Core Guard Post Proposal

[TechSY730]

I don't tend to end up stagnated anymore but that's because I had to modify the playstyle to meet the game.  My first 10-15 games or so ended up stagnated (at AI 6!) because I'd take too long, or take too much.

I realize the concern, but really, you need to be more aggressive to avoid the problem.  WRECK systems that are in reach.  Obliterate their ability to do more than barely reinforce alerted systems. This breaks the stonewall theory.  I even wreck the Jump Point for the AI for the whipping boy so that's not a problem either.

You can travel for four worlds from any neutral planet.  Use that.  Obliterate the enemy.  Lay waste to them before they build then use small patrols (Mobile Builders are great for this) to use redirection points (cloaking) for your MK I fleet to patrol and scour any alert worlds if you don't want to micro that part. 

That ends up leaving only coreworlds as a problem.  You are human, they are math.  Get angry. 

Hmm, true. I guess we not taking steps we should be. Using these sorts of things should help greatly.

But even this kind of "proactive" approach doesn't completely solve the issue.
One, the AI still builds up no matter how much you do. Sure, sometimes you can slow down how much they build up, or reduce how far they build up on a single planet, (both of which help a LOT) but eventually they will amass a force, if for nothing else then the slow but still there buildup on the command station from neutered roles, the build up of freed ships from retreated waves or ships that managed to escape your "wrath" when taking out a planet, CPAs, or other ways for aggressive ships to spawn (like exos).
Two, as you noted, the core worlds and homeworlds.
Three, if the AI grows so fast with AIP that you can't really take very much, then most of the content of the game (in any one game) is just sitting there useless because in many cases, you can't afford to take it. OK, so maybe its not so bad its most content missed in any one game, but the number of planets you can take before the AI starts forcing you to slow down no matter how good you are at preventing AI "walling" seems too low right now.
And four, the obvious style of play making a "stalemate" situation makes it hard to tell when you have pretty much given up a good chance to win, even if you can prevent losing. As Valve pointed out, stalemates are BAD, as they are not only not fun, but you can't really tell it is one until you have played so long that you are now bored.

[Wanderer]

But even this kind of "proactive" approach doesn't completley solve the issue.

It's not meant to solve it, just mitigate it to remove the stagnation and only have 'wars', not constant fleet drains as you get there.

One, the AI still builds up no matter how much you do. Sure, sometimes you can slow down how much they build up, or reduce how far they build up on a single planet, (both of which help a LOT) but eventually they will amass a force, if for nothing else then the slow but still there buildup on the command station from neutered roles, the build up of freed ships from retreated waves or ships that managed to escape your "wrath" when taking out a planet, CPAs, or other ways for aggressive ships to spawn (like exos).

My comment about using MK I fleets for scouring duty will remove (in most scenarios) the light command station build up that occurs per 'pass' of the patrol fleet.  Sometimes you just take your main fleet ball and do a clean-up job.

Wave retreats are a problem, but that's why I FF the heck out of the entry wormholes for the whipping boys.  If they can't leave, one of us has to die.

Freed ships usually attack something.  Either an outpost somewhere (ripping down your lovely Fact IV) or they head for the homeworlds, and should die to your defenses.  If they're camping send your patrol fleet or the fleet ball to deal with it and clear it before it gets to extreme levels.

CPAs are massive events and are a scenario to discuss in their own right.  I don't consider them 'hour to hour' activities.

Two, as you noted, the core worlds and homeworlds.

These don't build as much as you'd usually think, and at the worse are a massive fleet-war to get through.  It's not a stagnation component though if you DON'T sit on neighboring wormholes.  Raider mentality on a massively built coreworld is your friend.  Hit them, HARD.  Then run.  At least 2 worlds away so alert clears.  Deal with fallout.  Rinse/Repeat.

And finally, three, if the AI grows so fast with AIP that you can't really take very much, then most of the content of the game (in any one game) is just sitting there useless because in many cases, you can't afford to take it. (OK, so maybe its not so bad its most content missed, but the number of planets you can take before the AI starts forcing you to slow down no matter how good you are at preventing AI "walling" seems too low right now)

I agree, but that's an attraction to me.  One of the strategical components of this game is purposely NOT being able to be a completionist and having to choose your targets.  Using satellite systems as waypoints because the AIP gain is too high to take the 'highway' route.  The AI doesn't wall too badly, really, if you keep up on your maintenance.  I guess that's part of my point.  If you ignore the buildup, you end up against the 'wall'.  If you maintain it in some way (fleet resources, using fleet-ball time, whatever), it shouldn't build up past your ability to handle unless your AIP is WAAAY too high.  Well, or if you've got the AI cranked to a higher level then you're able to handle.

