• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

BaBoFantasy

Second Lieutenant
30 Badges
Jun 11, 2017
101
331
steamcommunity.com
  • Cities: Skylines Industries
  • Cities: Skylines
  • Stellaris: Synthetic Dawn
  • Cities: Skylines - Mass Transit
  • Stellaris: Distant Stars
  • Stellaris: Nemesis
  • Stellaris: Necroids
  • Stellaris: Federations
  • Age of Wonders: Planetfall - Revelations
  • Stellaris: Lithoids
  • Age of Wonders: Planetfall Season pass
  • Age of Wonders: Planetfall Premium edition
  • Age of Wonders: Planetfall Deluxe edition
  • Age of Wonders: Planetfall
  • Stellaris: Ancient Relics
  • Surviving Mars: First Colony Edition
  • Stellaris: Megacorp
  • Surviving Mars: First Colony Edition
  • Surviving Mars: Digital Deluxe Edition
  • Stellaris: Apocalypse
  • Stellaris: Humanoids Species Pack
  • Surviving Mars
  • Stellaris - Path to Destruction bundle
  • Stellaris: Leviathans Story Pack
  • Stellaris: Digital Anniversary Edition
  • Stellaris: Galaxy Edition
  • Stellaris: Galaxy Edition
  • Stellaris
  • Europa Universalis IV
  • Crusader Kings II
There has been a persistent issue where ships fail to properly pursue enemy targets, especially in battles that start at long range, and either when enemy ships retreat (such as those with Artillery or Carrier combat computers) or when the enemy does not actively approach (such as starbases or transport ships).
While this issue existed in previous versions, it has become more frequent in Stellaris 4.0 due to the addition of Deep Space Citadels.

The root causes of this issue are:
1. The return_to_formation_distance (RTFD) value is set too low relative to the updated fleet combat scenarios in Stellaris.
2. The 'fleet center of mass' determined by this value is placed awkwardly - somewhere between the fleet's current position and the center of the star system.
3. Ships are not allowed to move beyond the RTFD range from this fleet center of mass.

These issues result in the following problems:
1. Melee ships are unable to effectively close in on retreating artillery ships and are unfairly wiped out. This is most noticeable in battles between fleet of many corvettes and long-range cruiser fleet.
2. When battles start near the edge of a solar system, ships behave oddly by moving toward the system's center at the beginning of combat.
3. In fleets that combine long-range ships (like Titans) with close-range ships, the fleet often fails to approach edge-of-system targets (such as Deep Space Citadels) properly. (See GIF below)

a01_2.gif


Relevant forum links are listed below.


Maybe I can find more, but IMO the problems are well enough shown.













So I made a simple fix mod designed to resolve issues with ship combat behavior.


In this mod, the RTFD value is simply set to a very large number — 2000 — to prevent ships from behaving abnormally due to an incorrectly calculated fleet center of mass.

a02_2.gif


This mod is a partial extract from Better Ship Behavior and is not planned to receive further compatibility support or updates.
For more refined ship combat behavior, better compatibility with other mods, and ongoing support, I recommend using BSB.





I also reported this issue for the bug reports.

 
Last edited:
  • 22
  • 10Like
  • 3Love
  • 1
Reactions:
Thank you for the clear writing on this issue, I can concurr that this is an issue, especially with bioships, as one of the main ship types has extremely short range weapons. Having maulers die since a single enemy ship retreated to the edge of the system gravity well is not ideal behavious...
 
  • 1Like
Reactions:
Let me just stumble into this and say — if we would decouple the whole combat from on-map animations and have the animations just for presentation, we could solve three issues here:

No more odd behaviors.
Behaviors impacting combat performance/effectiveness.
Bad visual presentation by clipping and clumping ships.

In-system battles could be animated and designed to have great and immersive visuals.
Battle calculations could be improved and better balanced, as stuff like position and range can be abstracted — which in turn reduces performance impact.
Ship combat could be better interpreted by players and be more transparent in how fleets perform based on their power and stats, and not get annihilated because they get kited to death by artillery computer ships.
 
  • 11
  • 4Like
  • 4
  • 2
Reactions:
Will this mod disable achievements?

Most likely, yes - using this mod will disable achievements, as it modifies core game files.

Thank you for the clear writing on this issue, I can concurr that this is an issue, especially with bioships, as one of the main ship types has extremely short range weapons. Having maulers die since a single enemy ship retreated to the edge of the system gravity well is not ideal behavious...

