Stellaris II: I, For One, Welcome Our New Robot Fungus Hivemind Overlords

[GloatingSwine]

They mostly sound good, but I can see a bit of a contradiction in there: ships can retreat from battle below 50% hull, but hull damage also reduces the ship's combat effectiveness and speed, so it seems they're going to die pretty rapidly if they get as low as 50%. Will have to see how it actually works in reality.

Hull damage will reduce ships' damage output but won't noticably affect their ability to mitigate further damage (because evasion is a useless non-stat now anyway).

It will mean that big guns have more chance of scoring kills though.

[Kaptin Keen]

New Dev Diary. And I must say, I am optimistic about these changes.

To me, it looks like they started out with a combat system that ... essentially just doesn't work ... and now they're patching the patches on their patches in order to improve on something that should have been replaced. It's Frankenstein's monster, and it's not pretty.

In the real world, when a small force must fight a larger, you pick tactics and terrain that suits you and hinders the enemy.

In space, terrain would be immobile emplacements, or planets/asteroid fields. It could be mine fields. They have neglected the obvious chance to do this. They even have mine fields and immobile emplacements, they just don't work.

Tactics might be 'using multiple waves of his and run corvettes with torpedoes'. They've sort of hand-waved in this direction, but ... not really, right?

It's still a good game, and I've invested massive hours in it. But the combat/tactics are crap. And I'm convinced they'll remain so =)

[Aotrs Commander]

To me, it looks like they started out with a combat system that ... essentially just doesn't work ...

To an extent, I think you're right. So long as the battles are basically "charge towards enemy, firing" you've never going to get especially far.



That said, for this paradigm that they've chosen, working on the assumption that the battles are really only barely less abstract than the ones in EUIV or CK2, the RoF for up to 2:1 odds boost is a not a bad mechanical patch. (Would be better flavoured by them saying that the difference was due to the command/control/communications delays because of a larger fleet, it takes more time for the comms to reach them, mind.) As shooting is the only real meaningful mechanical effect of a movement-incidently system, placing the effect there is not bad. (What people often forget is that the EFFECT achived is often more important in a system thatn the way it is derived.)



But yes, to an extent, so long as they don't actually do some proper AI programming to have differet sorts of maneouver (i.e. kiting at suitable weapons range), then it's always going to struggle a bit. But this will hopefully be a DRASTIC improvements over what we've currently got.



Also, I feel somewhat vidicated in Acclerate & Attack's approach to armour, since Stellaris is more or less adopting the same line of thnking...!

[ShneekeyTheLost]

Hey, we can finally have torpedo 'vettes that won't suicide but hang back and actually use their weapon's range. Cool. And really, that makes a lot of sense. You want your ships together for mutual protection, especially when the missiles start flying and you've got PD ships mixed in there.

[Kaptin Keen]

To an extent, I think you're right.

I think there's little doubt that I'm overstating my case somewhat. I'm bitter because I have no control over units once combat is joined - but there are many other layers of tactics than that =)

It's funny. I feel that MoO is a better strategy game than Stellaris - and comparatively, I feel that Xenonauts is a better tactics game than X-Com. Regardless, I've played the prettier, but less good, games far more than the others.

[boomwolf]

But the last dev diary even mentions that each ship type has at least 2 Ai types to choose from and then you pile up your doctrine.
You can customize your tactics by setting up your fleets differently.

[GloatingSwine]

In the real world, when a small force must fight a larger, you pick tactics and terrain that suits you and hinders the enemy.

In space, terrain would be immobile emplacements, or planets/asteroid fields. It could be mine fields. They have neglected the obvious chance to do this. They even have mine fields and immobile emplacements, they just don't work.

Before we proceed, make a prediction for me.

If a fleet of 100 ships fights a fleet of 50 ships exactly the same, how many casualties should the fleet of 100 ships suffer in victory?

[Olinser]

Before we proceed, make a prediction for me.

If a fleet of 100 ships fights a fleet of 50 ships exactly the same, how many casualties should the fleet of 100 ships suffer in victory?

You haven't provided enough information to make a prediction, unless you mean this is in the Stellaris engine.

Casualties will largely be determined by who had better tactics and was able to focus their own fire while avoiding enemy fire. Which is not determinable by the information given.

In a 'pure' engagement, in which we say that all 150 ships are exactly the same and they are all able to fire on an opposing ship 100% of the time, then the 100 ships would be expected to lose I believe 12 ships (rounded down from 12.5), while killing all 50 enemy ships.

