Less focus on AI, more focus on multiplayer...PLEASE!!!!!!!!

After playing 3 games so far I haven't had much fun, this coming from someone who loved Supreme Commander 1 & 2 and TA plus many other RTS. The game feels heartless, like I am not really controlling anything at all. The graphics are right now i would say a 3 tops 4 Music is great alas it's the first thing i turn off Sounds I guess a 6 they literally sound generic. Gameplay 2 I could almost sleep through this game (constantly trying to see units within trees is a pain in the arse) Currently I would have to meta critic this game as a 3.5 I really hope there is a mother of all patch before launch. I will continue to play, and up date this post as I go. Just posting this from the eyes of a new player so please try not to be to critical.

Id really love to see a happening PVE scene of players teaming up to fight AI hordes in huge 40 player Vs AI galactic war games.

AI is critical to this game.

The apparent lack of progress on the game right now is probably because they got multiple people working on multiple thing things near completion, that they just aren't telling us about (I hope anyway)

What i think Uber really should do to put threads like this to rest (at least partially) is do another livestream, and tell us all where they are, where all those features are in development and in general, how are they doing. People have gotten so used to them sharing everything, they've been taking it for granted so now that they are a bit quieter, people are lost.

I have a feeling there are a lot of big changes either done or almost done, and they're waiting to output everything at once rather than trickle things in here or there to see how it all works together.

First part of your statement is correct - there's always a lot of work in progress. But we don't really trickle things in. Our build process is setup such that, once a big new change is ready, we integrate it into our main testing environment, put it through the paces, then release it. But it's usually on an individual feature/system basis, not a large hunk of them.

We ship things when we think they're ready, at least from a first pass perspective, or in the case of ongoing iteration, when it's actually accomplished the goal of improving the feature in question.

I agreed.

*nodnod* So I was close. Sorta. =p

You do know this is a Beta, right?

Do the graphics really bother you or do you just don't like the style? Yes, the game is a bit dull, but then again we are missing certain parts of the game that still aren't finnished.

BTW the 300k concurrent number as a failure is hilarious. If we had 300k concurrent that would put us amongst the most elite successful games in the world. That's seriously about 100x more successful than TA or SupCom. A game that has that kind of concurrency can generate 100's of millions of dollars a year.

That's why you have beta testers?

Really, balancing with AI is the worst idea possible. I never see an AI remotely half-decent.

You should read this article related to the subject.
http://aigamedev.com/open/interview/evolution-in-cityconquest/

Getting valid balance feedback from an AI is just a matter of making the AI good enough at the game.
Of course that isn't exactly easy for an RTS such as PA.

I said in my above post "before launch" so yes I know it's a beta. MWO fans used that excuse for months on end also and then launch and nothing . Dull indeed adding more to the game at this point would just make it a boring game with more options so no. Graphics do matter yes, why would you want a game where you move red and black pieces on a flat board? then go play checkers.

If you have a working AI you don't have to play with other people to quickly test stuff out. I doubt they use it to fine tune the game.

That must explain why nuke have bigger blast radius than anti nuke coverage...

I hope your AI gets to the point where I have to tone it down to win that would be amazing. I can get nuclear in under 15 though.

I guess this may have been misinterpreted : "....And honestly, a robust AI is one of the best ways for us to automate testing to work on MP quality."

I was gonna say.. That's a bit of a leap of logic to take my statement to mean we use it for balance. ;-)

Ok sorry

But I'm not the only who interpreted it like that

A member of the old guard here. Just have to clarify that competitive TA was fantastic, in spite of a bit buggy (I'm looking at you pelicans and moho metal makers...) and kind of unbalanced. So when gnug315 went air first as core on GoW (vs TAG_Rock ? not sure, but top player) and won we all had reevaluate what we though we knew about tactics.

edit: (I might be messing up details here, maybe rock was the one who went core? any other old timers around?)

Incredible depth in that game, much thanks to the fact that it was so well made. When players discover stuff devs couldn't dream of you know the craftsmanship was at an extraordinary level, and after a few years TA was full of stuff like that. Gnug-bombing, the hawk bounce, how useful t1 anti air was vs land everything if you had large enough packs (hello MT forests and samson hordes on PD) just to name a few.

To stay a bit more on topic i gotta tell how TA was a bit divided too, just like PA is. And i think this thread bears witness to that. Roughly there where two types of games to be found online in TA. One kind was with strict rules about build time (no rush i think it's called now, when players build without attacking) and often excluding units, like nukes or air. Often played on large resource rich maps. The other type was the one i enjoyed most; no rules, small maps, lots of action. The way it was played in competitions.

I think PA is going in the right direction when it's taking care of both kinds of players and i don't think it would have made it through kickstarter if uber had chosen to ignore those who enjoy to play just to watch stuff go boom and only focus on competitive PVP. As PA is coming closer to finalisation, i think it's more and more obvious that there's room for all playstyles in this game.