You're absolutely right. This issue has become even worse with the latest update, especially due to the addition of ships with extremely short or long ranges - by that, I mean bioships - and structures like Deep Space Citadels that engage in combat without being located at the center of the system.
Previously, battles mostly took place within similar weapon range bands, and although fleets would drift toward the system center, they still ended up clashing directly - so the issue wasn't as noticeable.
 
Last edited:
Let me just stumble into this and say — if we would decouple the whole combat from on-map animations and have the animations just for presentation, we could solve three issues here:

No more odd behaviors.
Behaviors impacting combat performance/effectiveness.
Bad visual presentation by clipping and clumping ships.

In-system battles could be animated and designed to have great and immersive visuals.
Battle calculations could be improved and better balanced, as stuff like position and range can be abstracted — which in turn reduces performance impact.
Ship combat could be better interpreted by players and be more transparent in how fleets perform based on their power and stats, and not get annihilated because they get kited to death by artillery computer ships.

I was already aware of your bold suggestion.
However, the real-time space battles in Stellaris - where every ship, missile, strike craft, and projectile is individually modeled - are just too beautiful to give up.
That said, I completely agree with your point about having ships of various sizes working together organically in combat. I would love to see that as well.
 
I was already aware of your bold suggestion.
However, the real-time space battles in Stellaris - where every ship, missile, strike craft, and projectile is individually modeled - are just too beautiful to give up.
That said, I completely agree with your point about having ships of various sizes working together organically in combat. I would love to see that as well.

I think they are passable with fewer than 50 ships, but above that, the visuals degrade into clumping and clipping, which just looks silly. And it's not even a matter of taste — it's objectively bad-looking when ships clip into each other and follow straight lines to a target, only to start a ballerina dance of perfectly synchronized behavior that doesn’t depict a believable and immersive battle, but rather the result of a computer program strictly instructed to make ships behave.

At no point am I suggesting to remove presentation from the game by decoupling animation from calculation. No — I suggest keeping every ship, missile, strike craft, and projectile individually modeled, but making it look good first and removing the inherent issues of combat animations affecting combat outcomes.

This would make life so much easier for all of us — and the devs — that it's hard to comprehend why we're still running this experiment.
 
  • 2Like
  • 1
Reactions:
Let me just stumble into this and say — if we would decouple the whole combat from on-map animations and have the animations just for presentation, we could solve three issues here:

No more odd behaviors.
Behaviors impacting combat performance/effectiveness.
Bad visual presentation by clipping and clumping ships.

It would be lame if we had short ranged ships near the edge of the sytem with fixed forward arc weapons, and the animation ignored this and made a random space battle animation. The animation must be based on the combat simulation parameters, surely? So I do not see how you can decouple the animation from the gameplay without breaking something.

edit Or in more clear words, if ship has fixed forward short range weapon, I expect to see animation of that, which means enemy must be in short range and directly ahead, and ... I don't see how they can do this without running a simulation and linking the graphics to it.
 
  • 2
Reactions:
It would be lame if we had short ranged ships near the edge of the sytem with fixed forward arc weapons, and the animation ignored this and made a random space battle animation. The animation must be based on the combat simulation parameters, surely? So I do not see how you can decouple the animation from the gameplay without breaking something.

edit Or in more clear words, if ship has fixed forward short range weapon, I expect to see animation of that, which means enemy must be in short range and directly ahead, and ... I don't see how they can do this without running a simulation and linking the graphics to it.
Sure, it must adhere to the models and gear, but things like constant position and range calculations can be dropped.
To be fair, I don’t know if this would reduce the impact ship combat has on the game, but at least you can make sure it looks good without breaking the combat calculations.
 
Sure, it must adhere to the models and gear, but things like constant position and range calculations can be dropped.
To be fair, I don’t know if this would reduce the impact ship combat has on the game, but at least you can make sure it looks good without breaking the combat calculations.

If you ignore the range calculation, how does the graphics side know if it should draw the attack or not?

Viewed the other way with graphics as primary, if you ignore the range calculation, how does the simuilaton know that it can/can't do damage that that it should fly closer?
 
  • 2
Reactions:
If you ignore the range calculation, how does the graphics side know if it should draw the attack or not?

Viewed the other way with graphics as primary, if you ignore the range calculation, how does the simuilaton know that it can/can't do damage that that it should fly closer?

As I said, I'm not sure if animation systems would still need positions, but the other pros for such a shift are still there.

