Does AI strength to AIP need adjustment? (aka, is AIP too restrictive?)

[chemical_art]

As long as they were truly CPA's, so the attacks came from already existing ships, I wouldn't mind that much.

Presumably it would use the normal CPA logic: pulls half the target amount from the strategic reserve, then pulls from barracks/guard-ships/etc of the target tech level, then it pulls from other tech levels, and if it's totally out it pulls the rest from the strategic reserve (and if that runs out, it just sends an understrength CPA).

So I guess you could put a dent in the strategic reserve just before a HW assualt with this, though you'd be paying a lot of AIP to do it.

I am of fan of what may seem to be a player nerf if it isn't a zero sum game: There are upsides.

It would also help clear out reinforcements, which I like.

[keith.lamothe]

As for how they enable low AIP game play, I think this might be fixed by changing something else to ignore negative AIP.  Right now the AI gets unlocks based on AIP gained, and ignores negative AIP.  What if CPAs ignored negative AIP?  You might be at floor, but those CPAs are still going to hurt.

Normal CPAs already have a time-based alternative-minimum-AIP, which I think largely achieves what you're suggesting would.

Anyway, I'm not fussed about Data Centers as they are, and if them triggering a one-time counter-CPA is a problem for folks I certainly don't have to do it.  I think having it be tied to advancement towards the AI HWs (i.e. taking out AI command stations nearer the AI than you) makes more sense anyway.  Though we don't have to do that either.

[Winge]

Rather than have DCs trigger counter-CPAs, I've been considering (all this only on if the Lazy toggle is off) :

1) whenever an AI command station (normal, not home) dies, it checks the distance of that planet to the nearest human home command station and to the nearest ai home command station
2) if distance-to-human is less than or equal to distance-to-ai, then nothing extra happens
3) else, trigger a 30-second-countdown counter-CPA based not on AIP but on proximity to the AI command station, capping out at as-if-AIP-were-300 if you take out a command station directly adjacent to an AI HW.  For the other planets perhaps it's 300-(300*(distance-to-ai/distance-to-human)), so if distance to AI is 2 and distance to human is 8, the counter-CPA would be as if it were a normal CPA at 225 AIP.

Basically just playing off the ideas some have had about making more of an obvious gradation when moving closer to the AI HWs.  Though I suspect this new idea will not be popular

Hmm...I wouldn't mind it on normal maps if the game shows you in advance whether you will get a CPA or not.  But I also think 'special maps' will be problematic.  I can just imagine the pain on a Snake map...  Other games, like Faulty's recent AAR might not even see a CPA (his Home Command Station was located so that he was within 4 hops of every planet...impossible to deepstrike, lol).  Maybe make it the first x planets taken within deepstrike range?

I think the Lazy AI option is the better one overall, and will be easier to balance.  Of course, that is entirely my opinion.

[keith.lamothe]

Hmm...I wouldn't mind it on normal maps if the game shows you in advance whether you will get a CPA or not.

Hmm, yea, I suppose some kind of indicator would be needed.  Though if you were only off by a little the attack probably wouldn't hurt all that much.

But I also think 'special maps' will be problematic.  I can just imagine the pain on a Snake map...

Well, there it'd probably actually be pretty easy, as:
1) during the early phase everything is closer to you than the first AI HW
2) even after that, it will take a while to get close enough where the CPA is actually much harder than a normal one would be, and you have a dead-simple way to single-chokepoint-defend it
3) past a certain point you'll have had to take enough planets to avoid deepstriking that you're probably only getting counters that are less intense than a normal CPA would be.

Of course, just the getting one at every planet at that point could be troublesome.

Other games, like Faulty's recent AAR might not even see a CPA (his Home Command Station was located so that he was within 4 hops of every planet...impossible to deepstrike, lol).

Those maps have their own challenges

I think the Lazy AI option is the better one overall, and will be easier to balance.  Of course, that is entirely my opinion.

If I did this it would be part of a non-Lazy AI, and not part of a Lazy one.

[chemical_art]

As a side benefit, the changes to strategic reserves will help prevent AS cheese.

[keith.lamothe]

Ok, the following changes are now in for the next release:

* Added a "Lazy AI" toggle to the AI Modifiers section of the lobby.
** If this is on (off by default), the AI will not care enough to be bothered with the following (newly implemented) self-preservation measures:
*** Maintaining a powerful strategic reserve to defend the AI homeworlds, even when it does not feel threatened.  Instead, its strategic reserve will be proportional to AIP (capping out around 200 AIP).
*** Increasing the AIP floor by 10 for every homeworld-guarding Core Guard Post that is killed.
** Decreased AIP-on-death of AI Home Command Stations from 100 to 20.
*** This applies to Lazy-AI games too, but is intended to counterbalance the increase the core posts give to the floor on Non-Lazy.
** The overall point of these changes is actually to make AIP feel less restraining: if you're playing a Non-Lazy AI you're going to need a certain amount of power to take down the homeworlds due to the strategic reserve, and if the core posts are going to bump up your AIP-floor anyway you may as well be above the floor before that happens.  At the same time, it's not actually forcing you "off the floor" until that final assault on the HWs, so it's up to you whether you want to start ramping up earlier or "just-in-time".  Of course, on high difficulty levels you may not be able to survive the escalation, but that's kind of the point of the high difficulty levels

So I didn't do anything with counter-CPAs (either from DCs or command stations).  Or harvesters.  In the end I think these changes are pretty intense as-is even though you don't actually encounter them until you attack the HWs.  If it looks like more will make sense, then we can do that later.

