Fighter/Wizard: 1000/2500
Wizard/Fighter: 1500/2000

You just apply the next level's experience total for when they hit character level 3. If they had stayed in their respective classes, it would look like this:

Fighter: 1000/2000
Wizard: 1500/2500

While it may let some people raise their character level a bit faster, there's still a cap on their max level (20 for most games), but the rate at which they reach it is either faster or slower depending on what kind of character class they want.

2009-09-11, 09:54 AM
The best way to do it is to apply a premium for certain classes. Like a wizard requires 50% more experience to get to the next level, rather than a set amount. This fixes the problem with multiclassing.