Please Make Cloth Physics!

[ensiform]

There is 1 very large glaring issue with adding cloth as a shader parameter in that you can no longer distribute the mod as working for the original game.  Because the original game's shader parser will see "cloth" and "go uhhhhh wtf is cloth", and mark the shader as defaulted.  There are a few ways around this and since this is likely something to go in a non-vanilla renderer, we can always add .mtr that will get loaded with newer shader parameters and the old .shader files are used as fallback for compatibility. (SmileTheory's renderergl2 for ioquake3 does this iirc)
 
Seems to me like it would make more sense to just fork the game and try implementing a full fledged new physics engine instead of only focusing on cloth.
 
@@Archangel35757 Neither of those implementations seem like they're even decent fit with the existing engine.

[Archangel35757]

 @@Archangel35757 Neither of those implementations seem like they're even decent fit with the existing engine.

Yes, I know... but sometimes it is useful to see examples of how others implemented cloth. The Open Dynamic Engine does have the ability for cloth... as it does have a spring-mass system; however, the ODE would need modification to add proper cloth buckling (see the papers I posted on my github request).

[eezstreet]

Let me explain how this plays out, using the analogy of a car. A physics engine is exactly that -- it's an engine. You have to pipe all the game's collision into said system, which isn't remotely feasible without a significant amount of work considering all the math that's hashed out on the side.. So you have the game engine and the physics engine. Essentially, they're both 4 cylinder engines, and you want to fuse them together to make an 8 cylinder powerhouse. That is quite the welding job, and you'll have to do a lot of welding yourself, and the exhaust and so forth, to the point where it would have made more sense to replace the engine or do other significant changes to it.

 

Listing papers is useless. We know how Google works, we know how to find what we're looking for, we don't need zillions of emails about it when only a few people are concerned about it. And we're way too early on to worry about it at this rate, especially when we have important milestones that yet need to be reached (like..a first release. Or Occulus Rift support. Or proper analog movement using joysticks. Or an OpenAL sound system that isn't hack-ridden, or OGG Vorbis support for sounds and XviD for video. Or actual wind in MP, which would be kinda important for cloth physics, would it not?)

 

For the record though, I'm just stating things as they are now.

[DT.]

Are there any plans for physics to be implemented? I only ask so I know if it's really worth it to add simulated cloth to JKA's existing animations.

[Xycaleth]

Unless someone wants to contribute their time to add it, then it doesn't look likely at the moment.

[Lamented]

That couldn't be truer.

[Grab]

Ekhm

http://jkhub.org/topic/2394-cloth-sim-experiments/

[Xycaleth]

That thread is based on baking cloth physics into the animation. There's no cloth physics involved on the game engine side of things.

[Futuza]

^^But it addresses a want that people have - even if it is just faking it.

[DT.]

It'll take a lot of time to do, but it CAN be done. I'll continue work on it then. B)