With the Stellaris AI, I'd actually expect them to lose significantly LESS than 12 ships due to a significant amount of 'empty' damage being applied by the smaller fleet that results in damage done but not an actual ship kill. Rough estimate I'd guess that if you actually spawned 2 fleets like this the 100 ship fleet would destroy the 50 ship fleet and lose 5-8 ships.

[Kaptin Keen]

Before we proceed, make a prediction for me.

If a fleet of 100 ships fights a fleet of 50 ships exactly the same, how many casualties should the fleet of 100 ships suffer in victory?

Since this is a reply to my post ... are you asking me to predict humanity's first space battle (involving said numbers)?

A thermopylae in space is absolutely within reason. It's not the expected outcome for a small force to beat a larger one, but it's possible. And - to me - it's a very, very good game that allows such a thing to happen if played correctly.

In one of my games, I beat a very much stronger FA fleet simply by placing my own fleet behind a space station/fortress combo. Their fleet wasted their time on my emplacements, while my battleships enjoyed free fire with their artillery.

But that is the only time I've seen actual tactics do anything in this game.

In a straight fight - lining up ships and exchanging fire until one fleet is gone, the 100 ships should beat the 50 100% of the time, with casualties on each side entirely predictable based on accuracy, damage and fire rate. Olinser likely has the numbers close to right.

In any other scenario, it would be hugely variable.

[GloatingSwine]

Since this is a reply to my post ... are you asking me to predict humanity's first space battle (involving said numbers)?

Given that we're in the Stellaris thread, I thought it would be relatively obvious I was talking about Stellaris....


So, we're all agreeing that in a situation where one side outnumbers the other by 2:1, the larger force will take only about 8-12% casualties.

That means that the force that wins the first battle, because it loses so few of its own ships, does not just win that battle but the whole war (and very likely all future wars as well). No matter how large the opposing empire, a single space battle spells either absolute victory or defeat for the rest of the game.

In that paradigm, there is only one sane response, which is to have all of your ships in one place. Because the consequences of being outnumbered are so monstrously negative, if you aren't fighting an opponent who you could still beat with your eyes closed, arms tied behind your back, and your hair on fire it is absolutely inconceivable that you should ever split even a small part of your fleet.

You might imagine fringe cases where positioning caused an upset*, but they will always be fringe cases. The majority of battles have to be assumed to be taking place in open space with nothing significant nearby, because that's how most battles in the game actually happen.

Hence, the casualty levelling bonus for the smaller fleet. It is necessary to reduce the absolute decisiveness of single battles to the fate of entire empires. It has to work everywhere all the time and it has to stop even rather small numerical advantages in single battles turning into immediate crushing victories on the strategic level.



* Your cited example is almost certainly a case of Fallen Empire ships being grossly overestimated by the fleet power calculation due to their large HP pool. Your fleet was almost certainly nowhere near as disadvantaged as you thought.

[Rockphed]

Given that we're in the Stellaris thread, I thought it would be relatively obvious I was talking about Stellaris....


So, we're all agreeing that in a situation where one side outnumbers the other by 2:1, the larger force will take only about 8-12% casualties.

That means that the force that wins the first battle, because it loses so few of its own ships, does not just win that battle but the whole war (and very likely all future wars as well). No matter how large the opposing empire, a single space battle spells either absolute victory or defeat for the rest of the game.

In that paradigm, there is only one sane response, which is to have all of your ships in one place. Because the consequences of being outnumbered are so monstrously negative, if you aren't fighting an opponent who you could still beat with your eyes closed, arms tied behind your back, and your hair on fire it is absolutely inconceivable that you should ever split even a small part of your fleet.

You might imagine fringe cases where positioning caused an upset*, but they will always be fringe cases. The majority of battles have to be assumed to be taking place in open space with nothing significant nearby, because that's how most battles in the game actually happen.

Hence, the casualty levelling bonus for the smaller fleet. It is necessary to reduce the absolute decisiveness of single battles to the fate of entire empires. It has to work everywhere all the time and it has to stop even rather small numerical advantages in single battles turning into immediate crushing victories on the strategic level.



* Your cited example is almost certainly a case of Fallen Empire ships being grossly overestimated by the fleet power calculation due to their large HP pool. Your fleet was almost certainly nowhere near as disadvantaged as you thought.

Somebody mentioned that how something is achieved is often secondary to what is achieved. The goal was to make it so that doomstacks are no longer just better than other strategies. Secondarily, the goal was to make it so that if two fleets fight, the difference in casualties is closer to linearly related to their difference in size, rather than quadratically or higher. They are probably going to mis-tweak, but their goal is that, should a 100 fleet and a 50 fleet fight, all else equal, both fleets will lose 50 ships. They have also said that the difference will peak out at about 100% when the smaller fleet is half the size of the larger one.

