Jump to content
  • Advertisement
Sign in to follow this  

Physic paper...

This topic is 4675 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Never came across this one, it's fairly new so you may not have too... http://www.research.rutgers.edu/~kaufman/ffdfrb.html

Share this post


Link to post
Share on other sites
Advertisement
Quote:
Original post by b34r
Never came across this one, it's fairly new so you may not have too...

http://www.research.rutgers.edu/~kaufman/ffdfrb.html


Haven't seen that one before: motion looks reasonable*.
Verlet (symplectic) integrator, impulse-based, anisotropic friction, QP approximations, conservation of momentum, not enforcing complete rigidity. Many interesting ideas...

*the chess pieces have unrealistically high damping/friction and should splay out much more (probably easily fixed be reducing damping/friction). The other videos look good.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Quote:
Original post by MrRowl
Highly unphysical IIRC...

How so?
will this be less physical that jlib?
Are they setting the mass to infinity after few sequential passes like in the shock way propagation algorithm?
I find setting masses to infinity aproach an extreme abomination, yet it has been accepted around here as a brand new law of physics that replaces the 300 hundred years old first, second and tird law.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
Quote:
Original post by John Schultz
Is the last law related to this book?

lol, I have to say that was very funny.
Now it is a lot clearer where you are getting those brand new laws physics of yours.

Share this post


Link to post
Share on other sites
Quote:
Original post by Anonymous Poster
Quote:
Original post by John Schultz
Is the last law related to this book?

lol, I have to say that was very funny.
Now it is a lot clearer where you are getting those brand new laws physics of yours.


I don't use that law in my simulations; I try to keep fudging to a minimum.
However, a popular cartoon character who says "Howdy ho!" might make an appealing physically simulated character using that law (though IP rights would have to be negotiated with Trey Parker and Matt Stone). A technical challenge would be related to CPU issues (mostly PU).

Share this post


Link to post
Share on other sites
Quote:
Original post by Anonymous Poster
Quote:
Original post by MrRowl
Highly unphysical IIRC...

How so?
will this be less physical that jlib?
Are they setting the mass to infinity after few sequential passes like in the shock way propagation algorithm?


I think, effectively yes - but after only ONE pass. However, I'm busy at the moment, so don't have time to read the details again... if I'm wrong someone please tell me!

Quote:

I find setting masses to infinity aproach an extreme abomination, yet it has been accepted around here...


Depends what you mean as "accepted". Nobody would claim that it gives highly accurate results although if implemented/used carefully can give results that are, to the eye, as good as the "real thing". Remember, this is a gamedev forum - not a pure physics sim forum.

Also, when using an iterative solver where you trade speed for accuracy, certain situations will in practice (e.g. very heavy object sitting on a light one) result in the system/constraints not being solved completely each timestep. The "shock step" is there only as a safety net for this case - if you don't use it (and you're terminating the iterative solver early), then you'll get interpenetrating objects, which is arguably just as, if not more, visibly "unphysical" as the results of using a "shock step".

Anyway, I thought you'd said in another thread you weren't going to post here any more?!? Welcome back, old friend :)


Share this post


Link to post
Share on other sites
thanks for the paper,

I missed this one on my travels, it does seem to have some interesting concepts that I need to evaluate before continuing with the non-convex RBWS paper.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!