Guardian numbers

[chemical_art]

I don't think you are following me.  Right now, on 10/10, the AI gets 3 Guardians per Guard Post.  But what is the cap?  If the cap is 10 Guardians per Guard Post, then the AI isn't at cap initially.  So are you suddenly okay with it getting 3 per Guard Post to start with if that isn't the cap?  Or is 3 per Guard Post at 1 second into the game too much?  That's what I was referring to.

Yes, I would be OK with that...my beef isn't the high numbers so much as them being at max. Although the high numbers are annoying.

It really ruins the concept of alerting worlds. I don't mind the AI making a fortress if it has been under alert for hours. But having it be so strong in a backwater world, every backwater world? It is just a grind.

[Hearteater]

Ok, just wanted to clarify.  I agree with your position.  I think after WHGPs aren't counted the initial spawns will be reasonable, but that can't be the cap or what's the point.

[keith.lamothe]

I haven't changed the logic for how many guardians start on each post... ever, actually.  So it should be whatever Chris set up last, back in the 4.0/5.0 days.

But I'll take a look at the current/cap and maybe make some revisions when I'm in there.

How do you think it should answer the question "how many guardians should planet X start with?" ?

[chemical_art]

I think that for 7/7 it should be around 1/4 of the cap, with 10/10 approaching 75%.

If possible, it would also vary by AI type though, with defensive AI types get more. So a turtle 7/7 would get 50% hile a turtle 10/10 gets 100%.


Also, if possible, I think just as key though would be variation of the cap. So for 7/7 you might get some worlds with no guardians, but on other worlds you'll get 50%.

[TechSY730]

I haven't changed the logic for how many guardians start on each post... ever, actually.  So it should be whatever Chris set up last, back in the 4.0/5.0 days.

But I'll take a look at the current/cap and maybe make some revisions when I'm in there.

How do you think it should answer the question "how many guardians should planet X start with?" ?

How about an expected value of .4 to .6 for every non wormhole guard post, with no more than 2 on a single guard post? And a planetary max of min(.9 * that planet's guard posts, guardians per world? (No planetary min, to allow some worlds to seed without guardians).

EDIT3: Some sort of extra logic would need to be made to ensure that galactic average does not deviate too far from the expected value, so the AI would still have some guardians on game start even if the RNG hated it during the "this is how many guardians you get" selection, and to stop the RNG from letting every world from having or nearing that min(.9 * that planet's guard posts, for almost every planet.

For AI homeworlds and core worlds, this should be closer to an expected value of 1.2 for every non-wormhole guard post for the core and home worlds. For the core worlds, no more than 2 per guard post, with a planetary max of 10, and a planetary min of 1.05 * guard posts on that core planet. For the AI homeworlds, no more than 3 per guard post, with a planetary max of 14, and a planetary min of 1.1 * core guard posts on that home.

These would only be for initial seeding. Not sure about how often reinforcements should give one. Caps for reinforcements should be higher than the caps for game seeding, though maybe not much higher.

EDIT: Oh yea, the constant caps (like max of 8 per non-core and non-home world, if that would be less than .9 * number of guard posts) should go up with difficulty. Any multiplier to the guard post counts in any of these should remain constant over difficulty, as the number of guard posts goes up over difficulty anyways, so no need to "double dip" with those.

EDIT2: For the above, count the command station itself as one of the guard posts. If the AI only gets a chance to have some of their guard posts have more than other on a planet, their first choice for one of those posts to get more of them should be their command station.

[chemical_art]

Digging through my logs, Keith, the logic really hasn't changed.

However, with this emphasis of stronger guardians, combined with me trying to play normal maps, is a one to punch to the face.

[Hearteater]

I wonder if Guardian cap per system shouldn't be X per Guard Post (including CS, excluding WHGPs), and go up by 2 each time a non-WHGP is destroyed.  So when you neuter a planet, you increase its Guardian cap by the number of Guard Posts it had.  I'm not sure what value to set for X, but my first thought is Difficulty/2.  Then I'd go with Chemical Art's idea of 25% of cap, though I probably wouldn't bring 10/10 up to 75% just because the cap is already increasing with difficulty.

EDIT: I guess this goes to show how noticeable Guardians were before

[keith.lamothe]

I think that for 7/7 it should be around 1/4 of the cap, with 10/10 approaching 75%.