Try an experiment I did that tought me some of these components.  Crank up one AI to 9, and drop the other to 2.  Use the exact same AI for both (I chose vanilla).  It helped me learn what you can and can't get away with in regards to left-over buildups, particularly in regards to AIP.  I literally could ignore the AI 2 worlds, but the AI 9 worlds I was doing constant sweeps on reinforcements to clear out, to make sure I never dealt with getting 'blockaded'.

I agree that all the 'little' things you mention can add up quickly.  The idea is to never let them add up.

[Wanderer]

And four, the obvious style of play making a "stalemate" situation makes it hard to tell when you have pretty much given up a good chance to win, even if you can prevent losing. As Valve pointed out, stalemates are BAD, as they are not only not fun, but you can't really tell it is one until you have played so long that you are now bored.

I missed this in my first pass of your comments, my guess is we just crossed posts.

I agree, true stalemates are bad, but I've used them as learning experiences.  What works, what doesn't.  What's cheesy and what isn't.    How do you bait the enemy.  How do you free the wall so you can engage it on your own terms, in your own turret balls (Hint: Raid Starships).  What is 'unbreakable' and what's just HUGE.

I won't say I don't stalemate anymore, but the chances of it are incredibly unlikely.  For example, if you've got that one world you just can't break... don't.  Raid the cmd center down and eat the backwash on a home planet with turrets and bonuses, then come back in with all your might and make sure it's buddies don't build up off the new alert.

*shrugs*  It might just be a way of looking at the game itself.  I found it to be a part of the tactical game itself and learned ways to deal with it.

[keith.lamothe]

Hmm, kinda going in a bunch of directions here, but the "stalemate" thing is worth discussion:

1) The "stalemate" outcome is an intentional part of the original intent of the game design, and in fact used to be considerably more pronounced in older versions: one of the ways to lose in AIW is to be unable to make progress.

2) That said, I don't think it's much fun either.  I don't mind can't-win scenarios as long as there's some satisfying way to measure how "well" I lost; like a pre/early WW2 scenario as Checkoslovakia or Poland: presumably if it were modeled correctly I couldn't come anywhere near near stopping Germany, but if I mixed the right concoction of audacity and natural-20s I might have the satisfaction of the dying rabbit seeing the look on the tiger's face as it realizes it no longer has any intact hamstrings and there are other predators out tonight.  But just being ground into the ground by inexorably growing AI forces doesn't have that feel.

3) That said, radically changing the game to remove the stalemate possibility is likely to have a net-negative impact on fun for the audience as a whole.  It's easy to not realize the importance of a certain part of the game being a certain way until things start flying around due to a central change.


But I do recall some ideas I tossed around a couple years ago about a "Showdown" AI plot where basically some trigger would cause the AI to just go all-out offensive and if you survived you could win without much further difficulty.  The idea was well received but we just never got around to doing it, and I wasn't ever all that clear on how to best do it anyway.  One general problem is that it's basically making the AI do something really stupid: abandon its well-entrenched position.  Stalemates win a lot of games for the AI, why would it stop?  Because it's not the DM

So let me throw out an idea; this is just off-the-cuff and might be an impressively bad idea:

a) Add a new kind of hacking that takes place on an AI Homeworld (specifically, on the same planet as a living AI Home Command Station).
b) Doing this hack takes 2 minutes, during which the AI spawns some moderate retaliation but mostly just as a "does the human have enough forces to even consider this" check.
c) Once the hack is complete, the virus or whatever makes the AI "do something really stupid"   Specifically:
- All reinforcements stop for 30 minutes (I'm thinking for both AIs, otherwise the effect is kind of diluted).
- 30 minutes worth of AI reinforcements get spawned all at once as threat, from both homeworlds and potentially a few different places (maybe even some spawning like a counterattack-guardpost does).  This will probably involve some exo-ish composition rules and/or wild-rolls to condense the available strength into not-too-many-individual-units to avoid this being a showdown between the AI and your CPU/RAM.
- During the 30 minute "there aren't any reinforcements" phase, the "wait at wormholes" behavior is also suppressed, so it won't just pile up into a bigger wall (and will "flush" existing threat-walls).  If that proves troublesome to implement (I don't think it will, but cannot be sure), just make the superspawns all zombies.

