The early game

[TechSY730]

Yes, something that really needs to be done is to sit down and quantitatively figure ou the pacing of the game (especially AIP and player economy at each major stage of the game), both how it is now and what it should be. It doesn't have to be exact, and of course things will vary based off of playstyle, bit we need some expected ranges. Then we can start talking about how tough the game is and what kind of economy the players can be expected to have at each phase, because then we would have a reference point.

Based on your questions Keith, I take it that you don't even have a rough idea of these expectations past the early game. I this is true, I'm a little disappointed that you two have not kept kept up with this VERY important part of the game, especially because it effects how you balance things so much.

[keith.lamothe]

In my own play, it's been fine.  From other reports, it's fine.  But when reports come up that it's not fine, I ask questions

[TechSY730]

Sorry. Upon rereading my second paragraph, I realize that I was overly critical and a bit condescending, and very much underestimated you two. For that, I am sorry.  

Anyways, I'm pleased to hear that you have been keeping an eye on pacing, even if it is nothing more than your own experiences in play testing (EDIT: Which upon reading your post more closely, I see that you do look at other reports too). We'll let you know our experiences with how the game is currently paced so good decisions can be made about what can be tweaked about it, if anything

[chemical_art]

I think I completely failed to make my point - I didn't have any experience with difficulty 8, but the rate of scaling for waves between difficulty 8 and 9 is known, and they both hit mk3 at the same time (I believe?), so from there it should be relatively easy to extrapolate down with regards to the necessary components to survive at difficulty 8.   Difficulty 8 has such-and-such a ratio of ships per wave to diff 9 at such-and-such an AIP, and I'm giving an assessment of what was needed to survive the diff 9 waves, so apply the wave ratio to the knowledge requirement on diff 9 and you'll have a ballpark estimate of the kind of situation a player will be facing at diff 8.

And yeah, I know diff 9 is suicide - I've kicked it down to diff 7 while I keep trying to work through the fallen spire campaign (tried it once at diff 9 and hahahahahaha I learned my lesson).

I think I am fundamentally missing something. With all my testing, between 8.0 and 8.3 the AI jumps a whole tech higher from the very start of the game. Let me pull it up:

Here is the data from the wiki:

http://www.arcengames.com/mediawiki/index.php?title=AI_War_-_AI_Progress

"
Here is how the level of the AIs are calculated:

    * Mark IV = 1200 - ( Difficulty * 10 )
    * Mark III = 800 - ( Difficulty * 10 )
    * Mark II = 300 - ( Difficulty * 10 )

For "Technologist" type AIs, if their level would be II or higher, they are instead Mark IV. If they would be below Mark II, they will instead be Mark III.

Additionally, when playing on difficulties greater than 8, the AI players will always be one Mark level higher than normal, until they hit Mark IV. "



Meaning the 8.3 AI hits MK III at 217. The 8.0 hits MK III at 720. So the difference between the two is 503 points; a MK II nuke! 10 MK I nukes! 25 planets!

...

I never thought the difference was so great. Remember that the power increase in tech is massive despite the "slight" decrease in corresponding wave sizes / reinforcements.


In conclusion, the difference between 8.0 and 8.3 is the most massive shift for 7.0+ if that is when the AI stops getting "smarter" at 7.0. SO the difference between 8.0 and 9.0 might be...massive.






[If we are posting direct notes to the devs] This game needs quite a lot of time to be dissected by players before any major changes happen in my opinion, so I'm quite happy with letting things settle for a while and only tweaking.  Frank discussion while investigating a "what if" puzzle like this one helps dissect the game I think. I'm sure you have considered a lot of these things already, but a second group of eyes helps. In addition, I feel that it is important to point out any unusual things that I notice in my game play since I know you have so many things to check and occasionally things go unnoticed. You have all done well!

[TechSY730]

From what I understand, the AI keeps getting smarter until 7.0. From 7.0 to 8.0, the AI gets a few extra tools they start using. (For example, I don't think that EMP guardians spawn on 7.0). But yea, 8.3 and up is when the AI starts "cheating".

EDIT: And I agree that before any changes to the game that are intended to majorly (or even moderately) change around the pacing of the game need to wait until people get a clearer idea about the current pacing of the game. I think a big part of the balance complaints that have come up is because the game is now harder overall, and they are still trying to use the same tactics against the same difficulties, where now they should go down a little in difficulty to get the same level of challenge they used to.
That is not to say there aren't some largish balance issues right now (in fact, a few have already been reported and fixed), but some of the ones that deal with the game as a whole may or may not be issues at all. We will need to see how the "meta-game" for the 5.0 series evolves first.

[Sunshine!]

I'd thought that diff 8 had the jump in starting mark from mk1 to mk2, guess I was wrong, so my entire statement of extrapolating down from 9 to 8 is irrelevant.  That being said, the jump from mk2 to mk3 at 8.3 compared to 9.0 is only 7 AIP difference, but that's not all that useful of a comparison anyway, I'm guessing.

Tech: I think the inclusion of some of the new guardians (even the ones that spawn at diff 7 - gravity rarely, zombie more frequently) make the game significantly more challenging.  Thankfully zombie guardians are relatively easy to kill.  On higher difficulties, EMP guardians, more prevalent gravity guardians, warp gate guardians (am I missing one?  vampire guardians aren't so bad) cause absolute nightmares and make the jump in difficulty between diff 9 without LotS to diff 9 with LotS very significant.

[c4sc4]

(For example, I don't think that EMP guardians spawn on 7.0)

Oh yes they do, I've run into them on 7.0.

