Here is a implementation sample, maybe helpful for you,but it''s not easy to read. http://cedar.intel.com/cgi-bin/ids.dll/content/content.jsp?cntKey=Generic+Editorial%3a%3aclothsample&cntType=IDS_EDITORIAL&catCode=0&path=3

Wow, someone remembers my post from November (some 8 months ago)!

Unfortunately I haven''t had enough time to invest in this problem and I probably won''t have for the next couple of months either, but I''ve already bookmarked that link and I also found a few other documents during these 8 months that will help me when I get the chance to get back to it.

It''s good to see that the discussions here are read by many people and that they take an active interest to help and find out the results.
I''ve certainly enjoyed many of them and learned a lot too.

Thanks for your input ''safe''.
Do you have experience with implicit methods? Or even with Physics in general?
Maybe we can exchange some ideas.
I''m sure ''uutee'' (Mikko Kauppila) will join in the discussion, if he is still around, as he was the main person that helped me with the original post.

Thanks again,
lusinho

Thanks for your reply,lusinho
I am a beginner with implicit. I was asked to compare it with explicit in second order euler. I just figure out that the step in implicit method can be large enough, I am trying to implement that with VC++,but it seems not easy.We need to implement implicit on a PC cluster.
We can share the results if I have any progress and welcome to discuss if you have time.

Hi fellows,

The implicit method is a great math coding practice, but I''ve found it''s not too great an enhancement for cloth simulation. Coupling with "buckling forces" of Choi ("Stable But Responsive Cloth", SIGGRAPH 2002) gives a great boostup because the structural instability is beaten. Accompanied with a good preconditioner, the implicit method should be faster than explicit methods for larger cloth models.

I wrote an article about implementing the implicit method for cloth simulation. I haven''t released it anywhere yet, so I''d be thankful if someone could "proofread" it. No images included (yet), so beware of ASCII math The article is based on a convenient, more "matrix-free" implementation of the solver.

Feel free to download it: http://www.rawhed.com/uttumuttu/temp/implicit.doc

- Mikko Kauppila