And in theory that would take care of it: if you die, well, the game's at least come to a conclusion.  If you survive you have a golden window to smack the AI around.  Of course, the response is probably "with what?!", which may indicate that the "30 minutes" needs to be longer.

Thoughts?  I realize this doesn't address the overall mid/mid-late stalemate potential before homeworld assault becomes possible, but one step at a time

[Hearteater]

I like it as an alternate attack method.  If you do go with 30 minutes only of no reinforcements, maybe have Force Fields in the home system turn off, or at least shut down Fortresses.  I'm guessing Raid Engines and CPA posts would need to be off once the hack is complete.  But that would be a good method of getting around an otherwise tough home world layout.

On the stalemate: One option would be an AIP over time increase once the first core guard post dies until the home command station is killed.  Not saying I like the idea, but some kind of build-up that keeps getting worse and worse once you start on the home world could force an eventual loss.

[Diazo]

An interesting proposal.

After 2 seconds thought, I'm wondering if this is misplaced.

I've gotten stalemated a lot more often in the mid-game then I have in the end-game while trying to attack the homeworlds.

Having said that, I have warheads for breaking me out of stalemates in the mid game, those don't work so well against homeworlds, especially the emp immunity.

Maybe some sort of funky warhead specialised for attacking homeworlds instead of a total re-work of the current mechanics?

D.

[TechSY730]

An interesting proposal.

After 2 seconds thought, I'm wondering if this is misplaced.

I've gotten stalemated a lot more often in the mid-game then I have in the end-game while trying to attack the homeworlds.

Having said that, I have warheads for breaking me out of stalemates in the mid game, those don't work so well against homeworlds, especially the emp immunity.

Maybe some sort of funky warhead specialised for attacking homeworlds instead of a total re-work of the current mechanics?

D.

Ah, warheads, a great, if extreme, mid game "wall breaker"

Multiple times I have used a well place lightning warhead or even an EMP warhead to weaken a well fortified planet long enough to actually be able accomplish my objectives on it.

But yea, mid game stalemates or even "grind" is, IMO, a bigger problem.

[keith.lamothe]

On the stalemate: One option would be an AIP over time increase once the first core guard post dies until the home command station is killed.  Not saying I like the idea, but some kind of build-up that keeps getting worse and worse once you start on the home world could force an eventual loss.

I don't think the rage is from "true" stalemates where there will never be a resolution, but rather because victory is impossible and defeat is going to take longer than, say, an hour or two to happen.  AIP-over-time from an event (like core guard post destruction) would need to be really fast to avoid that, and that feels a lot like railroading the player.  Some games can do that, and some "extracurricular" threats in AIW can do that, but the core game has to never do that.  One of the immutable parts of the design is that, in the core game, the player always controls any large changes in the pacing.

I've gotten stalemated a lot more often in the mid-game then I have in the end-game while trying to attack the homeworlds.

Having said that, I have warheads for breaking me out of stalemates in the mid game, those don't work so well against homeworlds, especially the emp immunity.

In mid-game I think there are two scenarios:

1) It's basically the core game, no game-changing threats, so the ball is supposed to be in your court.
2) You've got game-changing threats like hybrids, exos, etc and there's probably not a lot of stalemating going on.  Of course, exos can set up a sort of overall stalemate where you're stuck in a loop of defend-rebuild-defend, so maybe these don't change that as much as I think.

Maybe some sort of funky warhead specialised for attacking homeworlds instead of a total re-work of the current mechanics?

You mean like a warhead that suppresses reinforcements for 30 minutes and makes the AI put on blue face paint and charge you with a huge mass of ships?

[Volatar]

I think the direction to go is to take Keith's idea and apply it to the midgame somehow.

[Diazo]

Hmmm.

What about a "modified nuke", tags all units on one, adjacent, and all systems (same as the nuke progression) that frees them on to threat? With appropriate AIP, can't do this free of charge of course.

D.

[keith.lamothe]

I think the direction to go is to take Keith's idea and apply it to the midgame somehow.