[Zeyurn]

Really the only problem really early game is a massive bomber wave.  Maybe just make bomber waves not sent in the first hour of the game if you need a quick fix (or half/third their size)?

Beyond that, at difficulty 8, I feel that if people are having trouble, maybe they need more practice at difficulty 8?  I don't think the bonuses need swapped around or anything, the point of bombers is to blow up big targets and you need something with command-grade bonuses.  I could get behind the player command stations having more hit points but it'd be to deal with them getting killed by non-bombers.  I complained a ways back right after the initial nasty wave changes with AIP but now I find it really pretty easy to deal with and I play on schizo waves in multiplayer so there's pretty much always bombers in every wave hitting any planet getting attacked.

I'm leery of making the game much easier on difficulty 8 because I know I'm not ready for 9.

[TechSY730]

(For example, I don't think that EMP guardians spawn on 7.0)

Oh yes they do, I've run into them on 7.0.

Well I know that there are some guardians that don't spawn at 7.0 that do start spawning sometime before or at 8.0. Anyone know what those are?

[Draco18s]

Really the only problem really early game is a massive bomber wave.  Maybe just make bomber waves not sent in the first hour of the game if you need a quick fix (or half/third their size)?

Not just the first hour though.  I'm getting curbstomped at 2 hours, but still "low" AIP (under 200) as each AI sends about 250 ships, and 250 bombers at that point is not feasible to hold off.

[TechSY730]

Really the only problem really early game is a massive bomber wave.  Maybe just make bomber waves not sent in the first hour of the game if you need a quick fix (or half/third their size)?

Not just the first hour though.  I'm getting curbstomped at 2 hours, but still "low" AIP (under 200) as each AI sends about 250 ships, and 250 bombers at that point is not feasible to hold off.

On what difficulty?

Also, is that a problem with bombers, a problem with their supposed counters, or with growth of wave sizes? (There will be bomber waves, so don't complain about their existence, you will have to deal with them at some point.)

Would you support that all fleet ships with a significant bonus against structural get a small wave multiplier down? (Like if the an AI chooses fighters at some AIP, and later chooses bombers at that same AIP, assuming all other factors, including random factors, the same, the ratio of the fighter wave size to bomber wave size be 1:.75 or something?) It seems a little distasteful, but due to the low mass fleet killing potential players have until later in the mid game, and the disproportional threat of ships with a structural bonus, this may be a necessary evil.

EDIT: The ship wave size factor would not only apply to "pure" non-schitzo waves, but also affect relative composition of ships in schitzo waves. Both types of waves look at this factor.

[Draco18s]

Really the only problem really early game is a massive bomber wave.  Maybe just make bomber waves not sent in the first hour of the game if you need a quick fix (or half/third their size)?

Not just the first hour though.  I'm getting curbstomped at 2 hours, but still "low" AIP (under 200) as each AI sends about 250 ships, and 250 bombers at that point is not feasible to hold off.

On what difficulty?

7.0

Also, is that a problem with bombers, a problem with their supposed counters, or with growth of wave sizes? (There will be bomber waves, so don't complain about their existence, you will have to deal with them at some point.)

A bit of all three.  Fighters not being good enough (acknowledged), wave sizes being perhaps too large (acknowledged), and bombers being good against 90% of a player's defenses (under discussion).

Would you support that all fleet ships with a significant bonus against structural get a small wave multiplier down? (Like if the an AI chooses fighters at some AIP, and later chooses bombers at that same AIP, assuming all other factors, including random factors, the same, the ratio of the fighter wave size to bomber wave size be 1:.75 or something?) It seems a little distasteful, but due to the low mass fleet killing potential players have until later in the mid game, and the disproportional threat of ships with a structural bonus, this may be a necessary evil.

Might be a step in the right direction.

[Zeyurn]

I guess I'm not sure how 250 mk1 bombers isn't feasible to hold off two hours in?

Are you using laser/missile/lightning turrets?  Fighters?  Mk 2 forcefields?  Riot starships specializing in engine damage?

I actually would kind of support the small wave down for structural ships but weighted for lower difficulties probably.

[Draco18s]

Are you using laser/missile/lightning turrets?  Fighters?  Mk 2 forcefields?  Riot starships specializing in engine damage?

Lasers, yes, but they don't have a bonus against polycrystal.
Missile, yes.  Usually 10 per wormhole.
Lightning, no.  Their real bonus is against things with refractive hulls, so their cost isn't justified against bombers.
Fighters, yes.  Up to a full set of Mk1 and Mk2.
Mk2 Forcefields, no.  Bombers eat them only seconds slower than they do Mk1s.  Its not worth the KP cost early game.
Riots, sometimes.  Generally not early game due to their cost (120,000 crystal each, no modules and something like 5000 KP).

Also: what riot isn't specialized in engine damage?

[Zeyurn]

I don't know what to tell you then except that with the exception of riots, that's what I use (with tractors obviously) and can fend off waves triple that size by myself with the lower caps of a 3-player multi game :/  I rarely ever even research fighters until way late.  I don't let them attack my homeworld if at all possible so they're on a planet with a logistics station, and I try to make sure they can only use one wormhole to come in but it's not critical that that happens.

You're probably underestimating how good lightning can actually do against a swarm of ships.  Also I'm almost certain laser had 2.4 bonus versus polycrystal last I checked?  Did that change or am I crazy?

The main thing though is that you're going to need to spend some non-fleet research towards countering bombers even if it's just higher level tractor turrets.

EDIT:  I do not understand why I must always type polynomial instead of polycrystal.  Argh.