Author Topic: AI War state of the game  (Read 45573 times)

Offline KDR_11k

  • Hero Member Mark II
  • *****
  • Posts: 904
Re: AI War state of the game
« Reply #270 on: November 13, 2012, 02:16:21 pm »
Champion is plain OP, I don't think that's really in question. When I have one I tend to forget all about even having those paid-for fleets since 90% of the planets can be cleaned out with repeated champion application at zero resource cost. Well, Shadow Frigates are weak but once you get Destroyers or better the champ is really an army of one.

Offline Hearteater

  • Core Member
  • *****
  • Posts: 2,334
Re: AI War state of the game
« Reply #271 on: November 13, 2012, 02:28:33 pm »
At present, I find a really comfortable early-game (when I just have my homeworld) speed is at least Mk II Harvesters.  So if Harvesters go away, I'd like to see a bump in homeworld economic output a bit, just to keep the early game moving better.  I really don't like a single homeworld with no harvester upgrades for early game.  It just feels really slow and makes me wait a lot.
How do you normally start off?  I generally try to rush a world with just fighters (often just Mk I fighters) within the first five/ten minutes, I guess precisely to be able to get an econ station up (I haven't got used to using harvesters even though they are presently clearly better), but I can totally believe that that might be too dangerous a move on higher difficulties --- especially since up to diff 8 you can certainly survive the first three/four waves with pretty minimal turret placements.  Not arguing, just haven't had the same experience and am curious to hear about your playstyle.
My current game I'm 3 hours in and haven't taken a single system (1HW 9/9).  I've unlocked Harvesters III, Scout II, Shield Bearer II, and Gravity Turrets if I recall correctly.  But really I'm looking at the first hour or so.  If unlocking a lot of Starships or Mark IIs immediately isn't feasible with starting economy (b/c Harvesters are gone) then that option might as well not exist.  I'd like to see a lot of flexibility in terms of how you start.  I don't want the early game to be too constrained.

Offline Martyn van Buren

  • Hero Member
  • *****
  • Posts: 642
Re: AI War state of the game
« Reply #272 on: November 13, 2012, 02:55:43 pm »
What did you do before the harvester buffs?

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
Re: AI War state of the game
« Reply #273 on: November 13, 2012, 03:08:10 pm »
What did you do before the harvester buffs?

Econ stations.
In some cases, Youtube And Netflix enter the picture. ;)

Offline zoutzakje

  • Hero Member Mark III
  • *****
  • Posts: 1,052
  • Crosshatch Conqueror
Re: AI War state of the game
« Reply #274 on: November 13, 2012, 03:19:48 pm »
What did you do before the harvester buffs?

The exact same thing as I do now: Build logistics everywhere (military in some rare exceptions). It just causes you to be more careful when you have a low economy. And it means you probably can't build any trader toys or mercenaries. I have never used and probably never will use Econ commands, simply because I believe the advantages from logistics and military commands to be more important than a good economy.

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: AI War state of the game
« Reply #275 on: November 13, 2012, 03:27:57 pm »
I think reducing the number of harvesters on the player HW (with or without a boost to the player's HW econ through buildings) would help nerf harvester's huge burst early game while not nerfing high aip games that hard.

I do think if you sink 18k into pure economy (harvesters III and econ III)  you should reap a strong economy mid game, since you have cost yourself two of your triangle ships from going from I to III at least (increasing said fleet's abilities five fold) and maybe I - IV (A 9 fold increase in power)
Life is short. Have fun.

Offline TechSY730

  • Core Member Mark V
  • *****
  • Posts: 4,570
Re: AI War state of the game
« Reply #276 on: November 13, 2012, 03:30:41 pm »
I think reducing the number of harvesters on the player HW (with or without a boost to the player's HW econ through buildings) would help nerf harvester's huge burst early game while not nerfing high aip games that hard.

I think scaling down the number of harvester spots on a homeworld from 12 (or whatever) to 6 (or whatever half the current is), while buffing the pre-placed stuff's output (which ones are TBD) is my favorite solution so far, as it targets what seems to be the biggest point of concern, early to early-mid game, while giving minimal impact to late game.

Offline KDR_11k

  • Hero Member Mark II
  • *****
  • Posts: 904
Re: AI War state of the game
« Reply #277 on: November 13, 2012, 04:21:33 pm »
Perhaps the economy has to be rethought, resources over time plus infinite time can of course result in "I'll just wait" approaches. Other RTSes don't let you wait* because if you don't go to the fun the fun comes to you. If time is free and can be converted to resources without limit then resources are essentially free and so are all the things bought with them. AIP over time adds a cost to time but since it's optional it doesn't really have a role in the core balance.

*=Campaigns excluded which is usually the reason why RTS campaigns suck.

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: AI War state of the game
« Reply #278 on: November 13, 2012, 04:25:02 pm »
Perhaps the economy has to be rethought, resources over time plus infinite time can of course result in "I'll just wait" approaches. Other RTSes don't let you wait* because if you don't go to the fun the fun comes to you. If time is free and can be converted to resources without limit then resources are essentially free and so are all the things bought with them. AIP over time adds a cost to time but since it's optional it doesn't really have a role in the core balance.



CPA's are a very real mechanic to this problem.

On my 10/10 game the AI was lobbing 15k CPA's despite me being at 10 aip. I should have been murdered, but only through minor factions was I not.

The mechanic is independent of AIP in its strength and eventually will grow enough to eventually knock out any player who stagnates, although it may take a long while.


On a second note: The player's caps on units shall cause it to, FS or Champions aside, always be smaller then the AI as a whole unless you ride the aip and make very, very judicious use of mercenaries, which eventually invites the wraith of CPA's.
« Last Edit: November 13, 2012, 04:26:53 pm by chemical_art »
Life is short. Have fun.

Offline Hearteater

  • Core Member
  • *****
  • Posts: 2,334
Re: AI War state of the game
« Reply #279 on: November 13, 2012, 04:52:58 pm »
What did you do before the harvester buffs?
I'd be at 5-6 hours instead of 3 hours, for the same amount of progress.  Note that I do play with AIP over time, so the extra time does have an impact.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: AI War state of the game
« Reply #280 on: November 13, 2012, 05:24:30 pm »
If time is free and can be converted to resources without limit then resources are essentially free and so are all the things bought with them. AIP over time adds a cost to time but since it's optional it doesn't really have a role in the core balance.
Not exactly.  You can also put the AI on -300% handicap.  I don't think playing without autoaip is nearly that bad, but playing on zero autoaip and staying on just your homeworld for 6 hours is also something I don't really go out of my way to balance: it's a player choosing to play a bit cheesy.

And there are other mechanics that form a cost-over-time.  CPAs are the main one: they're basically a "force check" for that point in time and game-difficulty (if AIP is high enough AIP also boosts them, though) ; if you don't have enough total power you just die.  That doesn't happen a lot in this crowd but that's because people are playing fairly efficiently with their time.
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 Draco18s

  • Resident Velociraptor
  • Core Member Mark V
  • *****
  • Posts: 4,251
Re: AI War state of the game
« Reply #281 on: November 13, 2012, 09:56:02 pm »
And there are other mechanics that form a cost-over-time.  CPAs are the main one: they're basically a "force check" for that point in time and game-difficulty (if AIP is high enough AIP also boosts them, though) ; if you don't have enough total power you just die.  That doesn't happen a lot in this crowd but that's because people are playing fairly efficiently with their time.

And one of the problems I have with various plots and the exo-waves.  It ends up (or at one time was true) that there's no pause in the fighting long enough for me to go on the attack in the mid-game in order to push the end-game.  I build a fleet, I repulse an invasion, I repeat.
(Which is why I always put golems/spirecraft on easy when that option showed up)

Offline ZaneWolfe

  • Sr. Member
  • ****
  • Posts: 272
Re: AI War state of the game
« Reply #282 on: November 14, 2012, 12:25:58 am »
...a simply a well thought out post that shows you have thought deeply about many aspects of the game...

Hello ZaneWolfe, I'm not one to regularly welcome anyone to the forums.

I'm a bit closed off and unfriendly in regards in that I don't acknowledge people directly, but in the ideas they present on forums.

Your post here demonstrates a well thought out argument that demonstrates a lot of critical thought and analysis. I don't agree with all of it, and my post here isn't to analyze what I agree or disagree with (maybe another time)

However, your post moved me enough to recognize your effort, so I directly welcome you and your analysis (I'm sorry I can't be more specific, my time is limited).

You are the first person I've seen new whom I thought "wow! He knows his stuff well!" from the very beginning.

Well done!

The fact that my posts says quite blatantly that I cheat and yet someone else thinks it is well thought out and is complementing me is either a testament to me actually putting serious thought and effort into it, a sign that I can lie my ass off well enough nobody knows I am doing it, or a stealth troll...

For the life of me, I am not sure which it is. For now I'm going with option A and just saying thanks, by not sending the superweapons to destroy you. What? I am evil. Not killing someone IS the nice thing to do, isn't it? If it is option C, well done good sir, I fell for it.

Offline Histidine

  • Hero Member
  • *****
  • Posts: 581
Re: AI War state of the game
« Reply #283 on: November 14, 2012, 10:27:20 am »
Well, given that champions have been fingered as one of the principal balance issues, how do we tackle it? The problems are, I think, threefold:

1) Champions are a resource-free, powerful one-ship navy.
No resource investment (in M/C or energy), yet once leveled they'll cleave their way through anything in their path. As with KDR_11k I don't even care about my regular fleet for the most part once I get my champ going; for one, the champ costs nothing to replace when it dies. It's not the first thing I reach for when I need to clear a path of tachyon guardians, or neuter a planet, or deep-raid an AIP reducer - it's the only thing.

2) Nebula facilities give a bit too many resources.
Epsilon Eridani and In Memoriam Terra facilities give 100 M/C per second. For reference, my 4/4 harvester (Mk III!) whipping boy with a Mk I Military Station does 244 M/C s-1.
Shattered Pillar facilities produce 100k energy, or 2/3rds of a planet. Gray Spire are 25 M/C s-1 and 75k E, Neinzul Mourner 50 M/C s-1 and 50k E.
Yeah, the rewards are pretty good, especially on a low AIP game. Once all nebulae are done, you're getting like 825 M/C a second and 775k energy, or 3 (energy) to 5 (M/C) planets worth of resources on MkIII harvesters.

3) Nebula ships are (probably) too good for their cost.
Whenever I need a fleetblob that I can afford to sacrifice but needs to not die en masse before accomplishing its goals, I turn to the nebula fleet, simply because it's a fair bit cheaper to replace after a complete wipe, while retaining a good degree of effectiveness.

I made a spreadsheet comparing the basic stats of some nebula ships to their nominal fleet ship/starship equivalents here.
  • So, what does the Zenith Starship have over the EER Avalanche, in a one-on-one comparison? Well, it has radar dampening (certainly handy), and an exotic hulltype (a.k.a. doesn't die horribly to bombers). The Avalanche, on the other hand, trumps it completely in terms of speed and range.
  • The cap comparison is more telling: a cap of Avalanches has just under half the HP of the Zeniths, though it breaks even in terms of after-multiplier DPS... but it only needs 29% of the M+C and 5% (!!!) of the energy supply. The Gray Spire Dawn/Plasma Siege Starship comparison turns out similarly (worse, in fact, as the Plasma Siege has no multipliers or dampening).
  • The Mourner Lament/Bomber comparison is a bit interesting. The bomber fleet actually gets you more HP/joule, though not DPS/joule or HP/DPS per M/C, than the nebula bomberships; it certainly holds up better in the cap energy comparison than the starships. On the flip side, the Laments' huge speed and range advantage still holds, and they get all kinds of little (and not-so-little) perks: taking a fraction of the damage from AoE weapons, not dying to Ion Cannons, not setting off Eyes, not suffering from engine damage, less DPS loss from attrition, you name it.

