AI War 2
Too many fleets in a system increases AI progress?
Sooo what am I supposed to do then? Win the game using only two fleets? Surely this makes capturing new planets to get more ships much less valuable.

This is massively inconvenient. Its not just when attacking; I have to be careful when I move my fleets around the map through suppressed AI worlds.
แก้ไขล่าสุดโดย Ecthelon; 22 เม.ย. 2019 @ 5: 02am
< >
กำลังแสดง 1-15 จาก 24 ความเห็น
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 9: 16am 
Yeah, I'm not real thrilled with this implementation either, but we put this in as kind of a proof of concept first pass.

Basically the idea is that if you're putting all your fleets all at one place all the time, that's hugely a bad thing to do. The AI will notice and get angry, which seems to be the appropriate response.

In general we're giving you waaaaay more tools now, but the first thing a lot of people will do is try to ball those up in one system and that's something that we're trying to discourage in order to get people to use the right tool for each job.

"Ball of death of all my ships" has been a problem since the very start of AI War 1.0, and with fleets giving you so many more ships now that problem is not just a matter of "there's no strategy in that," but also "you'll make your simulation crawl to a stop while also being an unstoppable force" if you do that.

So we're trying to encourage you to spread your forces out, aka not have so darn many on one planet at once. Perhaps the number of fleets on an AI planet is too harsh at the moment though, I dunno. Thoughts are very welcome on this, I'm not super happy with this mechanic but it was a first pass idea.

You should be trying to defend your backdoor and so on anyway, and so having extra fleets to defend places that you capture or to distract the warden and do a couple of strikes at once is supposed to be a thing.

Beyond that... capturing every planet to get every fleet upgrade is also supposed to still be a Bad Idea as in the first game, but now you get way more choices all over the map on what to capture, so it's still a matter of going for the ones you want. You should be NOT capturing 80% of what is available to you to capture, or the AI will curb-stomp you. It's tempting, but the idea is to take what is most interesting and useful to you specifically.

It's trickier in the new build of the game because there are so many more tempting things now compared to in the past when it might just be some metal or whatever. We have a lot of interface work later to do to make this all clear, but for now that's not there.

As far as the anti-fleetballing... that's a big question mark on how best to handle that. And it might be too harsh right now.
I for one wound up increasing the AIP by 6 or 7 over my game by not noticing the notification for "too many fleets" a couple times, which is really not ideal.
A solution that punishes you in the short term rather than the long term would be preferable to me. Maybe it could bring additional special AI forces to back up that world?

If this system does remain then perhaps the player should be rewarded for attacking on multiple fronts simultaneously.
Perhaps it could be switched to a soft response "Too many fleets detected! The AI is spawning Threat ships" for a smaller number of fleets and a hard response "The AIP is going up now" for a large number of fleets. And the limits are different based on how strong the enemy forces on the planet are (or perhaps the planet mark level).
แก้ไขล่าสุดโดย themouthofsauron; 22 เม.ย. 2019 @ 9: 50am
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 9: 50am 
I generally prefer the carrot to the stick, and I also generally prefer things that don't put me on a timer, even temporarily. So yeah this current model bothers me for multiple reasons.

Rewarding you for striking on multiple fronts I like a lot. Also rewarding you for spreading out your battelstations well I also like. I dunno.

Maybe energy is more scarce, and you get an energy discount by not having too many things on a planet? Or something along those lines, which is a lot less scary. And energy management right now is basically a joke.
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 9: 52am 
I also do like the idea of a soft response in threat, although it still won't really help with cases where players try to do something like stack 8 citadels on one chokepoint world, thus ensuring they can never lose.

