Man, that dogfighting thing they are doing now… Wouldn’t it be great, if they just implemented it into KCD for creating online multiplayer mode there? Like knights tournament, maybe castle siege with two playable teams…
@213: C’mon didn’t you learn your lesson yet? I think you didn’t watch the video at all, so basically the dogfighting implements some seriously needed technology called AI behavior. If you think that you don’t need AI, then please go playing some MMOs, the AI consists only of pathfinding and you have lots of people you can flood with your cynicism and non-interesting talk.
And since you didn’t mention the part that I would personally disagree with @kdoman then it’s the following: The dogfighting is not an online multiplayer - only feature and implementing online modes is dependent on how you want to do it. But this is not exactly what both companies want to share in the expertise. More important is the layered clothing system for Star Citizen and the global AI tracking for KC: D. Those were the main reasons as far as I know, @Hellboy might correct me if I’m wrong with that information.
first of all, ai dogfighting in space has nothing to do with ai in a medieval game, secondly, they already have people from charles university working on the ai specific to what kcd is about. so why do you think kcd doesn’t have “ai behavior” already?
Good one. I could offer some more but I’m waiting for @213 to share his knowledge. The question is not only about the separated AI systems but also the connection between those systems (which lead to AI behavior).
Just to take the two you and me just mentioned: If the AI wants to attack me and wants to move to my direction, an obstacle is blocking its way. Question is: evasion of the object to the left or the right side? Maybe the left side is not so slippy than the right side, yet the distance to me is shorter on the right side: Will I take the risk of slipping or take the safe but slightly longer path? And while deciding on where to go, the target might move around: how often will the AI “change its mind”?
More variables mean more complexity. And more AI systems mean even more complexity to calculate the result, which is the AI behavior)
PS: Do not mistake AI behavior with Behavior-Based Artificial Intelligence (BBAI).
That’s how I basically thought of it. I wasn’t serious bout that multiplayer thing, because I don’t look forward to Star Citizen that much, so I have no real idea of what they are doing right now. Thus I just talk shit to have fun and see what smart people have to say about that. I don’t even need multiplayer actually at all. I’m just interested in development of the progect in general. But as far as I know, Warhorse are nailing the AI with specialists from Karl’s University and stuff, reading lectures about it, so no need to worry about AI here. Unfortunately new Dan’s blog post have no info on working with RSI, so look forward to next month probably.
as I understand him:
They are exchanging ideas about AI. (so both got things the other can use)
Warhorse got the layer system for clothing and armor
Star citizen have got a much improved system for “attachments” than the usual one in the engine.
SC also know the engine better since they do have a direct cooperation with Crytek…
So they are also sharing “tips and tricks” (alt least that was they announced back when the news about this broke)
They don’t have a direct cooperation but the CryEngine “guru” who developed big parts of the engine and even wrote a book about developing with the engine now works for Cloud Imperium.