Author Topic: (Thought Exercise) Reduce knowledge to reduce AIP scaling.  (Read 7328 times)

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,506
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #60 on: April 14, 2013, 06:43:45 PM »
Quote from: Valtiel
Good point. When you've got the bare minimum of stuff you need to take down the AI homeworlds, why take any other planets?

Serious question. I think we need to have an answer to that before we decide what to do with strategic balance.

My point is that the answer is "there is no reason to do so".  I don't see any problem with that being the answer.

Of course, it's not so simple to determine the minimum necessary to win those battles; there's probably a minimum you'd need to have a _chance_, but you would still have some motive to take more stuff to improve that chance (unless you don't mind savescumming to the hilt).

But once you know you can win, then balance questions leave the stage.  I'm not telling y'all "if you can win, you should"; if you like playing past that point or even possibly endangering your win by ramping up AIP further then that's totally your prerogative.  I'm just saying that I'm not really thinking it's necessary for that "I can win, but I'm choosing not to" phase of your games to be particularly balanced.  The part I need to balance runs from when you start to when your win is assured.  Or when you die in flames, but of course that never happens.

AI HW are sieges. Meaning if you can chip them away faster then they can regen, well, time is meaningless in ultra low AIP games. It is why I can still cheese low aip champ games. Sure, they have nemesis, but I can chip away 10% of your health each "volley", and each "volley" takes more health away then is regened per volley, eventually I win.
To clarify, is that just taking a single champ FF against the AI HW over and over and over again?

Quote
Idea: Follow your idea with the nebulas vaguely. Each time a core guardpost is knocked out, the HW releases a pulse of units on its planet, in the form of free threat. This pulse is meant to flood enough threat to knock out ultra low aip games, but not higher aip games.


EDIT: Make it devious, and make the units "special" in that they are meant to be siege breakers, also known as fortress world breakers. I'm thinking hunter killer (s) here for 7+ difficulties.
I like that idea a lot, actually.  Been planning some kind of counter-spawn from core guard post death for a while.  I don't want to overshadow the Core Raid Engine (the Core CPA thing is already overshadowed, I need to address that) but a one-time on-death pulse is different than something that will continue to hammer you until destroyed.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #61 on: April 14, 2013, 07:26:58 PM »
@keith:

AI champs are by far the worst abusers of it, yes. But the idea itself remains. That if I can do an attack that wears the AI HW down faster then it can regen, eventually, I win. Unless something else happens, which draws me too my idea...


Typically, for low AIP games, I have to go "all in" to destroy any guard post. Meaning I have nothing but my defenses to help me prevent counter attack. So a pulse that is weak against fleetships but strong against defenses is really threatening. In an higher aip game I generally have something (territory, extra defenses, strong econ, etc) to prevent a bee line to my own HW. But on a low AIP game, freak pulses are threatening. I think the AI should offensively capitalize on that.


« Last Edit: April 14, 2013, 07:29:33 PM by chemical_art »
Life is short. Have fun.

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,559
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #62 on: April 14, 2013, 07:31:42 PM »
Maybe the AI HW could have those pulses on each core guard post death in return of the core guard post raw AIP cost being slashed (from 2 to 1)?

Anyways, I think that a pulse per core guard post death accomplishes the "last ditch" retaliatory reaction from the AI than the +2 AIP (or whatever it is) currently does.

Offline _K_

  • Full Member Mark III
  • ***
  • Posts: 218
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #63 on: April 15, 2013, 03:51:28 AM »
Maybe the AI HW could have those pulses on each core guard post death in return of the core guard post raw AIP cost being slashed (from 2 to 1)?

Anyways, I think that a pulse per core guard post death accomplishes the "last ditch" retaliatory reaction from the AI than the +2 AIP (or whatever it is) currently does.

Yeah but if you kill a guard post and survive the counterattack, the AI should be more weary of your actions... which is kinda the purpose of AIP.
Making each guard post work as a small-scale MKV counteratack post sounds okay though, not entirely sure about nerfing the AIP they generate.

And speaking of "last ditch" thing, shis should happen when the actual AI Home command station goes down. Too bad right now there are too many easy ways to kill the second one once its neutered.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,506
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #64 on: April 15, 2013, 09:48:42 AM »
Making each guard post work as a small-scale MKV counteratack post sounds okay though, not entirely sure about nerfing the AIP they generate.
I was considering just removing its +2 AIP on-death for simplicity (while adding the on-death-spawns), since it already generates +10 AIP-floor (on non-Lazy-AI).