Huh, one other thing this does is it encourages concentrating power into large ships instead of just running around with a horde of naked corvettes. I didn't notice that until just now.

[Kaptin Keen]

* Your cited example is almost certainly a case of Fallen Empire ships being grossly overestimated by the fleet power calculation due to their large HP pool. Your fleet was almost certainly nowhere near as disadvantaged as you thought.

Oh no! No no no .... I fought them elsewhere first, lost terribly. Then reloaded, fought the same battle on my terms, won easily.

[LuckGuy]

Are doomstacks really a problem? I thought the existing war system took care of that. If you will lose the fleet battle, just avoid combat and go after planets, force a victory through war score. Isn't that why planets have a much larger effect than ships?

The quadratic nature of naval combat has been known for hundreds of years. More guns means you reduce their fleet faster, which slows the rate which yours is reduced, which gives you a bigger advantage, so even small differences snowball. Midway is the exception, Jutland is the rule.

[Silverraptor]

Are doomstacks really a problem? I thought the existing war system took care of that. If you will lose the fleet battle, just avoid combat and go after planets, force a victory through war score. Isn't that why planets have a much larger effect than ships?

The quadratic nature of naval combat has been known for hundreds of years. More guns means you reduce their fleet faster, which slows the rate which yours is reduced, which gives you a bigger advantage, so even small differences snowball. Midway is the exception, Jutland is the rule.

The problem is if one side completely destroys the other sides fleet, and you try to go for planets, they could easily split their fleet then and go for your ships in their territory and simultaneously go for your planets in yours. Also, the more ships you have, the faster the garrison on the planets decrease, so they will be easily able to capture planets faster then you, and you'll take so long to knock out their garrisons that it'll give them plenty of time to see you combat strength and send an appropriate force to sweep it up before you do any damage while they continue their conquest of you.

[GloatingSwine]

The other problem is that if you defeat an enemy fleet in a single battle, you haven't just won that battle, you've won that war and all future wars with that same empire because they will never reach the same level of power from as far behind as you put them.

[Kaptin Keen]

.. capture planets ..

Does that actually ever work, for anyone? It takes easily long enough to bomb your way through planet defences that even a fleet on the other side of the galaxy has enough time to react, before you can land troops (depending on drive tech, obviously).

Even more so if your fleet is divided into several un-doomstacks.

[GloatingSwine]

The AI will always prioritise getting its own planets back over taking more of yours, so a planet capping race will only work if its fleet is very far away and you can fastcap lots of planets.

It will also ignore distractions below a certain threshold now, so you can't shotgun corvette it.

What will work is cheesing it. Take one of its planets then snipe its transports, it is bad at protecting transports so this isn't too hard. It will then sit over the planet you have captured forever and keep bombing it with no transports to recapture it.

[Kaptin Keen]

What will work is cheesing it. Take one of its planets then snipe its transports, it is bad at protecting transports so this isn't too hard. It will then sit over the planet you have captured forever and keep bombing it with no transports to recapture it.

Really??

That's amusingly pathetic. I get it though, it's only a computer, there isn't any actual thinking taking place, just a list of priorities. It's the sort of thing I'd never find out though, because ... I just don't try to cheese. Never occurres to me. I try to slug it out, and if I lose I reload =D

[Driderman]

Really??

That's amusingly pathetic. I get it though, it's only a computer, there isn't any actual thinking taking place, just a list of priorities. It's the sort of thing I'd never find out though, because ... I just don't try to cheese. Never occurres to me. I try to slug it out, and if I lose I reload =D

If you're playing a strongly defensive race (which is arguably poor cheese in itself) it's also fairly easy to have enemy empires spend an entire war just bombing the **** out of one of your fortified planets but never invading, leaving you free to rebuild your fleet, wage war on a 3rd party with more or less impunity, or just wait for the warscore to tick to white peace while completely ignoring the fact that one of your planets is being bombed for years. Doesn't work so well if the enemy has "Armageddon" bombing stance though...

[Name_Here]

A more hit and run fighting style would be doable if my admirals could be convinced not to hurl everything at opposing fleets that have double their firepower

[Silverraptor]

Hey, I am actually curious about this. So I've been basically achievement hunting in my game and I'm getting close to the 100% achievements goal. I want to know, what is everyone's rarest achievements that you've got that you are particularly proud of in this game. For me, I have the 5 rarest achievements in rarest-most common:

• Retired Home
• Outside Context
• Rise of Machines
• Xenophage
• Voight-Kampff

I of course have a lot more, but those are my 5 rarest achievements in this game. What is everyone else's 5 rarest achievements when you compare it to the global achievements list?

