Doomstacked Doomstack Doom-Thread: ReDoox

[Legendsmith]

@Bozobub that kind of thing has been suggested and it doesn't actually solve the problem. Trying to squash doomstacks like that will just mean that the doomstacks will be a little smaller, and/or there'll be 2 sub-doomstacks fleets flying side by side with an admiral in each, ready to effectively merge during battle. Take a look at the most recent summary. What you're doing is solving doomstacks in name only.

@Drowe I like the proposal for rebuildable forts.

 

[Airowird]

Don't like "doomstacks"? Simply make the economy for them unsupportable, at least w/o a strong research investment in exactly that (thus, ignoring research in other areas). It would be quite easy to add this in as a selectable option during game generation.

Game generation is about creating the physical rules of that galaxy, not about gameplay options. Also, telling the player to pre-emptively look for the right numbers in the hope it'll create a balanced game is NOT good design. The way to solve this issue is have some sort of negative feedback when you the wrong way, not an invisible wall set at the start, because everyone will be hugging that wall getting as close to where they want to be anyway.

And to reiterate the counters that have been given before:

• "Soft" (current) ship cap.
  Obviously this isn't working as well as it should, or we wouldn't be on page 36 right now.
  
• "Hard" ship cap.
  Doesn't solve doomstacks, at most makes naval capacity more valuable. Considering it's a repeatable tech in society, doesn't actually solve much.
  
• Fleet caps. In other words, after a certain point any ship built is on its own and can only join an existing fleet. You would have to combine this with...
• Ships-per-fleet caps, modified by admiral skill, if one is present. Fleets with no admiral should have a set cap (say, 10-20 points).
  Nothing prevents me from having 500 single Corvettes instead of a single doomstack. This either makes Admirals useless (cfr. Generals) or only imposes a mandatory amount of admirals as requirement for effective warfare. (If I make max fleets of max size and just always send them together) And again, these are arbitrary numbers that make no logical sense within the system. It also leads to weird issues should the admiral die of old age for example. There is a reason that outside research projects, no leader levels are used anywhere as a requirement for a game function.
  
• Economic penalties (basically, selectable maintenance costs).
  Penalties already exist in the overcap cost + docking reduction = movement penalty. IF you're talking about just scaling maintenance cost, it would only shift economy more towards Energy Credits, but again, does not solve anything about the motivation of using doomstacks, only the size of ALL doomstacks.
  
• Ship build cost multiplier (say, 0.5x for beginners and 5x or even more, for bored masochists).
  This would actually benefit doomstacks more, as you always want to make sure to minimize ship losses (because they are so expensive) Losing a doomstack fight would then instantly be game over, because it would take you 100y to rebuild.

 

[JulienJaden]

Game generation is about creating the physical rules of that galaxy, not about gameplay options. Also, telling the player to pre-emptively look for the right numbers in the hope it'll create a balanced game is NOT good design. The way to solve this issue is have some sort of negative feedback when you the wrong way, not an invisible wall set at the start, because everyone will be hugging that wall getting as close to where they want to be anyway.

And to reiterate the counters that have been given before:

• "Soft" (current) ship cap.
  Obviously this isn't working as well as it should, or we wouldn't be on page 36 right now.
  
• "Hard" ship cap.
  Doesn't solve doomstacks, at most makes naval capacity more valuable. Considering it's a repeatable tech in society, doesn't actually solve much.
  
• Fleet caps. In other words, after a certain point any ship built is on its own and can only join an existing fleet. You would have to combine this with...
• Ships-per-fleet caps, modified by admiral skill, if one is present. Fleets with no admiral should have a set cap (say, 10-20 points).
  Nothing prevents me from having 500 single Corvettes instead of a single doomstack. This either makes Admirals useless (cfr. Generals) or only imposes a mandatory amount of admirals as requirement for effective warfare. (If I make max fleets of max size and just always send them together) And again, these are arbitrary numbers that make no logical sense within the system. It also leads to weird issues should the admiral die of old age for example. There is a reason that outside research projects, no leader levels are used anywhere as a requirement for a game function.
  
• Economic penalties (basically, selectable maintenance costs).
  Penalties already exist in the overcap cost + docking reduction = movement penalty. IF you're talking about just scaling maintenance cost, it would only shift economy more towards Energy Credits, but again, does not solve anything about the motivation of using doomstacks, only the size of ALL doomstacks.
  
• Ship build cost multiplier (say, 0.5x for beginners and 5x or even more, for bored masochists).
  This would actually benefit doomstacks more, as you always want to make sure to minimize ship losses (because they are so expensive) Losing a doomstack fight would then instantly be game over, because it would take you 100y to rebuild.

I agree with most of what you're saying. Obviously, lowering the overall ship capacity is stupid, considering the fact that Stellaris is about playing a space empire, so a hard cap at 20 battleships or even 40 battleships makes no sense when you have a population of several billion individuals, spread out over several planets, and many mining stations to boot - it would simply feel arbitrary to have that kind of economy but be told by the game that it can't support the military to match and protect it (makes even less sense when you add ethics to the equation - yeah, pacifists should probably have a lower fleet cap, but a militarist empire would pretty much always put the fleet first).

So, rather than having negative, arbitrary reasons why you can't have doomstacks, the solution has to be positive, an incentive, a strategic interest to split your fleets up so they can tackle multiple, vital tasks at once.
Looking at, say, Hearts of Iron is not the worst idea: The armies are huge but spread out along a frontline. Why? Because a single clumped-up doomstack would be quickly cut-off from supplies and communication, surrounded and destroyed, regardless of its size. Obviously, supplies play less of a role in Stellaris, but maybe communication could be part of a solution.


Follow me here for a moment and let's ask a question: How does communication work in the Stellaris universe?
Well, for all we know, they somehow figured out how to make real-time interstellar phone calls a reality.
But what if they hadn't? If you look at the lore of Mass Effect (the first one), you see that they recognized the problem and solved it through the use of communication buoys (basically interstellar phone masts) that, as a result, are strategic targets when conflicts break out.
Cutting the lines of communication within an empire's territory can send it into disarray and make people on the cut-off planets feel insecure, hurt morale, hamper production even if the mining stations are untouched.
A fleet operating behind enemy lines has its sensors, sure, but it's of strategic importance that they stay in contact with HQ and can both relay their information and status back to high command AND stay up-to-date on the rest of the conflict - i.e. setting up temporary lines of communication and defending them is important; being cut off could result in morale hits, long delays for new orders (think messenger ships trying to make their way back to the next connected buoy - and they might be intercepted too) and other possible impacts on their performance. Suddenly, having your entire fleet fly through hostile territory as a doomstack becomes a huge risk: If they are cut off, you might be practically defenseless for a while, perhaps long enough for the enemy to do significant damage, even if they have an inferior fleet.

It's not an easy problem to solve, but having a mechanic like communication and communication arrays in the game, with serious, and I mean SERIOUS consequences to having them destroyed, would emphasize the importance of defensive stations a lot more, it would make spreading out your forces throughout your empire practically a necessity (and full frontal assault with all fleets much more risky) and it would make the empire's space feel less empty because now you'd have to set up lines of communication - and the more important the line, the more important it is to set up redudancy to make sure that a single corvette taking a lucky shot can't just cut you off from half your empire.

Obviously, I'm not saying that it would have to be exactly as I describe it here, but communication is one of the necessities we will practically never be able to circumvent, so using that as a possible starting point for solving this problem and making warfare more complex might not be the worst idea.

 

[Drowe]

I agree with most of what you're saying. Obviously, lowering the overall ship capacity is stupid, considering the fact that Stellaris is about playing a space empire, so a hard cap at 20 battleships or even 40 battleships makes no sense when you have a population of several billion individuals, spread out over several planets, and many mining stations to boot - it would simply feel arbitrary to have that kind of economy but be told by the game that it can't support the military to match and protect it (makes even less sense when you add ethics to the equation - yeah, pacifists should probably have a lower fleet cap, but a militarist empire would pretty much always put the fleet first).

So, rather than having negative, arbitrary reasons why you can't have doomstacks, the solution has to be positive, an incentive, a strategic interest to split your fleets up so they can tackle multiple, vital tasks at once.
Looking at, say, Hearts of Iron is not the worst idea: The armies are huge but spread out along a frontline. Why? Because a single clumped-up doomstack would be quickly cut-off from supplies and communication, surrounded and destroyed, regardless of its size. Obviously, supplies play less of a role in Stellaris, but maybe communication could be part of a solution.


Follow me here for a moment and let's ask a question: How does communication work in the Stellaris universe?
Well, for all we know, they somehow figured out how to make real-time interstellar phone calls a reality.
But what if they hadn't? If you look at the lore of Mass Effect (the first one), you see that they recognized the problem and solved it through the use of communication buoys (basically interstellar phone masts) that, as a result, are strategic targets when conflicts break out.
Cutting the lines of communication within an empire's territory can send it into disarray and make people on the cut-off planets feel insecure, hurt morale, hamper production even if the mining stations are untouched.
A fleet operating behind enemy lines has its sensors, sure, but it's of strategic importance that they stay in contact with HQ and can both relay their information and status back to high command AND stay up-to-date on the rest of the conflict - i.e. setting up temporary lines of communication and defending them is important; being cut off could result in morale hits, long delays for new orders (think messenger ships trying to make their way back to the next connected buoy - and they might be intercepted too) and other possible impacts on their performance. Suddenly, having your entire fleet fly through hostile territory as a doomstack becomes a huge risk: If they are cut off, you might be practically defenseless for a while, perhaps long enough for the enemy to do significant damage, even if they have an inferior fleet.

It's not an easy problem to solve, but having a mechanic like communication and communication arrays in the game, with serious, and I mean SERIOUS consequences to having them destroyed, would emphasize the importance of defensive stations a lot more, it would make spreading out your forces throughout your empire practically a necessity (and full frontal assault with all fleets much more risky) and it would make the empire's space feel less empty because now you'd have to set up lines of communication - and the more important the line, the more important it is to set up redudancy to make sure that a single corvette taking a lucky shot can't just cut you off from half your empire.

Obviously, I'm not saying that it would have to be exactly as I describe it here, but communication is one of the necessities we will practically never be able to circumvent, so using that as a possible starting point for solving this problem and making warfare more complex might not be the worst idea.

This is in essence a variation of the supply lines mechanic, except with communication instead of supplies. I think supplies would be a much more useful mechanic than communication though.

Historically the lines of communication were very slow, and could be interrupted much more easily than today. Before fast long distance communication was possible, armies have been fighting far away from home with no contact to their HQ, so it is a problem that can be solved and we know how to solve it too. Just give the commanders of your fleets a lot of leeway to act without orders from HQ, in most cases simply an objective they should achieve and allow them to find a way to follow those orders. This would be an inevitable contingency if FTL communication is possible and interruptible, and standard operating procedure if there is no FTL communication. So no, it shouldn't work the way you described it, but that doesn't mean there can't be other consequences. You may be unable to give any new orders to a fleet out of communication range, except for emergency FTL. A fleet out of communication range will carry out its orders before returning, though if it suffers heavy losses, it will return home early. The problem with this is, that the AI would have a hard time dealing with such a mechanic, though it may be possible to do well. It would add a lot of strategic planning and thinking ahead, and it would probably mostly solve doomstacks, since going on the offensive with your whole fleet, without a way to recall it will force you to keep at least part of your fleet back for defense and reacting to enemy fleet movement. Though I suspect it would require a lot of changes to the AI so it doesn't become frustrating.

 

[fabius]

My guess is that the optimum gameplay solution lies in a balance of several suggestions being applied so that local and strategic combat is more fun and has more meaningful decisions.

Something tangential occurred to me:

How would soft capping the size of fleet that can travel to a system at the same time impact this. It would mean total fleet size is much bigger than actual fleets size.

The idea came to me from a Star Trek episode where they mentioned a Federation speed limit because Warping space was causing damage such as temporal rifts and anomalies.

It seems feasible that the mass of Fleet in Warp could be limited by safety.
Same goes for hyperspace and worm hole use.

Increasingly go over the safe limit, and increasing risk rising percentage chance of each ship being damaged or even destroyed by temporal rips or worse. You could do a Hannibal and fluke a big fleet all in one, but luck need be with you.

The could be a cool down for warp or jump into a system. This could add tactic of building a fleet with less damage output but staying power to pin an enemy fleet for reinforcements to arrive.

Also, I think moral of ships captains and crew should help with the all or nothing battles.


I like the sound of a weighted supply range system; and flanking bonuses.

 

[Drowe]

I like the sound of a weighted supply range system; and flanking bonuses.

Flanking has already been talked to death, that only increases micromanagement but does nothing to solve doomstacks.

My guess is that the optimum gameplay solution lies in a balance of several suggestions being applied so that local and strategic combat is more fun and has more meaningful decisions.

Yes, that is the current consensus.

Something tangential occurred to me:

How would soft capping the size of fleet that can travel to a system at the same time impact this. It would mean total fleet size is much bigger than actual fleets size.

The idea came to me from a Star Trek episode where they mentioned a Federation speed limit because Warping space was causing damage such as temporal rifts and anomalies.

It seems feasible that the mass of Fleet in Warp could be limited by safety.
Same goes for hyperspace and worm hole use.

Increasingly go over the safe limit, and increasing risk rising percentage chance of each ship being damaged or even destroyed by temporal rips or worse. You could do a Hannibal and fluke a big fleet all in one, but luck need be with you.

The could be a cool down for warp or jump into a system. This could add tactic of building a fleet with less damage output but staying power to pin an enemy fleet for reinforcements to arrive.

It would work only if both the cooldown and the soft cap were implemented. But I'm not sure it's a desirable approach, in essence you would have smaller doomstacks or feed in your fleets piecemeal, unless you crank up the cooldown so that it blocks movement for a significant amount of time, which would be terribly annoying. While a soft cap is better than a hard cap, and you give a good reason for it, it's preferable not to use caps where it can be avoided.

Also, I think moral of ships captains and crew should help with the all or nothing battles.

A moral system might be part of a solution, though in my opinion there are better ways to solve the all or nothing combat, for example by being able to retreat in a controlled fashion instead of a scattered retreat like we have now. The moral system could work with that, so that if you lost enough moral you automatically use the emergency FTL.

 

[fabius]

If flanking would increase micro- what about combined with cut the player control of fleet movement in system. Maybe just set tactics.

Re soft cap and cool down. I meant have a soft cap; and then cool down before the end of that fleets mass being counted for the increasing random chance of damage.

Not to sure about that it would lead to smaller doom stacks. Smaller fleets is the consensus desire, right? So that the game has multiple fleet manoeuvring on the strategic scale. Whatever comes, there will always be a rough optimum fleet size. The point is that it would be much more fun if it was just all the ships possible in one fleet.

 

[EntropyAvatar]

• Slower FTL
  • This doesn't change much, and could make colonisation and exploration slower for no reason.

I appreciate your summary posts, but I think this line is fairly misleading.

Things Slower FTL changes:

1. Takes longer for a doomstack to chase down small fleets
2. Takes longer for a doomstack to reach core worlds
3. 1 and 2 mean that there is more time to build more ships before the war is decided
4. Makes it easier for a player to simultaneously control multiple independent fleets
5. Gives more time for an outmatched fleet to run away (and for a player to notice that an outmatches fleet is in danger)

Slower FTL doesn't slow down colonization much: The travel time of colony ships is right now a very tiny factor in overall speed of colonization. Travel time could therefore be greatly increased with minimal influence on overall colonization speed.

Slower FTL doesn't slow down surveying much: As above. Most of the time in surveying is moving about the system and scanning.

Slow FTL does slow down "poke head into system" exploration: This is true, but I think it would be a good thing. Especially if using hyperspace, you can know where all the habitable planets are within scores and scores of systems very very quickly in Stellaris in comparison to other games. Why is this a good thing? Slower exploration is more strategic because you have to make decisions in the absence of complete information. Do I send my colony ship to system A, which is a marginal location, or hold off until I know more? Basic exploration in Stellaris is perilously close to a "reveal map" button and would benefit from being slower.

One last thing: Does anyone want FTL in Stellaris to be FASTER? Consider how a game of Stellaris will play out if FTL is 3x faster. Managing multiple fleets is harder, your fleet's position in a system's gravity well matters more than it's position in the galaxy, there is less time to react to enemy fleet movements and you opponent from across the galaxy can be at your homeworld 3x faster. Your starting corvettes can map the galaxy in the first year. Does that sound like it wouldn't change much?

 

[Drowe]

If flanking would increase micro- what about combined with cut the player control of fleet movement in system. Maybe just set tactics.

Having just one fleet or multiple within the same system doesn't really make a difference, it's still a doomstack. Read the summary one or two pages back.

Re soft cap and cool down. I meant have a soft cap; and then cool down before the end of that fleets mass being counted for the increasing random chance of damage.

Ah, OK so basically the more ships you send, the higher the risk, whether you send them one by one or all in one fleet doesn't matter unless you send them after the cooldown has elapsed.

That leads to the question, does that only apply to your own ships or is it irrelevant who the ships belong to? Provided two empires use the same FTL method, then they would logically have to use the same cooldown. Which would be easy to exploit in order to harm an enemy fleet. On the other hand if they don't, why not? Can you get around that cooldown by jumping in from a different system? If yes, then you have doomstacks again, if no, why not? How will the endgame crisis deal with that? Are they exempt? What about FE/AE? Can you still reasonably fight them with such a limitation if they are exempt from the cooldown? Are they still a challenge if they are not?

 

[fabius]

I've re-red the summary and I don't see it that way. A combination of changes that promote the use of many fleets vs just one doom stack is not 'still doomstacking'.

Regarding the impact on other areas, as so often the answer will likely involve balancing.

 

[Drowe]

I've re-red the summary and I don't see it that way. A combination of changes that promote the use of many fleets vs just one doom stack is not 'still doomstacking'.

Regarding the impact on other areas, as so often the answer will likely involve balancing.

If you have a single fleet with 100 ships in a battle or two fleets with 50 ships each does not really matter, since you don't have control over them there is no real difference. You can already do that right now, not that you get any benefits from it except keep one fleet from being targeted by spinal mounted weapons maybe, but you can do it.

The main problem with your idea however is that it treats the symptom but ignores the root problem. There are two fundamental issues that make doomstacks the best strategy. The first is the lack of valuable targets, the way you win is by smashing your opponents fleet and spaceports, capturing planets is just for warscore and anything else is not even worth paying attention to. The second is the nature of the battles, they only end in one of two ways, either a fleet is destroyed completely, or it retreats with emergency FTL, which removes it from your control for a significant amount of time. Those two things have doomstacks as consequence, since you have to protect your fleet, or you lose the war, and if you get into a fight, you better make sure you win or you're in trouble.

 

[Hawticecubes]

I think the problem of doomstacks can be easily solved by incentivising splitting up fleets to cover many systems and lessening the blow of losing spaceports through:
1. Allowing the building of ships outside of planetary spaceports. This would incentivise splitting up fleets to attack ship production centres after a battle, while lessening the blow of losing spaceports.
2. Allow occupation/capturing of stations and utilisation of enemy resources so that rapid raiding of economy is a viable way to win or profit off a war
3. Add more depth to land battles so that conquering and defending planets can become prolonged affairs. This would allow using planets as forts to delay attackers after a loss in order to rebuild a fleet.

This would also help the ai since its currently doomstacking and runing the entire stack to defend whichever system your 1 corvette appeared in.

 

[JulienJaden]

This is in essence a variation of the supply lines mechanic, except with communication instead of supplies. I think supplies would be a much more useful mechanic than communication though.

Historically the lines of communication were very slow, and could be interrupted much more easily than today. Before fast long distance communication was possible, armies have been fighting far away from home with no contact to their HQ, so it is a problem that can be solved and we know how to solve it too. Just give the commanders of your fleets a lot of leeway to act without orders from HQ, in most cases simply an objective they should achieve and allow them to find a way to follow those orders. This would be an inevitable contingency if FTL communication is possible and interruptible, and standard operating procedure if there is no FTL communication. So no, it shouldn't work the way you described it, but that doesn't mean there can't be other consequences. You may be unable to give any new orders to a fleet out of communication range, except for emergency FTL. A fleet out of communication range will carry out its orders before returning, though if it suffers heavy losses, it will return home early. The problem with this is, that the AI would have a hard time dealing with such a mechanic, though it may be possible to do well. It would add a lot of strategic planning and thinking ahead, and it would probably mostly solve doomstacks, since going on the offensive with your whole fleet, without a way to recall it will force you to keep at least part of your fleet back for defense and reacting to enemy fleet movement. Though I suspect it would require a lot of changes to the AI so it doesn't become frustrating.

That's the thing: I decided against the idea of supplies because that makes less sense to me than communications.
Maintaing communications throughout the empire are a base necessity - if you're cut off from everybody else, and are blind and deaf, you can't fight, let alone win.
Maintaining communications with the fleet isn't important for the tactical side of it (that's what admirals are for), so you're absolutely right with that you wrote on that, but it's something that's vital for
a) knowing what the fleet is finding (if they gather intel that could be of strategic importance but they have no way of letting HQ know, it's worthless), and
b) issuing orders pertaining not only to their specific mission but the entire theatre of war - an admiral who's preoccupied with his fleet can't see the bigger picture but somebody with information from ALL fleets and station might.
And while I understand why it might make sense from a gameplay POV to be able to order emergency FTL even when they are otherwise cut off, I'm not sure that should be up to the player - if the fleet were to be cut off, maybe everything should be up to the admiral, including decisions like that (which would make thhe choice of the admiral that much more important, especially if they should ever introduce character traits to Stellaris [that actually have an impact on behavior], kinda like they did to EU4).
It would be relatively easy to represent ingame (through proper stations within the empire and buoys in systems your fleet passes - connections between communication arrays could be easily highlighted on the galaxy map and would probably kinda look like the hyperlane map).
The consequences of losing this connection can be fairly straightforward and logical, and their importance would be easily recognizable.

But most importantly: It would organically make doomstacks a highly risky option without arbitrarily forbidding players to use them, whereas supplies are a lot more complicated and heavier on micro-management than that. As long as your ships aren't operating outside your empire for long, they wouldn't really run out of supplies, since space ships are built for just that: Operating in the hostile environment that is space for extended periods of time. Sending your fleets home to refuel every couple of jumps sounds to me like it could more easily devolve into the kind of micro that makes flanking a bad approach (besides, all generators ingame are of the kind that doesn't really require a lot of fuel, if restocking on food became an issue, the logical problem would be that ships of that size couldn't store noms for more than a couple of years, which is not exactly a long span of time in the game, so it could very easily feel like too big of a range limit, especially for research vessels etc.).

That said, I can see how giving the player more control over something like stockpiling supplies, setting up depots near the front and splitting up the fleet because it's more difficult to support one doomstack in one system, rather than three spread-out fleets, could be beneficial. I just feel that, if it came down to choosing one approach or the other for a new mechanic, communications would make more sense, both gameplay-wise and lore-wise.


But generally speaking, I agree that a new mechanic like this would only deal with a part of the issue that is warfare, so only a combination of multiple possible changes to the game can really improve it.

 

[Zer0k]

Directly increase fortress HP/Damage/Power

• This doesn't solve doomstacks itself because the increase in power means that fleets want to concentrate more firepower in order to beat the strong forts, but combined with other factors it could have a place
• They'll still get outclassed if this is all.

You don't use them to defend, you use them to "lock" the engaging fleet long enough that your smaller inferior fleet can do some damage elsewhere. If you add this on top of slower FTL (but only for military ships) and add high-value targets, then no one will risk going to war with "everything" they have.

But for this to happen, HP needs to increase by a lot (damage would probably need to be lowered).

 

[Legendsmith]

Summary of Reasons the Doom-Stack is King™
p37 Edition
"Every battle in Stellaris is Midway, only more decisive and with more damage. Every battle is like Midway with US carriers parked in Japanese ports a week later" - @durbal

1. Every engagement is a full pitched battle. This makes doomstacking the best because in a pitched battle you want your maximum force there. There's no squadrons of fast torpedo corvettes raiding the lumbering enemy battleship blob with bombing runs. Also related to Issue #4 as every battle ends with total, or near total annihilation of the loser.
2. The Enemy fleet is the only meaningful target for each side. Doomstacking is the safest way to win any war because hostile fleets that are not targeting your fleet can be ignored, and then mopped up. Nothing has an meaningful impact on your current war situation except losing your fleet. Nothing impacts your post-war status except a loss or win. Pyrrhic victories are virtually impossible.
3. Non-fleet defences are useless. Fortresses and starports are speedbumps at best."The impotency of starports and fortress' means that stellaris does multi-front or pan-galactic war very poorly, as of now your only defensive force is your blob fleet, which is also your offensive fleet." - wastedswan
   We could also include defensive armies in this category, as they're irrelevant at best.
4. A losing fleet loses hard, and is quickly wiped out, this makes smaller fleets suicide because they die so quickly with no chance of reinforcement. Admirals apparently never fight delaying actions in Stellaris. (Related to #1) The speed at which a doom stack can mop up smaller fleets contributes to its superiority.
   At the moment unless a reinforcement fleet warps on top of the fight, by the time they get to the fight it'll already be over. Therefore sending reinforcements is useless because the friendlies will already be dead.
5. You can't outrun the enemy except with superior technology. Every fleet has the same strategic movement capability and thus the same capability to respond to threats. This is related to #1. Again, there's no squadrons of fast torpedo corvettes raiding the lumbering enemy battleship blob with bombing runs. (if there was, defence stations/forts/platforms might be actually useful.)
6. Smaller fleets are currently too risky. See #1, #4 and #5, for reasons why smaller fleets are discouraged, and additionally #2, because there's no incentive.
7. Rebuilding Speed Once a fleet is destroyed it cannot be rebuilt in any kind of timeframe relevant to the war. (Suggested by @Summin Cool ).
   This might not matter so much if there was less need to rebuild entire fleets.

These reasons are quite varied, and no single change can address all of them. Thus critiquing any suggestion because it will not solve doom-stacking completely is not a valid critique, it must be shown why it would only slightly change the problem, such as that of admiral fleet capacity caps turning the problem into everyone having 2 doomstacks instead of 1.

Summary of Suggested Solutions (and some Problems/Rebuttals)
These are in no particular order. Not all of these are necessarily good, but must be included in the summary so they're not re-suggested for the Nth time.
Red topics have been discussed to death and need to stay dead.
Green topics have been regarded as near universally good moves by the denizens of Paradox Forumland
White topicscould be good, could be not good, or where the consensus is unclear.

• Slower FTL For Military Ships
  • Might be a good idea. Details here by @EntropyAvatar.
    1. Takes longer for a doomstack to chase down small fleets
    2. Takes longer for a doomstack to reach core worlds
    3. 1 and 2 mean that there is more time to build more ships before the war is decided
    4. Makes it easier for a player to simultaneously control multiple independent fleets
    5. Gives more time for an outmatched fleet to run away (and for a player to notice that an outmatches fleet is in danger)
  • This change could scale with ship size, corvettes (And civilian ships) being the fastest, while the increasing sizes of other ships = progressively slower FTL, whether it be windup, cooldown, or transit time. This partially addresses #4
• Admirals as a size limit for fleets
  • Not a solution because everyone just has their doomstack turned into 2, 3, or 4 mini doomstacks that behave exactly the same way. Even combined with other changes it would be those other changes having the impact, rather than this one.
  • @Airowird's way of putting it: A flat fleet cap is pointless, because 2 half-doomstacks flying together still have the same effect as a single one...It provides no incentive to split fleets, only rules. Game Design 101: Any arbitrary rule to limit players only limits enjoyement of min-maxing, not the reason/source of the problem.
• Rebuildable/Reinforcable/Redeployable Fortresses & Other Static defences.
  • Making defences less of a waste of minerals that can't survive encourages people to build them. A fortress that could actually be defended by a friendly fleet rushing to its aid allows smaller fleets to have a 'home turf advantage' from the fortress's firepower.
  • This could make defences more annoying
  • But defences are supposed to be inconvenient.
  • What if fortresses could be captured?
  • Redeployable fortresses makes them less of a permanent minerals sink that can never be moved from a position that could become strategically irrelevant later.
• Directly increase fortress & other static defence HP/Damage/Power
  • This doesn't solve doomstacks itself because the increase in power means that fleets want to concentrate more firepower in order to beat the strong forts, but combined with other factors it could have a place
  • They'll still get outclassed if this is all.
• Faster retreat times
  • Nobody really wants this, but it's been suggested. Faster retreat times are extremely frustrating and turn warfare into a game of "chase down the enemy fleet" or "run from the enemy doomstack" as soon as one side starts to lose the first battle, especially due to the AI's ability to hit it ASAP.
• Flanking Bonuses.
  • As far as I can see, the general response to this is that it is a post-hoc mechanic that has more elegant solutions. It's also very situational and suffers from the same problem that @Drowe elaborated on with AOE weapons (below); it stops doomstacks in name only.
  • "Flanking has already been talked to death, that only increases micromanagement but does nothing to solve doomstacks." - Drowe
• Planets as high value targets aka Consequences of invasion & bombardment
  • This seems to be universally accepted as a good idea.
  • THIS DOES NOT JUST MEAN MAKING PLANETS WORTH MORE WARSCORE, though it could include that.
  • Suggestions that fall under this:
    • Long term happiness debuffs for any planet that's invaded, or bombarded.
    • Harsher Bombardment penalties.
    • The ability to destroy buildings with invasion and/or bombardment. Planetary combat has to happen somewhere, why doesn't a mine get wrecked or a farm shelled to bits?
    • The ability to capture and, destroy, steal, or render-inaccessible portions of the enemy's stored resources upon capturing a planet.
    • Loot resources, possibly from enemy's stored reserves by destroying buildings on occupied worlds - Suggested here by @Undar
    • Auto-enslavement of occupied populations.
    • Diversion of produced resources to occupier's empire.
• Supply limits/chains. This could really be done well or awfully.
  • Done well, supply chain/supply limits discourages sending a doomstack around for every single task, and makes sending a fleet deep into enemy territory a costly endeavour.
  • Done poorly this just creates another variation on the doomstack theme that will be immediately min/maxed out again.
  • A proposed implementation is Weighted supply range by @EvilKnievel82
• Diminishing firepower returns for large vs small fleets (Slow down the rate of death for losing fleets)
  • This can be called combat width, coordination penalty, or whatever. Basically it means that larger fleets will still defeat, but not immediately 'delete' smaller fleets. They will kill them more slowly, up to a point (unless the smaller fleet is very significantly smaller in which case it'll still be deleted). This slows down battles a bit and makes splitting fleets up a less risky move. The slowing of battle also means that it's not a great idea to send your whole fleet to kill something a quarter of its size because it'll be tied up for too long in a battle that yes it will win, but it's just so much overkill.
  • Rebuttals: Doesn't make sense, everyone's easy to hit in space.
  • @Airowird's rebuttal: "(Relative) Fleet power reduction does nothing outside of making fights lasts longer, as you still want to build up a doomstack as much as possible just in case the other guy brought more friends than expected." (But making fights last longer still helps mitigate doomstacks)
• Hearts of Iron TFH style combat tactics.
  • Pretty sure someone suggested this, it seems like it might be good, it could help address issues #1 and #4. It could also make admirals more important. It's related to to "Diminishing firepower returns for large vs small fleets.", as the different tactics that the combat AI uses could slow down the combat with 'fleet manouvers' that provide -#% damage to enemy, and similar things.
• System Wide Auras for Stations.
  • Suggested by @Drowe and expanded here by Legendsmith. This concept allows stations to be a meaningful kind of defence without encouraging doomstacks. Defence auras affect whole systems, and yet do not require a doomstack to kill, thus achieving the goal of delaying the enemy. Every station contributes to an aura score for the system, which maxes out at 100%.
• Auto-Retreat/Morale mechanics Fleets currently fight to the death every single time unless the player hits emergency FTL. Is every captain and crew a fanatic? Apparently so. There's no way to defeat an enemy without just crushing them physically, which means there truly is no recovery for the losing side. This is related to Issues #1 and #4 .
• Faster Ship building Suggested by @Summin Cool, this change would make ship building faster and allow a loser to recover faster. (Details are apparently to come.)
• In-Combat Controls/RTS Controls. I'm putting these two things under one header because they are both a similar thing; the ability to affect fleet behaviour during combat (other than emergency FTL which just ends combat).
  • @Airowird 's post here has some suggestions. Summary: "The option to set a fleet to Evasive during combat. Evasive fleets have 30% more Evasion, but 30% less Accuracy and Damage, and will try to move out of the gravity well to jump back to the last 'safe' system."
  • This seems unlikely to happen as Stellaris is not an RTS and does not operate on an RTS scale, plus it would create a lot of micro if doomstacks weren't the go-to strategy, because there'd be more fleets to manually manage.
• Movable/Redeployable stations/Fortresses
  • Stations that can FTL but are rendered inoperable during and for a period afterwards
  • Alternately, allow construction ships to deconstruct and then reconstruct stations for an energy-credits cost.
• More cost-effective Defensive fleets that can't leave an Empire's borders.
  • Do we really need to make fortresses and defence stations even more irrelevant though?
• Organisation/Fleet Cohesion loss After a big engagement, your ships suffer from cohesion loss resulting in diminished combat stats. Fleets regain cohesion, and when they do it at a spaceport, the cohesion gains are a lot faster.
• Moving under fire, aka "Not charging into battle just because you're in the enemy weapon range". This basically removes Issue #1 altogether.
• Greater Abstraction/Automation:
  • Strategic box variant Suggested by Hammer54: One possibility is to introduce a strategic "box" for each system, where you can leave raiders in enemy systems and gurilla warfare ships/escorts in your own. This could force the larger empire to disperse fleet power in more systems, and gives the weaker party something to do when they cant engage the blob. The point with a strategic "box" is that you dont have to micro it, you can just leave the corvets there. They'll do damage over time, and wont be killed off instantly.
  • System-Battlefield Suggested by @skydiver1: ? Why not treat the entire system as one big battlefield where the player can influence events but ships can manage complex behaviour on their own, attacking targets, chasing and being chased. Then mechanics can be introduced that allow smaller fleets to serve as effective "partisan" harrassers, that cannot be ignored and quickly dealt with but require multiple sizable fleets to keep them at bay
• AOE weapons. To discourage large blobs, AOE weapons that damage large numbers of ships a once could be used. Suggested by @REJS7
  • This would just make AOE weapons the most OP thing in the game, and what's to stop a larger AOE equipped doomstack from just killing a smaller fleet faster?
  • "This has already been brought up multiple times, it doesn't solve doomstacks except for in the strictest sense of the word. People would get around it by RTS style micromanaging multiple smaller fleets within the same system, which technically is still a doomstack in all things that matter." - @Drowe

New things from Last Summary: Added @EntropyAvatar's details on slower FTL.

Summary end.

I saw some discussion of what counts as solving doomstacks. I believe that that just encouraging the players to split their huge fleets into 2 and have one giant decisive battle made out of 4 or 5 fleets that still keep together and then duke it out together in a single winner-take-all decisive engagement is not solving doomstacks because the fights will be exactly the same, except with flanking or whatever. We're solving that single decisive engagement that has the victor win at that moment; everything after is mopping up. That's why I made the Summary's header quote "every battle is like Midway with US carriers parked in Japanese ports a week later". Doomstacks are the best way of conducting that decisive battle type of warfare, which is most efficient & least risky type at the moment in Stellaris.

You don't use them to defend, you use them to "lock" the engaging fleet long enough that your smaller inferior fleet can do some damage elsewhere. If you add this on top of slower FTL (but only for military ships) and add high-value targets, then no one will risk going to war with "everything" they have.

But for this to happen, HP needs to increase by a lot (damage would probably need to be lowered).

I think that's frustrating gameplay. if a fortress is a giant chunk of HP that you need to just wait for your fleet to whittle down, people are going to stack as much as possible on it for it to have some effect, and that HP would have to massively increase for it to be able to have meaningful delays, even with slower FTL (Because the smaller fleets it is delaying for will have slower FTL too). If it's low firepower and high HP not much interesting is going on.

This is one of the reasons I suggested system wide auras, because it provides a way for defences to slow down the enemy while actually providing some interesting strategic choices, AND without requiring a huge fleet to kill it. Big tanky forts will just encourage large fleets that can kill them as fast as possible. Forts do need a firepower/survivability buff, but it shouldn't be all.

 

[Airowird]

Slower FTL doesn't slow down colonization much: The travel time of colony ships is right now a very tiny factor in overall speed of colonization. Travel time could therefore be greatly increased with minimal influence on overall colonization speed.

Slower FTL doesn't slow down surveying much: As above. Most of the time in surveying is moving about the system and scanning.

Slow FTL does slow down "poke head into system" exploration: This is true, but I think it would be a good thing. Especially if using hyperspace, you can know where all the habitable planets are within scores and scores of systems very very quickly in Stellaris in comparison to other games. Why is this a good thing? Slower exploration is more strategic because you have to make decisions in the absence of complete information. Do I send my colony ship to system A, which is a marginal location, or hold off until I know more? Basic exploration in Stellaris is perilously close to a "reveal map" button and would benefit from being slower.

One last thing: Does anyone want FTL in Stellaris to be FASTER? Consider how a game of Stellaris will play out if FTL is 3x faster. Managing multiple fleets is harder, your fleet's position in a system's gravity well matters more than it's position in the galaxy, there is less time to react to enemy fleet movements and you opponent from across the galaxy can be at your homeworld 3x faster. Your starting corvettes can map the galaxy in the first year. Does that sound like it wouldn't change much?

What if larger fleets get a longer cooldown/travel period?
Something like sqrt(fleet_size) extra days to re-establish communications. Or for calculations during windup on not hitting eachother in warp space, similar to wormhole tech penalty.
The point would be that a larger fleet requires more internal communication and management as well, so there is a reason to not lob everything together.
Ofcourse, this penalty would be for all windups occuring in the system towards the same one, so your 5-fleet doomstack can't actually catch up to the smaller fleet and is forced to bait them in. Alternatively, you could only allow 1 fleet at a time in a lane, this trickles out doomstacks and makes them vulnerable for traps, as you can pick off a part before reinforcements come. (similar to combat width actually)

What if you had to move to the relevant 'side' of the system before you could jump?
Sort of make systems the intersections of interstellar travel. This would make mining stations far more vulnerable, as the enemy fleet wouldn't be skipping them anymore, and means science & colony ships still travel faster than fleets with cruisers & battleships, so fleets travel slower than civ ships.

PS: This is all assuming FTL is 'forced' into hyperlane only, but maybe with slight variations on windup/travel/cooldown speeds for different techs.

 

[Drowe]

But generally speaking, I agree that a new mechanic like this would only deal with a part of the issue that is warfare, so only a combination of multiple possible changes to the game can really improve it.

That's the general consensus of this thread too.

That's the thing: I decided against the idea of supplies because that makes less sense to me than communications.
Maintaing communications throughout the empire are a base necessity - if you're cut off from everybody else, and are blind and deaf, you can't fight, let alone win.

If you read a few novels in settings that don't have FTL communication, you will see that is not exactly true. Or look at our own history, the British Empire fought wars in India when the fastest way to get new orders were messengers that took months to arrive and they did just fine. Fast lines of communication can actually be harmful too, because it encourages HQ to micromanage operations, which is terrible. Even if fast lines of communication are cut, you still have courier ships to send back and forth with updates, it's slower, true but on the scale of things not terrible.

Don't get me wrong, I like the idea of it, but I don't see how it could be practically implemented. You would either have to improve the AI by a lot to act more intelligently and don't allow for control of fleets out of communication range, or allow you to control your fleet but with a time delay on both, sensor data and responsiveness. So it would most likely turn out to be a less interesting mechanic.

 

[DavetheGreat]

Awesome summary Legendsmith

 

[JulienJaden]

Don't get me wrong, I like the idea of it, but I don't see how it could be practically implemented. You would either have to improve the AI by a lot to act more intelligently and don't allow for control of fleets out of communication range, or allow you to control your fleet but with a time delay on both, sensor data and responsiveness. So it would most likely turn out to be a less interesting mechanic.

Agreed - the bottomline is that whether it would be fun or not hinges on the execution and AI implementation.
That said, the game would benefit from smarter AI regardless, so here's hoping that, even if they never implement a supply/communication mechanic, they will work on the AI some more.
If the AI doesn't doomstack anymore but spreads out its forces to hit you everywhere at once, that could already be a step in the right direction.

 

[Zer0k]

I think that's frustrating gameplay. if a fortress is a giant chunk of HP that you need to just wait for your fleet to whittle down, people are going to stack as much as possible on it for it to have some effect, and that HP would have to massively increase for it to be able to have meaningful delays, even with slower FTL (Because the smaller fleets it is delaying for will have slower FTL too). If it's low firepower and high HP not much interesting is going on.

Fair point, I think from a mechanical point of view it would work, but it wouldn't be fun (which is the end goal of this discussion).

This is one of the reasons I suggested system wide auras, because it provides a way for defences to slow down the enemy while actually providing some interesting strategic choices, AND without requiring a huge fleet to kill it. Big tanky forts will just encourage large fleets that can kill them as fast as possible. Forts do need a firepower/survivability buff, but it shouldn't be all.

+1 on this idea, regardless of the doomstack issue.

What if you had to move to the relevant 'side' of the system before you could jump?
Sort of make systems the intersections of interstellar travel. This would make mining stations far more vulnerable, as the enemy fleet wouldn't be skipping them anymore, and means science & colony ships still travel faster than fleets with cruisers & battleships, so fleets travel slower than civ ships.

Well, I recall Wiz hinting (could be wrong though) on fixing the issue of mining stations (I think by simply removing the weapon slots), so you can only destroy them if you tell the fleet to attack (even with aggressive turned on).

PS: This is all assuming FTL is 'forced' into hyperlane only, but maybe with slight variations on windup/travel/cooldown speeds for different techs.

You don't need to force it to hyperlane, I would argue only wormholes and jump drives don't need the directionality. But for warp, you can play with multiple things, heck you I would argue you don't even need to slow anything down. Merely forcing the fleet to move to the other side is enough to "slow" FTL.

However, I feel a problem along the way. It might be difficult to "retain" the perceived advantages of each FTL type when you slow things down, that assuming PDX doesn't restructure everything.