Some ideas
  • Maybe we need to bite the bullet and charge people resources to respawn their champion, and the champ could have an energy upkeep like everything else. To throw the players a bone, reduce the cost of respawning based on the champion's health at "death" (for warp-home or switching to a different hull, which is intuitive and makes gameplay sense.
  • Nebula ships could cost a little more (and use more juice), and/or...
  • Reduce the resource reward of nebulae a bit. In my 7 planet empire, 30-40% of the metal+crystal and energy come from the nebulae; quite a good bargain at the "price" of 20% higher effective AIP.
  • Apply Wanderer's idea of having nebula scenarios raise the AIP floor, to add some cost of all that firepower and income for the cheese monkeys who currently run a Shadow BB at 10 AIP.
  • Let the AI use the nemesis fleet more aggressively. Have only one or two patrolling the homeworlds; the others should be stationed in Strategic Reserve Space, waiting for the player champion to come out of his hidey-hole. Every time the Champion leaves the safety of allied territory, the nemeses should hound him to the ends of the galaxy. The deeper the champion goes into AI space, the more angry Shadow frigates (and bigger ships!) should be swarming him. Especially if he's triggering deepstrike threat. Especially if he's going for an AIP reducer or a nebula wormhole. That is what a Nemesis does, after all.

    The idea is that the player should make the tactical and strategic consideration of using their regular ships to assist the Champion, and/or paying the AIP to kill an AI Command Station and create an "island" to reduce the nemesis anger (and deepstrike threat), when sending the champion off on deep raids. Though for this to work optimally, you might have to increase the population and respawn time of the nemeses (so there's more of them waiting for you, but they stay dead a bit longer once killed), and increase the AIP "ceiling" for the nemesis pop cap.
« Last Edit: November 14, 2012, 11:16:19 am by Histidine »

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: AI War state of the game
« Reply #284 on: November 14, 2012, 10:51:03 am »
The fact that my posts says quite blatantly that I cheat and yet someone else thinks it is well thought out and is complementing me is either a testament to me actually putting serious thought and effort into it, a sign that I can lie my ass off well enough nobody knows I am doing it, or a stealth troll...
I just figured chemical_art was drunk again.

Well, given that champions have been fingered as one of the principal balance issues, how do we tackle it? The problems are, I think, threefold:
Yea, I agree with basically all of those three points.

I saw champions as partly a thing for new players trying to get into MP with just a champion, and I saw them as partly something for players to use specifically during the times when they don't have the resources to do much (refleeting), so overall I was basically trying to be nice. 

When it comes to game design I don't have a lot of practice being nice ;)

So some fairly big changes are probably necessary to reign in the balance impact of champions.  I don't think an m+c cost will be one of them (defeats one of the core goals: something to do while refleeting), and probably not e either just because of the challenges of a champion-only player's unit pulling from a main player's energy, etc.  But I have an idea for making dying and respawning more of something you want to avoid, though it may conflict with the "fast travel system" you get from just warping home (i.e. scrapping it).

Beyond that, the nebula rewards (m+c/e and ships) just need a nerf.  The ship stats will probably stay the same because I don't want to change the intra-nebula balance, but they'll probably be made to cost more and have a lower cap for the humans.  We'll see.

And finally, yea, some kind of offensive nemesis component.  I don't think having it hunt your champ whenever you leave your front yard is the right way to go, but it does need to hit back with more than just the +20%-to-most-AI-responses it currently get (that could be increased, but I think there are more interesting things to try first).
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!