Does the +2 AIP really make much difference at that stage?
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline _K_

  • Full Member Mark III
  • ***
  • Posts: 218
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #65 on: April 15, 2013, 10:10:57 AM »
Does the +2 AIP really make much difference at that stage?
I was about to make the argument that its adds up, and as result is significant for low-AIP games... but then the +10 to the floor on each of those posts probably has much, much bigger impact on those games.

And on medium-high AIP, 12-16 AIP ist really that much, yeah. Still a little noticable, but i agree it is not really significant.

Do we maybe want to switch the whole guardpost mechanic from "Lazy AI" to be global then, with simple +10 to floor, +0 to AIP? I can see the point of static strategic reserves being an option, but this looks like a direct buff to the higher AIP games and nerf to the low-AIP games.
Since its twofold, i'd rather see it as a fact, not an option.

Offline Diazo

  • Master Member Mark II
  • *****
  • Posts: 1,710
  • I love/hate Diff 10
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #66 on: April 16, 2013, 09:30:00 AM »
Ok, I've thought about this for a while and while I know the thread was a thought experiment to begin with I think the basic idea is worth trying, so here's what I'm thinking of doing (near-term; if it's well received this could be in 6.019, though I suspect there will be opposition, which is fine) for ratcheting back the "arms race" a bit to bring the expected AIP numbers (and by extension planets-taken count) back closer to how they used to be:


1) Reduce the effect of AIP to 2/3rds of current in the following uses:
- When determining the size of waves (including counterattack waves, etc).
- When determining the size of reinforcements.
- When determining the size of a CPA.
- When determining the rate at which the Special Forces grow.
- On Lazy-AI, when determining the rate at which Strategic Reserves grow (still capped at 200; Non-Lazy-AI would stay constant at 200).
- When determining the size of a Core CPA Guard Post's response to being triggered.
- When determining the frequency with which Broken-Golems-Hard/Botnet-Golem-Hard/Spirecraft-Hard exos come (there's a time-based floor to this that would remain unaffected, though).

These places where AIP is used would remain unaffected (this combined with the above is, to my knowledge from just now searching the code, a comprehensive list of things that AIP influences) :
- When determining the AI's tech level for ships in waves, etc.
- When determining when the AI should get new bonus ship type unlocks.
- When determining whether the human has too advanced a champion at too low an AIP and thus the nemesis response needs to ratchet up (so this remaining the same favors the player)
- When determining the Dyson's "population cap" for dysons in the galaxy.
- When determining how many Roaming Enclaves to spawn per event.


2) Reduce the max knowledge a player can gain from a planet from 3000 back to the old value (pre-4.0) of 2000.


The main caveat I can think of is that Fallen Spire would be nerfed somewhat by this as you'd probably need to take more planets to get the K you need to upgrade your FS stuff, and the FS exos would not be impacted by the above (not being based on AIP).  But I don't think it'd be a very big nerf because taking those extra planets doesn't make FS exos bigger either, and honestly once you have an FS fleet the AI's not exactly going to be able to stop you from taking those extra planets or punish you very much for the extra AIP they represent.


I'm thinking that is pretty much what my intent was in starting this thread, but lays it out better.

Especially if the K cost changes being talking about in the other thread happen at the same time, actually both happening at the same time would probably be necessary.

If K is dropped to 2,000 per planet, 6,000 for the Mk IIIs and the high-mark turrets probably need to come down.

On the FS stuff, is it possible to give the City Hubs the Spire Archive mechanic so the player gets an extra 3,000K per city hub? That would take care of FS being short on K nicely.

D.

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #67 on: April 16, 2013, 11:42:27 AM »
Keith I don't feel these changes really address the underlying issue.

Meaning, if you cut AIP for some, but not all things by 2/3rds, but decrease the K by 2/3rds as well, the result is that the "power curves" of ai power and human power doesn't change, except now more of it is front loaded to low planets since the player homeworld now gets 6x (!) the knowledge on its first world compared to additional ones (10k + 2k K) compared to (2K). The result is the player curve of increasing power is shifted so even more sharply to low planet amounts.

Never mind that since things like wave strength (via techs) are not reduced, even discounting the player HW, it still shifts more to the ai's favor for high AIP games. Again, shifting the power curves to lower aip games...which certainly don't need it.


I feel this simply does not address the issue, that low aip games are favored, in no some part, because the 1st world is so powerful. If anything, I think the player HW should get less base K in reserves, so the power curve is shifted away from the first world.
« Last Edit: April 16, 2013, 11:44:12 AM by chemical_art »
Life is short. Have fun.