[Cinth]

As a side benefit, the changes to strategic reserves will help prevent AS cheese.

HW nemesis fleet doesn't do this well enough already? 

[keith.lamothe]

As a side benefit, the changes to strategic reserves will help prevent AS cheese.

HW nemesis fleet doesn't do this well enough already?

He may never have had the pleasure of sailing 8 BB champs to an AI HW at AIP 10.

[Cinth]

As a side benefit, the changes to strategic reserves will help prevent AS cheese.

HW nemesis fleet doesn't do this well enough already?

He may never have had the pleasure of sailing 8 BB champs to an AI HW at AIP 10.

Probably not    It is a suicide mission to do that.  I would think that even a single BB would produce quite the fleet (comparatively).

[keith.lamothe]

For later, but wanted to raise the question now, and related to two things we've discussed here:

One of the things I'm wanting to do leading up to the next expansion is to replace the normal (non-home) AI command stations with a variety of types randomly selected during mapgen.  Not entirely dissimilar to how during the run-up to 4.0 guard posts were changed from always-the-same-thing to the different kinds we have now, but with a focus on utility and support functions rather than just a ton of tank or dps (if you want that, the Fortress King already awaits you).  The expansion would add new station types, but the base game would get a moderate selection too.

One such kind could be a "surveillance command station" or something like that, which is only ever seeded if Lazy-AI is off, and:
- Any planet closer to it than any other surveillance command station is considered "assigned" to that station.  This assignment is made during mapgen and doesn't change if any surveillance stations are killed or whatever.
- That "if Data Center is killed, launch counter-CPA with strength proportional to total number of times this rule has been triggered" rule would be in effect, but only if its planet's assigned surveillance station was still active.
- That "if Command Station is killed and nearer the AI HWs than the Human HWs, lauch counter-CPA with strength proportional to how near it is to the AI HWs" rule would be in effect, but only if its planet's assigned surveillance station was still active.
- There'd be a new form of hacking whereby a surveillance command station could be rendered "inactive" (for the purposes of the above rules) without destroying it and thus without incurring AIP.
- Just to be clear: destroying a surveillance station would render it pretty thoroughly "inactive"

Anyway, this would give the AI some kind of sensible response to you blowing up important stuff, and you'd have four ways of dealing with that:

1) Just eat the counter-CPAs, which shouldn't be terribly hard unless you're playing low-AIP (and it's fine if you want to do that, but this particular option may not be pleasant).

2) Just kill the surveillance stations of any area you want to avoid the counter-CPAs; costs AIP but if the surveillance station(s) in question are on planets you already want, well, there you go.

3) Hack the surveillance stations of any area you want to avoid the counter-CPAs; ramps up hacking antagonism (making it harder to hack for K or Bonus picks or whatever later) and may be a hard fight depending, but no AIP and no CPA.

4) Just turn on "Lazy AI" in the lobby   Perhaps for added challenge you could hang a "Kick Me" sign on the AI Home Command Station?


Thoughts?

[chemical_art]

Sounds cool but beware too many interplanatary effects. Otherwise youv step on ai types turf.

[Winge]

It does sound interesting, and it makes it a bit easier to see whether you are in FUN! range or not.  A few recommendations:
1.  Similar to how "External Invincibility (x i) is listed, CCs under surveillance or other CC effects should indicate both the effect and the planet of the station.  The exception would be if Extra Fog of War was turned on, in which case you wouldn't know the location CC unless you scouted it.
2.  For the hacking, I would rename the Ship-Design Hacker to AI Structure Hacker, and indicate what AI structures it can hack (at this time, ARS and special CCs).  That should keep the 'extra stuff' to add to a minimum.
3.  Make the stations immediately adjacent to the Human Home be unable to trigger the effect, to try and keep those planets 'simple'.
4.  For the reason chemical_art mentioned, limit the number of "Special Command Centers" that can seed based on the number of planets.  Naturally, defensive and turtle AIs get more.  Possibly add a new defensive, non-turtle AI type at a later time (expansion material?) that gets a multiplier to the number of Special Command Centers it can seed, or even special "Backup Special Command Centers" (forcing you to kill or hack a Second Command Center if you want to avoid the effects).

Some other CC ideas (all using the same Data Center and Command Station conditions and counters to determine their strength):
1.  Missile Silo Command Center.  Launches multiple warheads (Lightning and EMP, much like the NRC) that immediately head for Human worlds.  Relative difficulty:  easy.
2.  Warp Counterattack Command Center.  Launches a counterattack on a random human world (similar to Warp Counterattack Post).  Relative difficulty:  easy-moderate.
3.  Enraging Command Center.  The next few waves from the AIs gain an AIP bonus.  The number of waves and AIP bonus both increase with the counter.  Cannot be seeded by turtle types or if No Waves is selected.  Relative difficulty:  variable.
4.  Raid Command Center.  Also known as  >D.  Launches waves at the planet in question.  Wave strength and number of waves is based on the counter.  The waves have the same wave time and cooldown as a Raid Engine (60 second warning, 4 min between waves).  Relative difficulty:  FUN!
5.  Rally Command Center.  Heavily reinforces every planet within its range.  Relative difficulty:  variable.
6.  Special Forces Command Center.  Immediately reinforces the Special Forces to the current cap plus an amount based on the counter.  The boost would be temporary but, much like Champion Nemesis units, you would need to kill the Special Forces back to the lower cap.  Relative difficulty:  moderate.

Obviously, this is a very rough sketch of ideas--balance would probably be very difficult for some of these (not number 4, of course...that one is clearly perfectly balanced and fair).  Just trying to help turn some creative wheels here .