I think maybe just ranging from 1/4 at 7 to 1/3 at 10 would be fine with the cap increases that already happen, as Hearteater mentioned.  But yea, in general sounds good.

If possible, it would also vary by AI type though, with defensive AI types get more. So a turtle 7/7 would get 50% hile a turtle 10/10 gets 100%.

That would be easy to do, yea.  It'd be a slog for the player, but I guess that's what written on the tin.

Also, if possible, I think just as key though would be variation of the cap. So for 7/7 you might get some worlds with no guardians, but on other worlds you'll get 50%.

I think it'd be best if the variation weren't totally random, so an ARS/AdvFact/ASC/CSG planet would have more, and stuff with nothing important could have less.  It could also vary with the subcommander personalities in the expansion.

I wonder if Guardian cap per system shouldn't be X per Guard Post (including CS, excluding WHGPs), and go up by 2 each time a non-WHGP is destroyed.

Hmm, but wouldn't that basically be "never destroy WHGPs"?  They're already a serious hassle to kill (unless you happen to have a few FS cap ships just laying around).  But I suppose with this rule change they could actually be made much easier to kill and just make it clear in their description that "by destroying these you can control the shape of the AI's reinforcements here, but the AI will station more guardians on the command station" or something like that.  As long as both choices (to destroy or not to destroy) are good ones in enough different circumstances they don't have to be giant balls of HP.

[chemical_art]

Oh yes, the changes to guard post logic influences their strength a lot.

Which is why in my 7/7 games I didn't notice, but at 9/9 I feel like I'm playing hybrids again (during 8/8 w/adv hybrids)

[LordSloth]

I wonder if Guardian cap per system shouldn't be X per Guard Post (including CS, excluding WHGPs), and go up by 2 each time a non-WHGP is destroyed.

Hmm, but wouldn't that basically be "never destroy WHGPs"?  They're already a serious hassle to kill (unless you happen to have a few FS cap ships just laying around).  But I suppose with this rule change they could actually be made much easier to kill and just make it clear in their description that "by destroying these you can control the shape of the AI's reinforcements here, but the AI will station more guardians on the command station" or something like that.  As long as both choices (to destroy or not to destroy) are good ones in enough different circumstances they don't have to be giant balls of HP.

Non-WHGP. So basically, you're suggesting an inverse relationship between regular guard posts and guardians, with WHGPs not factoring into the count at all?

[chemical_art]

To be fair, I do care about guardian types now.

My AI had Tractor guardians.

Ooooo! I'll use my zombards and enclaves.

But then the AI also had enclaves. {enclave vs enclave fights are very enjoyable.}

Ooooo! I'll use this curse golem I found.

But the AI also had implosions.

And that was that.

[chemical_art]

Keith I'm actually liking your idea of WGP being used, but my worry is that it is too map dependent. With both lattice and snake maps, it seems too hard to balance.

I don't think guardians should be judged based on them. Low entry point maps (snake) get enough breaks as is.

[Hearteater]

I wonder if Guardian cap per system shouldn't be X per Guard Post (including CS, excluding WHGPs), and go up by 2 each time a non-WHGP is destroyed.

Hmm, but wouldn't that basically be "never destroy WHGPs"?  They're already a serious hassle to kill (unless you happen to have a few FS cap ships just laying around).  But I suppose with this rule change they could actually be made much easier to kill and just make it clear in their description that "by destroying these you can control the shape of the AI's reinforcements here, but the AI will station more guardians on the command station" or something like that.  As long as both choices (to destroy or not to destroy) are good ones in enough different circumstances they don't have to be giant balls of HP.

Non-WHGP. So basically, you're suggesting an inverse relationship between regular guard posts and guardians, with WHGPs not factoring into the count at all?

Correct.  Only killing regular GPs would increase the Guardian cap.  So when we neuter worlds, we face more Guardians as the AI tries to spawn more at the neutered systems to protect them.  Killing WHGPs would be neither here nor there.

[keith.lamothe]

Non-WHGP.

Ah.

Oops.

Sometimes I do actually read, really

[chemical_art]

[Epic side bar]

Cursed golem has a build time of 720:00

It builds at 62 m + c per second.


Cant its build time be lowered, so I don't have to use whole caps of engineers to repair it?