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,559
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #68 on: April 16, 2013, 12:07:52 PM »
Yes, how the human and AI "power curve" grows is important, but also (and possible more importantly so for low AIP games) their base values (where they "start") is also very important.

Hmm, so how can we encourage people to take more planets without making the very early game straight up "too much" harder in "normal" (not ultra-low AIP) strategies?

Maybe something like increase the "base" AI strength, but make the very first wave from the AI (for each AI player) be less than what the normal base scaling would imply. So the very first wave an AI player sends (aka, the first wave from AI 1 AND the first wave from AI 2) you would be around the same strength it is now, but after that, the increased "base" AI strength will kick in, putting you in more pressure if you don't try to get more resources after that very early stage, which usually doesn't happen in mid to high AIP strategies.

Then again, seeing how some maps make the very early game slow (ever been seeded next to high mark planets?), maybe it should be the first two waves.

Say, 1.5x the "base" AIP strength, but have the first wave (per AI player) be about 66.7% strength (which would bring it about to the level it is now), and the second wave be about 83.3% strength (putting it about 25% more than it is now). (The base AI strength increase may need to be more severe)

Then we can do the 2/3 reduction in AI scaling in return for this (if the base AI strength is buffed more extremely, then this will need to go down a bit more).

As the low AIP has been made harder, we don't need as much if any knowledge reduction per planet. Maybe knock it down to 2500 knowledge a planet (if the reduction in AIP growth rate is made more severe than 2/3, then perhaps 2000 would be acceptable)? That way, the human can actually grow in strength faster for more stages of the game (as the human reward per planet has not been nerfed as much), until finally you start to plateau near the limits of what knowledge can get you, and the AI starts overtaking you again.

Yes, this does make the early game harder (a little bit for mid to high AIP strats, quite a bit so for low AIP strats), but the late game easier. IMHO, this is a good thing, as right now, most late games devolve into "grind fests", and most early games are just too "boring" as the AI isn't doing much (which is the design intention, but right now, it is taken a bit too far IMO). Hopefully, the new non-lazy AI option should keep the homeworld from being a pushover. (Speaking of which, that AIP for the strategic reserves may need to come up some if something like this is done)

The exact numbers need tweaking, as I am sure the numbers I proposed will lead to a too early mid to late game without a proper increase in difficulty for the low AIP strat early games.

EDIT: Also, I do think base knowledge could stand to go down a bit, but not a huge amount.

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #69 on: April 16, 2013, 12:22:23 PM »
Current values of K:

Base K: 13K
Additional K per planet: 3K

Number of K needed to double K from base: 4.3 planets
AIP increase in response: (4.3 * 20 )  = 86
Relative increase in K power: 2
Relative increase in AI Power: 8.6 times


Number of K needed to triple K from base: 8.6 planets
AIP increase in response (8.6 * 20) = 173
Relative increase in K power: 3
Relative increase in AI Power: 17.3



New changes
Base K: 12K

Number of K needed to double K from base: 6 planets
AIP increase in response: (6 * 20 * .66 )  = 8
Relative increase in K power: 2
Relative increase in AI Power : 8 times


Number of K needed to triple K from base: 12 planets
AIP increase in response (12 * 20 * .66) = 160
Relative increase in K power: 3
Relative increase in AI Power: 16


These new changes, if everything was equal, marginally decrease the sharp increases of AIP response to human's K response. But, wave strength doesn't increase linerally with AIP on higher difficulties, the very difficulties that need the low aip games, so it fails on that front.

No matter what, it makes the relative power increases of the players over planets even less, and when combined with the AI still getting better units in waves, hardly makes me think this is an improvement in any meaningful way even for 7/7.
« Last Edit: April 16, 2013, 12:25:35 PM by chemical_art »
Life is short. Have fun.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,506
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #70 on: April 16, 2013, 12:52:48 PM »
if you cut AIP for some, but not all things
To clarify, as I said in that post there are only two parts of the AI response that would be unaffected: when it gets higher-mark units, and when it gets new bonus unlocks.  Are either of those a problem to not be affected by this?  Part of what I was getting from feedback was that the bonus unlocks weren't really happening much due to how much AIP you could actually take anyway.

If the tech level thing is a problem that could be adjusted too.

Not that this is the main problem you have with the idea, I just wanted to clarify that the parts I left unaffected were because I thought that's what folks wanted to see more of.


On the more general question I think we're still talking about two different things.  To sum up, here are two questions for you:

1) If you have the power and position (including CSGs, if on) to kill the AI, would you ever take another planet?

