AI Homeworld Defense

[Draco18s]

It doesnt really change anything though for thre better. It increases the fact that the AI HW is a strength check, and that overlaping fields of fire results in nightmares

Do blob-on-blob fights result in nightmares?

[Diazo]

Okay, wow.

I think we need to back up a step.

What is the issue with the AI HWs we are trying to solve?

The last page of posts is pretty much people talking past each other and we are losing the thread here.

As I see it:

1) AI HW's are too grindy and throwing a fleet at the problem and then netflixing for a couple hours while your fleet rebuilds is the simplest solution to the problem homeworlds currently pose. Not the design solution where you are supposed to raid the guard posts on the HW down, but the simple straightforward one.

2) HW difficulty varies too much depending on which guard posts spawn, and where those guard posts spawn in relation to each other. A Wrath Lance and Grav Drone guard post covering each other is magnitudes harder then a Core Raid and a Core CPA post on opposite sides of the system.


Regardless of what happens, the 'combat' guard posts need to be brought into line with each other. I should not rather face 10 Needler guard posts then a single Wrath Lance post.

Then to solve 1), move all the posts next to the AI command station, rebalancing them as needed, so that the HW fight is only one or two battels.

To solve 2) was my suggestion of moving from 'brutal' and 'non-brutal' to 'combat' and 'support' picks for the Guard posts with all the combat guard posts being rebalanced to all be roughly equal.

D.

[chemical_art]

It doesnt really change anything though for thre better. It increases the fact that the AI HW is a strength check, and that overlaping fields of fire results in nightmares

Do blob-on-blob fights result in nightmares?

Blob on blob fights are very different from AI HW offenses.

To start, the player usually choices where (and often when) the blob-on-blob fight occurs. AI HW ultimately forces the player to not decide, they have to attack.

[TechSY730]

I actually like the current system. I'm just wary of the current stats.

I like the idea of brutal picks. However, I agree that the brutal picks are a bit too...um...brutal. I guess the issue is that the difference in difficulty between non-brutal picks and brutal picks it too high. Also, some brutal picks are more threatening then others. (Thus, nerf the brutal picks, and buff the non-brutal picks somewhat)

Could having some fancy extra system in place to prevent too much variance in structure placement making it hard to balance be good? Maybe. However, I think forcing a seeding strategy to such a narrow range (like evenly spaced or all next to command station) is going too far.

Some of the ideas of integrating core shield gens or core shield gen "lookalikes" as an optional way to weaken the homeworld (like blowing up the shield gen or shield gen replacement will blow up a linked core guard post, and vica-versa) would be a nifty way to have a fallback if you get "extra brutal" RNG layouts.


Again, I would like to say for the short term, before any stat balancing needs or new logic needs to be done (as a short term stopgap):
-Reduce brtual pick counts the AI getes by 1 for all difficulties
-Make the number of brutal picks constant for all difficulties (no 1 or 2 brutal picks, depending on chance, use the middle of current bounds, rounding up. Combined with the first thing, this makes difficulty 7 go from 1 to 2 brutal picks to always having 1 brutal pick. Yes, that means any difficulty <6 will get none. That is fine)
-(LESS IMPORTANT)Make the AI superfort a brutal pick, and brutal pick exclusive so they don't seed anywhere else anymore (though the Z trader can still sell it to the AI, if it is already on the list of the things the AI can buy from it)


Some of these can be rolled back as other, more extensive options are explored (if any)

[chemical_art]

Tech, your ideas are simple, but they don't address the core issue that it is possible for combinations of posts that result in strategic gridlock.

Usually I am in favor of quick fixes but for some reason I am not in this case: The core issue has to be addressed, overlapping brutal picks result in bad situations. This has to be addressed, reducing the number of brutal posts and streamlining their number does not.

[Toranth]

Moving all the guard posts next the the AI's home command makes the AI HW a strength check, did you play the game up till now correctly? This then turns the AI's home command into a one-off fight that is over in a single attack (or two) that is resolved very quickly.

Stop taking suggestions out of context.  The suggestion to move everything to the home command station would INCLUDE a rebalance of the guard posts.

Was this directed at me?  Because that quote wasn't something I said, and my post made no assumptions about rebalancing.

I actually like the current system. I'm just wary of the current stats.

I like the idea of brutal picks. However, I agree that the brutal picks are a bit too...um...brutal. I guess the issue is that the difference in difficulty between non-brutal picks and brutal picks it too high. Also, some brutal picks are more threatening then others. (Thus, nerf the brutal picks, and buff the non-brutal picks somewhat)

This is my stance.  Ideas that balance the intensity will easily win my support; ideas that want to change things dramatically are being met with skepticism.  I don't think it is broken, so why should it change?

Could having some fancy extra system in place to prevent too much variance in structure placement making it hard to balance be good? Maybe. However, I think forcing a seeding strategy to such a narrow range (like evenly spaced or all next to command station) is going too far.

Some of the ideas of integrating core shield gens or core shield gen "lookalikes" as an optional way to weaken the homeworld (like blowing up the shield gen or shield gen replacement will blow up a linked core guard post, and vica-versa) would be a nifty way to have a fallback if you get "extra brutal" RNG layouts.