That's the bad side to the flexibiltiy that I introduced with the fleets, is that now you can put them way too close together and be far too effective with that. I wonder if the economy, rather than the AI, isn't more the way to go in terms of carroting people out of doing that.
Too many fleets on one world decreases ship speed and fire rate?
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 10: 57am 
Something like that, but I'd like to make it a carrot rather than a stick if possible. Ala this: https://www.psychologyofgames.com/2010/03/framing-and-world-of-warcrafts-rest-system/

That's always been such a killer story about the beta from WoW.
The other problem with the multiple fleets penalty is that it makes un-upgraded fleets completely worthless. In the later game a fleet with low mark ships can't really do anything alone and now they can't even be used to help in a fight as utility.
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 11: 12am 
Yeah, I hate that also. It really makes me (as a player) try to mentally balance something I don't want to be thinking about in the first place.
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 11: 21am 
Coming tonight for 0.853:

* Got rid of the "too many fleets at a planet" logic entirely, at least the current version.
** It simply felt punitive and un-fun, and there was no version of that style which was likely to be fun. Nonetheless, we still just commented it out for now so someone can build something on that if desired in the future. We are also going to be working on some things to prevent players going absolutely nuts with how they stack things, but they won't be so opaque or so likely to unexpectedly bite you.
Kelp  [ผู้พัฒนา] 22 เม.ย. 2019 @ 11: 58am 
Perhaps a spawning of an attack wave would start when enough fleets get on a planet (say a 10~ or so minute timer, decreases faster based on # of fleets there), which spawns from the planet. Once spawning, it splits up evenly among all fleets you had there, allowing you to 'split' up the attacking force?

I 'unno, just throwing out a random idea I had.
x-4000 (Chris McElligott Park)  [ผู้พัฒนา] 22 เม.ย. 2019 @ 3: 12pm 
The main challenges usually boil down to how to communicate to what is happening and why to the players, how to not make them feel punched in the face for daring to play a certain way, and yet how to keep them from going to extremes that break the game.

I'm wary of anything that involves the AI too much (in terms of then spawning a threat or wave force), because the best players will just use that as further fodder for salvage, or further tank their framerates by getting into insane amounts of AI forces.

I prefer something mechanical -- in Chess only one piece per game tile of course (but look what happened to Civ 5 with that idea) -- or something tactical at the highest levels of play -- ok in Risk you can put all your armies in Australia, but is that really particularly smart most of the time?

Ideally a bit of both, but I need to think on it more. And ideally something that psychologically feels like a reward for being efficient or stealthy or whatever the case is at the time, versus something that feels like a penalty for making a certain choice.

It's a really hard question, so we just did a basic first idea and naturally it was hated. ;) That's OK. I think the answer lies somewhere in metal or energy, or in giving some sort of bonus to fleets for operating on their own, etc.

I like how in Dead Cells there are the contradictory goals of "kill 30 things taking no hits" and "clear the first level in 2 minutes" in order to get a bonus. The chances of getting both at once are almost nil. But both require a different opposite playstyle and reward you for doing it well. Quite often you get neither goal but think "darn, that was my fault but I was close and I'll get it next time!"

It affects how players play even if they aren't getting the rewards most of the time. That sort of thing is just really tight design if it can be pulled off.
Maybe the AI could respond to the "death ball" tactic by using warheads?

actually, have warheads been implemented yet?
Kelp  [ผู้พัฒนา] 22 เม.ย. 2019 @ 3: 31pm 
In that case, how about we bring back the gate mechanic from the first game, but put a timer on them?

Aka, Killing all C tier gates require you to kill them all within 10 minutes, all B within 5, and so on, and so on?

In the case that you didn't, it would still fall, but with an AIP increase for each gate you failed to kill in that group.

To stop the 'neuter the planet before doing the objective' problem, start the gate timer on system entry?
แก้ไขล่าสุดโดย Kelp; 22 เม.ย. 2019 @ 3: 33pm
< >
กำลังแสดง 1-15 จาก 24 ความเห็น
ต่อหน้า: 1530 50

วันที่โพสต์: 22 เม.ย. 2019 @ 4: 58am
โพสต์: 24