[GloatingSwine]

Rarest 5 (from rarest to least rare):

Last, Best Hope
Rise of the Machines
Deus Vult
Does Not Compute
Towards Utopia

[factotum]

Don't think I've got any achievements--don't you have to play in Iron Man mode to get them? I might try it at some point, but I'm not planning on playing the game again until all these latest changes get implemented.

[GloatingSwine]

Stellaris is predictable enough that there's no reason not to play in Ironman.


I think I might try to get the Peacekeeper/Very Open Borders achievements next. Make a xenophile pacifist and go and kick everyone until they behave (and sign migration treaties.

Probably also go tall and spam habitats.

[Aotrs Commander]

Stellaris is predictable enough that there's no reason not to play in Ironman.

I'd have to disagree - I don't see there's any reason to play ironman in anything, myself.

[GloatingSwine]

Well, the only reason not to is so you can reload an older save if you do something wrong.

But if everything is super predictable you won't end up making catastrophic decisions anyway,

[factotum]

Stellaris is predictable enough that there's no reason not to play in Ironman.

Until they release the next patch and completely change up the game mechanics again, you mean? I really don't like that...sure, add new features in new patches, but completely changing existing mechanics just means I have to learn how to play the game again.

[mythmonster2]

One of the biggest reasons I don't play Ironman in Stellaris is all those moments where I build a Frontier Outpost and miss the system I really wanted to include in there by a couple pixels, or vice versa when I delete an outpost I figured wasn't needed anymore and end up losing a valuable system. Seriously, why do we not have a way to see what we're going to get with an outpost yet?

[Rockphed]

One of the biggest reasons I don't play Ironman in Stellaris is all those moments where I build a Frontier Outpost and miss the system I really wanted to include in there by a couple pixels, or vice versa when I delete an outpost I figured wasn't needed anymore and end up losing a valuable system. Seriously, why do we not have a way to see what we're going to get with an outpost yet?

One of the reasons cited for the change to how systems work is to put an end to that sort of thing.

[Silverraptor]

One of the biggest reasons I don't play Ironman in Stellaris is all those moments where I build a Frontier Outpost and miss the system I really wanted to include in there by a couple pixels, or vice versa when I delete an outpost I figured wasn't needed anymore and end up losing a valuable system. Seriously, why do we not have a way to see what we're going to get with an outpost yet?

I mean yeah, knowing exactly what would happen with a frontier here or one taken out there would be nice, but I never found it game breaking. If I see the territory just missing by inches, I know that with the proper steps (That everyone should be taking), you should be able to have your territory naturally grow where that doesn't matter. Missing a few years of benefit from one system doesn't mean very much. Also, with removing the Frontier Outposts you need to eyeball your territory and account for friction. If your frontier outpost is near the edge of your territory with very little distance between it and the end of your territory, you don't have enough border projection to afford decommissioning it. Whether it's because you have a low population on your planets or not enough territory researched in border growth, or you have a neighbor right up against your territory where only the frontier outpost is holding it back; if you don't have anything else on the other side of the frontier outpost that directly extends your borders (ie. a colonized planet) then you shouldn't decommission the frontier outpost.

Essentially, look at your frontier outpost as having the same border projection as a very, very small planet (as in a planet you would never colonize in 200 years. Maybe in 300 you might.). The whole purpose of that frontier outpost is to expand you borders in a little jut into space and snatch up a few close together valuable systems (often times its nebulae because they have higher then average strategic resource drops. Or a small cluster of systems with high amounts of minerals or other resources you covet). If you've expanded your territory with actual planets and have a large space from the frontier outpost to the edge of your borders, you can safely decommission the frontier outpost for the spare half-influence a month you could want (After expansion tradition. And maybe turn that around into buying another frontier outpost on the frontier edge of your nations borders).

Rarest 5 (from rarest to least rare):

Last, Best Hope
Rise of the Machines
Deus Vult
Does Not Compute
Towards Utopia

Wow. Other then rise of machines and Does not Compute, those are the 3 hardest I don't have currently. Anyone else have any other hard achievements they want to share?

Stellaris is predictable enough that there's no reason not to play in Ironman.


I think I might try to get the Peacekeeper/Very Open Borders achievements next. Make a xenophile pacifist and go and kick everyone until they behave (and sign migration treaties.

Probably also go tall and spam habitats.

Peacekeepers is actually a very deceptively easy achievement to get which I'm surprised it has such a low percentage earned on global. All you need to do is make two species that are passive. Make one of them always appear in your galaxies, play as the other, and reduce the empires in the galaxy to 2, you and the other guy. After you unpause for a few seconds, you instantly get the achievement where you get to close the game, and play another race you want to play with.