Again, I would like to say for the short term, before any stat balancing needs or new logic needs to be done (as a short term stopgap):
-Reduce brtual pick counts the AI getes by 1 for all difficulties
-Make the number of brutal picks constant for all difficulties (no 1 or 2 brutal picks, depending on chance, use the middle of current bounds, rounding up. Combined with the first thing, this makes difficulty 7 go from 1 to 2 brutal picks to always having 1 brutal pick. Yes, that means any difficulty <6 will get none. That is fine)
-(LESS IMPORTANT)Make the AI superfort a brutal pick, and brutal pick exclusive so they don't seed anywhere else anymore (though the Z trader can still sell it to the AI, if it is already on the list of the things the AI can buy from it)

I don't think the SuperFortres should be a random pick - It's a special AI-type goody.  Otherwise... Setting a constant number of brutal picks is a good idea, I think.  However, I think that Diff 9+ should still get 3.  At the very least, Diff 10 still should.

Tech, your ideas are simple, but they don't address the core issue that it is possible for combinations of posts that result in strategic gridlock.

Usually I am in favor of quick fixes but for some reason I am not in this case: The core issue has to be addressed, overlapping brutal picks result in bad situations. This has to be addressed, reducing the number of brutal posts and streamlining their number does not.

The RNG will sometimes result in bad situations; this is a given.  I had a wonderful experience where at the beginning of a new game, the Z Trader moved through a Mk IV AI world next to my homeworld - and sold a OS Raid Engine when it left!  2 minutes into the game, I was facing 100 ship Mk IV wave.  That game was short.  Screwed by the RNG?  Oh, yes.  But, oh well.  That's the RNG for you.

As long as we can reduce the number of times an impossible situation occurs, rather than a merely difficult one, then I think that addresses the problem.
Reducing the number of brutal picks does that, by reducing the difficulties at which you will face multiple Brutal posts.  At higher difficulties (9+), it seems a safe assumption that players are willing to deal with difficult situations.
And at Difficulty 10?  Well, nearly impossible is what you're asking for, right? 

[chemical_art]

And at Difficulty 10?  Well, nearly impossible is what you're asking for, right?

No.

Which is where the conflict happens.

10 has enough already, it doesn't need theoretically impossible situations.

[LintMan]

If it was not random, a pattern could be found, and the defense is fairly useless.
If it was not random, but the area of decloaking is so large that you couldn't sneak by it by watching it, it is in effect global.
If it was random, one could save scum to get around it.
If save scumming doesn't work because the decloaked area is too large, it is in effect global.

Getting back to my earlier point, what if we allow cloaking, but made the AI smarter about responding to it?  Let's say the HWs start out with just standard tachyon sentinels.  The player could take some out and start doing some cloaked operations, but then after annoying the AI enough (ie: some damage threshold is reached, or maybe a cloaked vs non-cloaked damage ratio is exceeded), the AI starts producing more detection-capability: pulling tachyon-capable ships from reserve, warping in more tachyon sentinels, introducing a new "tachyon guardian" (there isn't one of those already, is there?), or "purchasing" a tachyon pulse from its normal reinforcement points, etc.  The more pain generated by the cloaked ships, the stronger the AI response.  In any case, though, the AI would use these detectors to aggressively to locate and destroy the cloaked ships.

[Draco18s]

Stop taking suggestions out of context.  The suggestion to move everything to the home command station would INCLUDE a rebalance of the guard posts.

Was this directed at me?  Because that quote wasn't something I said, and my post made no assumptions about rebalancing.
[/quote]

I screwed up a quote tag somewhere, fixed the original.

[TechSY730]

Stop taking suggestions out of context.  The suggestion to move everything to the home command station would INCLUDE a rebalance of the guard posts.

Was this directed at me?  Because that quote wasn't something I said, and my post made no assumptions about rebalancing.
[/quote]

I screwed up a quote tag somewhere, fixed the original.

We're messing up quotes all over the place here.

[KDR_11k]

Sneaking around the HW is fun but with the AI's weakness in terms of tachyon handling (planetary coverage seems like a band-aid) that should be more fleshed out. Especially things like the Wrath Lance ask for cloaking since it will start wiping your fleet rapidly once it uncloaks.

[Vacuity]

I'm still a little confused as to how the idea of seeding clones across the map is supposed to work.  Based on what I understood other people writing, the idea is to essentially have core guard posts seeded all across the map?  That would make the game as a whole a bit harder, particularly in the early stages.  Quite how much harder, I'm not totally sure, but facing core guard posts a couple of steps out from your homeworld would make initial expansion a much bigger challenge.  And considering that it's not *that* rare to get fortresses on regular worlds depending on the AI personality, it won't be too long before there will be complaints of "there's a core needler guard post and a fortress together (under a shield) and I can't get past them."

If brutal picks are also showing up outside of the core and homeworlds and with multiple copies then that's making life massively harder.

[LaughingThesaurus]

The clone guard posts would only be on the core worlds. It's essentially a microcosm of CSG-like guard post defenses that you peel away within that microcosm to get at the actual homeworld-- which simulates something you've done throughout the entire game up until then. Strike the AI's weak spots, and sneak around their strong spots.

[KDR_11k]

Special rules tend to make the game harder to learn though.

[LaughingThesaurus]

"Special rule"? I wouldn't think it's that complicated. The core guard posts already protect the home command station, all I'm thinking is adding a weakness-- the same weakness another major game mechanic has-- to the core guard posts and spreading them around more to add more angles of attack.