2) If you do not have the power and position (including anything non-AIP increasing like hacking) to kill the AI, would you ever not take another planet?
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Diazo

  • Master Member Mark II
  • *****
  • Posts: 1,710
  • I love/hate Diff 10
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #71 on: April 16, 2013, 01:51:35 PM »
@Keith: I think we (being you and me) are now talking about two different things (if related).

On  your questions:
1) If you have the power and position (including CSGs, if on) to kill the AI, would you ever take another planet?
Depending on  how you define position, sometimes yes. I can technically be in a position to stomp the AI, but my defense is so weak that the AI's response would kill me before I can turn to take out the second AI. This would be rather uncommon, but it does happen to me due to how I planet hop out my forward bases.

Quote
2) If you do not have the power and position (including anything non-AIP increasing like hacking) to kill the AI, would you ever not take another planet?
No (err, yes?). By this I mean that if I don't have the strength needed, I have to take another planet, that is the only way to make my fleet stronger. I could K-hack a system or two I suppose if there's only a small strength difference left, and I might delay taking more planets to deal with something like a CPA, but I'm going to need more planets before endgame.

@chemical_art: On power curves, my original contention when I started this thread was that the power curves are fine, progress along them just happens too fast on both sides, so I posted this as a method for slowing down how fast the 'power' of each side escalates.

On the starting K, I never actually stated it outright, but I was assuming that would get knocked down by 1/3 also to keep things in line.

If Mk III fleet ship costs (and some of the turret costs) get reduced as is being talked about in another thread, dropping starting K from 10,000 to 7,000 should not be a big deal.


@Keith (again): It sounds like this is something you are actually seriously looking at. However, before committing this to a patch, can you post the changes exactly (as in, copy-paste from what the patch-notes would say?) as I don't feel anyone is really on the same page here.

D.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,506
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #72 on: April 16, 2013, 02:03:40 PM »
On the starting K, I never actually stated it outright, but I was assuming that would get knocked down by 1/3 also to keep things in line.
I'm happy to do that, but there was opposition to that so I'm on the fence on it.  The general idea that the player's starting planet is too strong does seem valid, though to some extent that's just how the game has been for a long, long time, and I don't know if it's really of primary relevance (as opposed to "how much more do you need to win?").

Quote
@Keith (again): It sounds like this is something you are actually seriously looking at. However, before committing this to a patch, can you post the changes exactly (as in, copy-paste from what the patch-notes would say?) as I don't feel anyone is really on the same page here.
Sure.  I'm not decided on doing this or not at this stage, but it seems like it would address several concerns.  The current roadblock is that some folks think the starting HW needs a nerf (to the extent that that is more important than K/AIP scaling) and others think the starting HW most definitely does not need a nerf.  Though I'm still not convinced we're focusing on the most important pieces of the puzzle (the actual motivation for taking planets).
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline LordSloth

  • Sr. Member Mark III
  • ****
  • Posts: 430
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #73 on: April 16, 2013, 02:14:02 PM »
I really need to get more practice in. I can only manage at most an hour a day during this week, my play hours are dropping.

However, as I mentioned elsewhere, on a ~7ish game, I'm finding the Fabricator positioning and variety dramatically increasing the incentive to look outside the essentials. Combine that with a decreased cost for taking territory* and I'm no longer dismissing them out of hand.

*What about the 'reinforcement pulse' mechanism. Are there any other mechanics that count the number of Non-AI systems?

Vanilla, vanilla, vanilla. I need to keep that in mind, but putting it aside for a moment, decreasing the energy generation would also promote the value of a Zenith Power Generator capture.

The carrot is probably going to be a bit more attractive, but that doesn't particularly help the stick part of the equation. The AI likes Vinegar more than Honey after all. Bitter, bitter vinegar of defeat.

But yeah, a variety of the ship and turret balance changes being discussed should make the rewards of territory grabs and advanced factories and fabs more attractive.
« Last Edit: April 16, 2013, 02:17:21 PM by LordSloth »

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,506
Re: (Thought Exercise) Reduce knowledge to reduce AIP scaling.
« Reply #74 on: April 16, 2013, 02:18:23 PM »
However, as I mentioned elsewhere, on a ~7ish game, I'm finding the Fabricator positioning and variety dramatically increasing the incentive to look outside the essentials. Combine that with a decreased cost for taking territory* and I'm no longer dismissing them out of hand.
But if you have enough that you know you could take down the AI HWs, would you take another fabricator? 

It makes sense to take one if you're not sure you'd win (or survive) with your current force.

It also makes sense to take one "because it would be fun" but if that makes the win harder that's not really a balance concern, because in that case you're not making decisions primarily to win.

Is there another reason?
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!