Procedural animations: presentation of concepts

[Julius]

Awesome presentation on procedural animations development (and responsive game animation in general):

http://www.gdcvault.com/play/1020583/An ... e-Approach

Would love to see an open-source implementation of that in the good old quake engine

[toneddu2000]

Forget about it. I would be happy enough to understand how works skel_build() in Darkplaces! We can of course hope that some day will arrive a brave new programmer who introduces next gen features like skeletal interaction with physics and skeletal attaching features (camera, weapons, etc.)

EDIT: something like this and that

[leileilol]

EDIT: something like this

[mankrip]

EDIT: something like this

@ 10:35

[hogsy]

Oddly enough we were planning on looking into doing something like this after OpenKatana was finished, only to a more extreme level that would be more comparable to what was done in JP:Trespasser.

[toneddu2000]

only to a more extreme level that would be more comparable to what was done in JP:Trespasser.

My favourite game ever. Same realism on skeletal model/damage inflicted I've only seen in Skyrim and Ryse: son of Rome. If you should do something like that on a open source framework I'd be the first to sponsor you

[leileilol]

My point is, 'revolutionary new tech' is not a solution for game design issues and shoehorning new things doesn't fix games.

[toneddu2000]

completely agree leilei, the problem is: "if you want to create a game in 2014 that wants to have a(any kind of)audience, you MUST obey to some graphics/physics nowadays standards....Or you can do MineCraft (which, after 1 year, had mods that emulated those graphics/physics nowadays standards...)"
After that you can start thinking at your supercoolmagicfaboulous game(the gameplay I mean), but prerequisites remain.

[Julius]

Yes, this isn't really revolutional technology but rather clever use of exisiting one to lower the workload to create nice looking animations.

[mankrip]

On a semi-related note, I hate how "procedural" became a buzzword for any gimmicky feature nowadays.

My point is, 'revolutionary new tech' is not a solution for game design issues and shoehorning new things doesn't fix games.

Expanding upon that: It could work, but only if properly done. New features should be implemented in a way that they doesn't feel out of place when implemented.

In Quake's case, almost all models would have to be recreated, along with all of the code for the animations, and much of the physics code. Because even if the gameplay don't change, the physics code would need to generate extra data for the animations. Otherwise, there's no way to make it work that well.

So, it isn't just a matter of implementing such a tech in the engine. The design of the game would have to be studied in order to create replacement content that would work well with it. Otherwise, it's pointless.

[toneddu2000]

In Quake's case, almost all models would have to be recreated, along with all of the code for the animations, and much of the physics code. Because even if the gameplay don't change, the physics code would need to generate extra data for the animations. Otherwise, there's no way to make it work that well.

I'm trying to use procedural animations with DP and IQM and, believe me my friend, it's so frighteningly difficult that recreate all bosses/enemies/player/items QUake models in Skeletal ones would be for me a breeze!