Hmm, so basically "wave the red cape at the bull" to "shake things loose" and either the AI busts you up badly or you have a window in which to make progress without much resistance?  As long as it doesn't just let you beeline the AI homeworlds, I could see that working.

*Deploys hacking ships to write trolling posts on the AI's internal bulletin board

What about a "modified nuke", tags all units on one, adjacent, and all systems (same as the nuke progression) that frees them on to threat? With appropriate AIP, can't do this free of charge of course.

For one system, I thought the Spirecraft Attritioner already existed   More seriously, just bring something in to shoot at each of the guard posts (or the command station, carefully to not kill it) and that should free most/all of it.

For multi-system... just freeing them would probably cause carrier pileup, but I suppose that's ok.  In general I'm not sure the AIP-just-to-free-stuff option would really appeal to a lot of players, though.

[Hearteater]

I don't think the rage is from "true" stalemates where there will never be a resolution, but rather because victory is impossible and defeat is going to take longer than, say, an hour or two to happen.  AIP-over-time from an event (like core guard post destruction) would need to be really fast to avoid that, and that feels a lot like railroading the player.  Some games can do that, and some "extracurricular" threats in AIW can do that, but the core game has to never do that.  One of the immutable parts of the design is that, in the core game, the player always controls any large changes in the pacing.

Yeah, that's true it would do nothing to end the game "quickly" and I certain wouldn't want to see something that ramped up to finish me of in 60 minutes from when I first engaged a home world.

[Wanderer]

Just a side note: I'm still here and watching this conversation avidly, but I'm not involving myself because I honestly don't see this as a significant problem.  It's one of the ways to lose and it's something to learn to watch out for.  My personal indicator is if I'm on a churn and burn with no movement on the whipping boy(s), where it's an econ attrition war and my fleet + turret vs. waves are basically stagnating.  If that's happened, I know a CPA or other galactic event will end the game.  Stalemate.  These are usually the only stalemates I run into often.

However, afaik using an EMP 1 will fire up an entire planet after you, for a mere 2 AIP, so that may be an option already existing in game.  For multi-system 'annoyance' try the following combination: Cloaker starships, 2x transports, Raid SSs, A handful of engineers for repairs.  "Buzz" the guardposts in question, just get barely enough in range to activate the post, then run like heck back to your cloakers.  Rinse repeat, until you've emptied a system.  Nerf at whim, move to next system.

[chemical_art]

I somewhat like the hack idea...but...

in my experience, if my fleet ball can survive two minutes, it can survive long enough to blow up one guard post. If the fleet can blow up one guard post, even if fleet is lost (not starships, fleetships) I still come out ahead, because further reinforcements are reduced. Next fleet ball 90 minutes later can smash another post. Etc, etc. Game over.

Unless said hacking is cloaked (or can be cloaked) then you can at least have the option of a matyr fleet (not a fleet of matyrs  ) distract for 45 seconds, and the station and cloaker run out of the fire. But otherwise, it still runs into the problem of "AI has several hundred V's on this planet. Fleet eliminated in 90 seconds if it doesn't flee."

It just seems like it defeats the purpose. If my fleet can survive to hack, it can survive to blow up a post. So why risk a short term attack when I can play it safe? Unless said hack can be done on a core world, or the hack hinders for more period of a time...

<another problem with the hacking mechanic is that right when you put out the fires of the counter attack like mechanics, and rebuild the fleet lost holding the worlds (assuming  you have much of a fleet left after the hack)  the effect is already over.>

Or...

lose the counter attack post mechanic. Make the ai behavior aggression permanent, but make it so that every 30 or 45 minutes another reinforcement "burst" that also increase 10% from the normal amount the current AIP would spawn. Player can now attack worlds rapidly, but the timer is ticking to defeat AI before doom comes. A blitz of taking worlds is even more short term solution, because the "base" spawns are increasing and the multiplier is increasing too.

So it would be like

x * <1 + (y /10)>
x = reinforcements for period using aip, etc
y = number of burst, including this burst.

With the effect permanent, it is worth justifing amassing the whole fleet to buy time for the hack, including lightning warheads, because if the whole fleet is lost you still permanently altered the course of the game.

[Hearteater]

I agree with chemical_art's assessment on the 2 minute hacking.  Maybe have the hacking device hack any NON-homeworld AI system which "charges it" and then you send it in as a bomb to the home world.  This procedure could be made a little escort-y (if desired) by preventing the hacking device from going in a transport.