Fat simulation issue

sovaaa

Hi, I have been working on setting up the arm sim and ran into a problem.
I am still new to Ziva, been using it for a few weeks and trying to learn and understand as much as I can.

Anyway here is the issue. When I run a fat/skin sim I get a strange indentation in the skin mesh (shown in the image), but only around this area of the mesh. I ran a fat sim on the same model but different muscle sim/animation previously. The model stayed the same I just adjusted the muscle mesh volume and adjusted a bit the position of the muscles before cashing out a different sim where the arm gets bent a bit more to test out how the elbow area is going to behave trying to get a good result.

I have tried to find what is causing the issue by changing the fat materials, turning on and off different attachments and bone geo. I am probably missing something so hopefully someone can take a look at the file and help with resolving the issue. Thanks in advance.

Here is the file.

arm-fatsim-updatedmuscle-v2-002a.zip

4MB

dreams

Not going to open the file, I am too busy simming my own work and it would take me too long to debug it even if I wasnt busy but here are some quick questions

Do you have self intersections on any bone/muscle at any frame? You must deal with this if you do

What is your solver?
What is your integrator?

I find these solver / integrator pairs don't equal the same result

Have you tried increasing substeps? What is your current substep?

What is your collision point spacing? Have you considered lowering this?
What is your max newton iterations? Have you considered increasing this?

Have you considered putting selective sliding attachments on that indented area?

sovaaa

Thx for the reply.

I guess I should have provided more screenshots of the settings. I thought the file might be easier to look at. Anyway to answer your questions.

I used Direct solver with BackwardEuler for integrator.

I tried going to 8 substeps and increasing max newton iterations to 12. I even put the collision point spacing to 0.001 (not sure if i can go lower than this). None of these settings did anything I haven't even remotely noticed any better behavior in that area so I didn't bother putting some even bigger values. I have no idea how high with these numbers people go in their final sims.

At the start of the sim I have 0 intersections of bones and muscles geo. Once the sim starts running on the 1st frame they start to appear so I suppose that is the major issue. Thing is they appear more less all over the place but only that area is problematic. I tried painting both fixed and sliding attachments with different settings but to no avail.

I did run the fat sim together with the muscle sim and it worked fine. I stumbled across this information in another thread. Even though this is working it is not the most optimal way because the sim time increases but it is manageable for now.

vpavkov

you had your muscle and bones sim checked for self penetration and stuff like that? To me, looks like some issue of that nature...
You want to have muscle sim without any problems before you get into tissue sim. Sometimes you can get away with it, but in general is better to avoid it.
You have scripts that go through your timeline and alembic, and check for stuff like that.. You can apply deltamush, or fix it manually per frame with average vertices, if you cant fix it in sim itself. But i am pretty sure that your problem is connected to that.

dreams

vpavkov

How are you properly using Deltamush on bones and muscles to solve self intersection?

For me, I am only using it on completed, simmed fat mesh. If I have still a very small number of areas on the simulated fat mesh creating indents and hole I paint select those areas, create the deltamush or multiple deltamush, and key the envelope only on the bad frames. What works for me is to pretty much only set the iterations and smoothing step and then set "Displacement" to 0. I will do this deltamush after I've pretty much maxed out my settings and done a long, quality sim.

I don't know if this is correct use of deltamush, though.