If you have free hand to reduce the battle into a breakdown of phases, task forces, and engagements, you can just use chances and stages where ships engage and when. It can all be abstracted while the visuals remain great. It can also add much deeper ship and fleet composition, as stages and chances would introduce a new layer of design choices and specializations.

The details are tried and tested, and anything is possible then — without adhering to anything related to real-time positions and behaviors.
 
  • 1Like
  • 1
Reactions:
As I said, I'm not sure if animation systems would still need positions, but the other pros for such a shift are still there.

If you have free hand to reduce the battle into a breakdown of phases, task forces, and engagements, you can just use chances and stages where ships engage and when. It can all be abstracted while the visuals remain great. It can also add much deeper ship and fleet composition, as stages and chances would introduce a new layer of design choices and specializations.

The details are tried and tested, and anything is possible then — without adhering to anything related to real-time positions and behaviors.

I don't understand your post. How would it work?
 
I don't understand your post. How would it work?

As I understand it, he’s not referring to Stellaris’s current RTS-style fleet battles, but rather to a phase-based or turn-based combat system like what you see in games like Endless Space or HOI. It seems like he wants a system where a star system is treated as a single point, and combat begins immediately when hostile fleets are present in the same system.

That said, I don’t agree with his suggestion, because I believe there’s still a lot of room to improve Stellaris’s fleet battles without completely overhauling them in the way Imp0815 proposed.
 
  • 3
Reactions:
As I understand it, he’s not referring to Stellaris’s current RTS-style fleet battles, but rather to a phase-based or turn-based combat system like what you see in games like Endless Space or HOI. It seems like he wants a system where a star system is treated as a single point, and combat begins immediately when hostile fleets are present in the same system.

That said, I don’t agree with his suggestion, because I believe there’s still a lot of room to improve Stellaris’s fleet battles without completely overhauling them in the way Imp0815 proposed.
Yes, this is more or less what I imagine. And I think the RTS system has run its course — anything more built on it is making the issue worse, not better.

A system where we decouple real-time positions from the backend calculations would open up so much potential to create immersive and good-looking animations. It would also eliminate the need for in-system micromanagement — which, by the time engagement ranges exceed system limits anyway, is already unnecessary. We'd create a system that’s enjoyable to watch and removes the awkward parts of the RTS nature of things.
 
  • 4
  • 1Like
  • 1
Reactions:
I totally agree with Imp0815.

Battles are one of the few elements of Stellaris that hasn't undergone a thorough overhaul.
In order for the changes to bring any improvement and add strategic and decision-making depth, the mechanical part must be separated from the animation.

I don't understand people who think this could ruin the look of the fights. If anything, the animations and appearance can be improved and we will no longer have to look at a large ball of doom but at something much more pleasant. Especially if the battle took place throughout the entire planetary systems.
 
  • 3
  • 1Like
Reactions:
Interesting find. Good fix. It's been a while since I played, I had to go look up Deep Space Citadels.

The awkward half-abstracted half-simulated battle system has long been one of Stellaris' original sins, dating all the way back to launch. I feel like the devs long ago gave up on improving it; I don't think much has really changed since disengagement was added. So much of the battle system is good-enough kludges to fix fatal flaws. It's workable, certainly, but does keep causing problems. However, that's not to say it needs to be ripped out. You could change a *LOT* in the battle system design-wise without changing the technical side of things very much.

Long ago I tried to rebalance the whole system via modding, but my whole design got broken overnight by a backend change that hardcapped evasion.

IMO, the biggest thing the current system is lacking is nonlinearity. The gritty details of ship design will change how things match up, but you're ultimately just applying multipliers to fleet power. Fighting a Leviathan is just a matter of eyeballing whether its armor or shields and bringing enough ships, there's no like hard technological requirement or game changing technology that gets you in the fight.

An example of nonlinear balance would be damage reduction armor that requires big guns to pierce, or weapons that scale up and hit every ship so you can't win by flooding in with numbers. I'd love it if combat strategy was a mix of fleet mass and 'chess pieces' and the leviathans were like extremist outliers. As is I barely see them. On a lesser scale you also have problems with like mining drones basically being nonentities because of how rapidly they get 'outdated' in fleet power, but thats a tougher problem to handle

edit: The unfortunate downside is that people would get mad if their 100k fleet got uselessly chumped because they didnt understand the mechanic. A system where you can make big differences at the edges but ultimately you can solve all your problems by throwing numbers at it has broader appeal in this manner, since you can play it successfully with less knowledge.
 
Last edited:
  • 5Like
  • 1
  • 1
Reactions: