A Paradox Employee Contacted Me Regarding Current Backlash on Forum

  • 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.
but a smart enemy can lure you
Have you seen the stellaris AI?

Any solution to doomstacking that involves additional logic will be DOA (Dead on Arrival) without an AI rewrite.
Lets be real, if it's not happened in the last two years, it's probably not happening in the next two either.​
The "Real" solution to doomstacking is to remove player agency (Reducing agency NEVER leads to good gameplay, so I dont endorse this, it's just a concrete solution). Make it so that fleets are guided to targets by running SOTS2/EU4 style "missions" or "sorties", rather than players stacking things up.
  • exploration sorties (mark clusters to scan)
  • raid/skirmish sorties (mark planets to abduct, undefended systems to cripple, on a per system or cluster-wide basis)
  • patrol sorties (mark out a star cluster to patrol)
  • invasion sorties (mark systems or planets to invade, allocate armies to the plan, armies go off and invade them as needed)
  • attack sorties (mark a cluster, star or several stars within a cluster to attack)
  • defend sorties (mark a cluster, star or several stars within a cluster to defend)
Whatever you want.
A system like that would be more useful for the AI as you could tell Allies "Go do X Sortie on Y star/Cluster" and it creates more hooks for the AI to work off and a way for it to "lock in" orders and form larger battle plans. (rather than the weird system-weights for targets they use currently, seemingly acting more like a chaos engine than a battle AI).
It'd also remove the players "unfair" advantage of microing fleets to either stack them up or dogpile them on the enemy. As fleets would be largely AI guided so the AI would be managing ships and would fan out as needed.

It'd also be frustrating as hell when it glitches out and your ships get caught out-of-place with no way to control it.
 
Last edited:
  • 7
  • 3Like
  • 1
Reactions:
Have you seen the stellaris AI?

Any solution to doomstacking that involves additional logic will be DOA (Dead on Arrival) without an AI rewrite.
Lets be real, if it's not happened in the last two years, it's probably not happening in the next two either.​
The "Real" solution to doomstacking is to remove player agency (Reducing agency NEVER leads to good gameplay, so I dont endorse this, it's just a concrete solution). Make it so that fleets are guided to targets by running SOTS2/EU4 style "missions" or "sorties", rather than players stacking things up.
  • exploration sorties (mark clusters to scan)
  • raid/skirmish sorties (mark planets to abduct, undefended systems to cripple, on a per system or cluster-wide basis)
  • patrol sorties (mark out a star cluster to patrol)
  • invasion sorties (mark systems or planets to invade, allocate armies to the plan, armies go off and invade them as needed)
  • attack sorties (mark a cluster, star or several stars within a cluster to attack)
  • defend sorties (mark a cluster, star or several stars within a cluster to defend)
Whatever you want.
A system like that would be more useful for the AI as you could tell Allies "Go do X Sortie on Y star/Cluster" and it creates more hooks for the AI to work off and a way for it to "lock in" orders and form larger battle plans. (rather than the weird system-weights for targets they use currently, seemingly acting more like a chaos engine than a battle AI).
It'd also remove the players "unfair" advantage of microing fleets to either stack them up or dogpile them on the enemy. As fleets would be largely AI guided so the AI would be managing ships and would fan out as needed.

It'd also be frustrating as hell when it glitches out and your ships get caught out-of-place with no way to control it.

To be honest, that would actually be pretty cool. I like it.
 
  • 3
Reactions:
To be honest, that would actually be pretty cool. I like it.
Same. To my mind this idea of giving orders for specific clusters would naturally lead your design to making clusters - and by extension sectors - being created in a way that actually uses a galactic geography and theryby also solving the sector creation problem.
 
  • 1Like
Reactions:
The solution to doomstacking should be:

- The larger the ship, the slower it travels, and the difference is significant
- Larger ships are significantly more powerful (to prevent dominating with Corvette swarms)
- Space travel is slower in general

That way, you're forced to have multiple fleets. If you have a Doomstack, you'll win any individual engagement, but a smart enemy can lure you to one side of your empire while attacking another.

With your way you're just multiplying the issue by two. You propose one stack of fast corvettes to lure enemy out of position (harassing) and one of powerful battleships to (deliver pain).
To which the natural answer is fielding 2 stacks: one of fast corvettes to counter corvettes stack and one of battleships to counter the battleships one.

More or less what happens now... to a certain extent. Since the battleship stack is the only true valuable asset in a war (the only thing that requires half a game or more to rebuild if gone) one can simply ignore corvettes harassing here and there for little to no effective damage. They can occupy some fringe sector but they can't get past planetary FTL magnets so in most of the cases the harassing corvette stack only practical effect is some noise spam in the alert menu.

I'd say instead that ship class should not influence speed. Instead there should be other modifiers to promote fleet diversity (anti piracy value and evasion were nice but failed attempts). Using speed only suggests to the player to not mix ship classes at all.
 
  • 2
  • 1Like
Reactions:
Have you seen the stellaris AI?

Any solution to doomstacking that involves additional logic will be DOA (Dead on Arrival) without an AI rewrite.
Lets be real, if it's not happened in the last two years, it's probably not happening in the next two either.​
The "Real" solution to doomstacking is to remove player agency (Reducing agency NEVER leads to good gameplay, so I dont endorse this, it's just a concrete solution). Make it so that fleets are guided to targets by running SOTS2/EU4 style "missions" or "sorties", rather than players stacking things up.
  • exploration sorties (mark clusters to scan)
  • raid/skirmish sorties (mark planets to abduct, undefended systems to cripple, on a per system or cluster-wide basis)
  • patrol sorties (mark out a star cluster to patrol)
  • invasion sorties (mark systems or planets to invade, allocate armies to the plan, armies go off and invade them as needed)
  • attack sorties (mark a cluster, star or several stars within a cluster to attack)
  • defend sorties (mark a cluster, star or several stars within a cluster to defend)
Whatever you want.
A system like that would be more useful for the AI as you could tell Allies "Go do X Sortie on Y star/Cluster" and it creates more hooks for the AI to work off and a way for it to "lock in" orders and form larger battle plans. (rather than the weird system-weights for targets they use currently, seemingly acting more like a chaos engine than a battle AI).
It'd also remove the players "unfair" advantage of microing fleets to either stack them up or dogpile them on the enemy. As fleets would be largely AI guided so the AI would be managing ships and would fan out as needed.

It'd also be frustrating as hell when it glitches out and your ships get caught out-of-place with no way to control it.

Essentially this creates an auto battler system where you are picking the board. While it could get a bit screwy when you are you opponent pick different total areas with some overlap, the auto battler is a solid gameplay style that is also fairly popular at the moment. I would argue that auto battlers are designed to average not reduce player agency, so for some players they increase their skill, for others they reduce it.

For example, if you had to control each unit in a game of an auto battler, my friend who played starcraft at a master/grandmaster level would wipe the floor with me. By removing the micro player agency, and emphasizing the other aspects of play, I have a much better chance of winning.
 
Last edited:
  • 2
  • 1
Reactions:
Essentially this creates an auto battler system where you are picking the board. While it could get a bit screwy when you are you opponent pick different total areas with some overlap, the auto battler is a solid gameplay style that is also fairly popular at the moment. I would argue that auto battlers are designed to average not reduce player agency, so for some players they increase their skill, for others they reduce it.

For example, if you had to control each unit in a game of an auto battler, my friend who played starcraft at a master/grandmaster level would wipe the floor with me. By removing the micro player agency, and emphasizing the other aspects of play, I have a much better chance of winning.

It's more of a game scope matter. Do I want player to keep attention to macro or micro, economy or war, diplomacy or internal affair... how often, how much and so on.

RTS games traditionally have a strong micro component. Grand Strategy tend to abstract that away an let player focus on the bigger picture. Stellaris is an hybrid of the 2 genders and (imho) didn't find the right balance between these two souls.

Stuff like ship designer is clearly RTS. Create optimized design for specific tasks, put right weapon in right slot, make a fleet with x% class A and y% class B.
Battle mechanics are clearly grand strategy. Send fleet to system, battle and read recap (optionally: click retreat button... if you make in time).

There are two main things to consider when building that:
-How much player attention is requires: does player have enough time left to handle other things, or has to constantly "babysit" (micro) the mechanic?
-It is interesting (fun): there is agency in the process, does the process present viable alternatives, can player influence the end result.

Imho currently only the first requirement is met.
Fleet combat is already an auto battler. You have 0 agency on that. During a war you can just issue commands from galaxy view and wait for battle report while you work on something else (most likely pops an jobs :confused:).
Ship designer is instead very micro but is an asynchronous time sink. You don't constantly design ships during wars. You just occasionally update classes and that is. It's mostly there to give you something to do during quiet times.

What it fails is giving effective agency by presenting viable alternatives that can influence the end result. Unless you play for losing we all know by now you have to single stack all you fleets, prefer battleships over other classes and use whatever weapon load that patch is more effective. Auto follow the slowest fleet, move that fleet. Whoever has the biggest stack wins. Some "geography" configuration may delay the first/final showdown, especially in MP, in that case there will be a little bit of (arguably annoying) cat and mouse micro. But eventually it will come and the war is set.


Personally I wouldn't like micro intensive solutions. For that end Auto resolvers may look like a good solution. But actually they are not necessarily so. Because with auto resolvers a battle outcome is decided before battle starts. And if we pair that with a doomstack gameplay as it is now war itself is decided before it starts. And more or less that's what all people complain about.

That's why I still stick to my idea that maintenance cost per month is the right lever to use when paired with easier (cheaper, faster) ship production/replacement. It offers an interesting trade off between amassing the biggest stack you can to dominate an enemy in a battle and spreading your forces while keeping most of them anchored at home ports to not run out of supplies before war ends. It makes undesiderable for large empires to commit all they have against small targets allowing for more interesting wars, it naturally advantages defenders without boring and hard to balance static defenses (not to mention they are a big culprit of the hyperlane only rework). It also suggests to players a natural conclusion for the war (when their economies are getting impacted beyond repair or the next enemy prize it's just not worth the cost).

Plus it can be adjusted with all kind of modifiers (fleets in system, fleets in nearby systems, fleets engaged in combat/bombardment, enemy fleets in nearby systems, jumps away from border, buildings/space structures in range, traditions, policies, techs, etc.) that offer a number of alternative choices to the player and ways for developers to balance them. With right modifiers it should be possible to win wars even after losing some initial battles if that means that you were able to make the enemy consume to much supplies.
 
maintenance cost per month is the right lever to use when paired with easier (cheaper, faster) ship production/replacement.
You'd have to make going negative on upkeep actually mean something, The starving utopian empire and purge world strategies are extreme examples that show off how weak the effects of "hitting zero" on resources - or even stability - are in Stellaris.
And if you make the negatives too strong, the poorly designed AI's fleets and economies will death spiral even faster. And if you only make it effect players it'll just feel cheap.
Dammed if you do, Dammed if you dont,

it makes undesiderable for large empires to commit all they have against small targets
But the AI picks its targets via an absolute weight, it considers your(+allied, at a reduced weight) entire fleet vs its entire(+allied, at a reduced weight) fleet, and if it's Fleet power advantage is 1.X times greater, it'll go to war.
Committing more ships to a battle means I'm more likely to walk away with fewer losses, meaning even as a large empire it's in my interest to doomstack, as (usually) half the galaxy will be in a federation opposing me.
For this reason, travel-times arent an issue either, I'll just shuffle all my ships up to the jump point closest to their space and spill-over at the start of a war, why waste time DOW and then mobilising when I can issue a DOW 1 hour before my fleet arrives in enemy space?

There are a lot of game systems that actually exacerbate doom stacking,
  1. how the AI evaluates enemies is one,
    • Realistically, PDX would need to go away and shatter the galaxy map up in to proper pre-made sectors or star-clusters which the AI can then hook in to and evaluate on a case-by-case basis
      • (OK the Imperium has 500 ships empire wide [assign N value]
      • BUT they only have 3 ships in our adjacent cluster assign [-N1 value],
      • and 40 ships in a cluster once-removed from us [assign -N2]
      • and our ally is also adjacent to this target_cluster with, say, 60 ships, assign value +N3)
      • If ΣN > X, declare war.
      • This would lead to small empires picking and choosing when to attack a far larger empire if that empire lacked decent defences on its fringe sectors.
      • (whilst the current version of the AI would only evaluate N:X, and if the player is bigger, they'll never attack, even if they are a fanatic purifier on GA or starnet (as even starnet is constrained by these limits, it just marginalises the power disparity)).
      • This could be calculated on a per-system basis, but that potentially leads to a lot of iteration, thread locks and performance issues, better to handle it on a "Aggregate Enemy Fleet power In Cluster N" basis.
  2. and player micro is another
    • hence my above post about fleet missions/sorties,
    • you'd set restrictions like you can't launch an attack/invasion mission from one of your/federal starbases unless war has been declared, giving both sides a little farer mobilisation time.
    • But could have special civics like criminal megacorps or despoilers that could issue "delayed CBs", at some higher cost, only notifying their enemy of war as their ships reach the border, for suprise attacks and raids.
      • And GC resolutions and sanctions that could build off this.
I could probably sit here all night and explain how dozens of game systems incentivise or reward doomstacking, because the game is fundamentally built around conquest or growth at any cost, with few to no counters to this
  • I personally still think Area of effect weapons, minefields and limited-range auras need to make a comeback as a means of "thinning the horde" and stopping stacking-up fleets from being so damn lucrative, it'd certainly be less work [as paradox actually had this in-game once, till they ripped it out in 2.0] than "sectorifying" how the AI handles the galaxy (I dont forsee any major AI adjustments coming before stellaris 2... probably in 2026 or something lol) or rewriting fleet movement from the ground up for a mission system (which would be as contentious to players, I think, as FTL changes were - though I do think that console players might actually benefit from a mission system, if it's well implemented, it might make for less "fatiguing" gameplay with a gamepad, I'm not a consoleplayer so thats just a guess, though).
 
  • 3
  • 2Like
Reactions:
In any case, these unknown, but long patch notes seemed to have slightly quieten the forum discontent for now (until the Bananoid portait DLC reveal, I guess, although Xenonion got there first).
 
  • 1
Reactions:
In any case, these unknown, but long patch notes seemed to have slightly quieten the forum discontent for now
To make conclusions about "unknown", but "long" notes is a bit too risky, don't you think ? Paradox has the tendency to tune its notes in such a way that they're just looking THAT impressive. For example, instead to write something like "Corrections in regards to various localisation-issues", that's ironically something in which Paradox becomes pretty detailed ( --> "long" notes ) in order to point out even the tiniest issues they've corrected. Paradox has also the tendency to claim that something is completed, fixed, optimised or / and balanced ( in its notes ) before it turns later out ( by players in actual playthroughs ) that it's still a thing.

And don't make the usual mistake that interprets "quietness" as nothing else but satisfaction.
 
Last edited:
  • 14
  • 1Like
Reactions:
In any case, these unknown, but long patch notes seemed to have slightly quieten the forum discontent for now (until the Bananoid portait DLC reveal, I guess, although Xenonion got there first).

There is no reason to not be furious about the state of the game. Having less players voice their opinion is a sign of players quitting. Me and my big playgroup haven't touched the game since 2.6 when we realized the devs don't actually care about fixing nearly 2 year old issues like the Crisis. We are not going to return unless the base game is fixed. Nothing else will make us come back.

Remember guys, 2.6 was the patch where they worked on the Crisis! The number 1 feature thats broken since 2.2. And they chose to NOT fix the Crisis being unable to conquer the galaxy and expand. This is the state the game is in. Its obvious. They keep ignoring feedback which has been coming in since Megacorp release. They don't care. The game is dead and they will simply move to other, more profitable projects. Because they are unable or unwilling to fix it.
 
Last edited:
  • 9
  • 1Like
Reactions: