3 ms·

Straight from the oven (paper from SIGGRAPH 2023): "Multi-Layer Thick Shells" (https://youtube.com/watch?v=z1Wc5DvC2Wk https://youtube.com/watch?v=z1Wc5DvC2Wk)

Straight from the oven (paper from SIGGRAPH 2023): "Multi-Layer Thick Shells" (https://youtube.com/watch?v=z1Wc5DvC2Wk https://youtube.com/watch?v=z1Wc5DvC2Wk) Extends the FEM model to accommodate for thick shell-like structures, like yoga mats, leather, and wrinkles on matrices. Also you might want to check out the IPC papers (particularily C-IPC, since you said you're interested in cloth: https://www.youtube.com/watch?v=jvBSDXxsgRA https://www.youtube.com/watch?v=jvBSDXxsgRA). A new class of simulation methods has arisen over the years which allow you to simulate detailed collisions in an incredibly robust way (though computation time is still an issue).

guhidalg 3y ago
> though computation time is still an issue This is what I got hung up on. I was doing some independent research on real-time cloth simulation for sailboats and I reached the conclusion that the physically accurate techniques are not real-time. I was looking at simulating tensions applied at specific vertices of the sail, the stiffness of the cloth, wind pressures, etc... and I gave up on finding anything that was doing it in real-time. Am I wrong? Or is there a SIGGRAPH paper I haven't found yet :P
- pistachiopro 3y ago
  Have you looked into Small Step XPBD (introduced in the paper "Small Steps in Physics Simulation ")? It's the most efficient nonlinear dynamics integrator/solver I've come across, and luckily one of the simplest, too! (Keeping in mind that simplest nonlinear solver is a relative metric.) I've been able to simulate very stiff materials like bone by updating the sim at 6000 steps/second. The exact number of steps/second you'll need will depend on both the desired spatial resolution and physical accuracy of your sails, but I wouldn't be surprised if 6000 steps/second is more than sufficient. And computers are fast enough where you could simulate quite a few large, detailed sails at that rate in realtime.
  - guhidalg 3y ago
    Thanks for that link, I had not seen this paper before. I will add this to the backlog of things to try!

guhidalg 3y ago

> though computation time is still an issue This is what I got hung up on. I was doing some independent research on real-time cloth simulation for sailboats and I reached the conclusion that the physically accurate techniques are not real-time. I was looking at simulating tensions applied at specific vertices of the sail, the stiffness of the cloth, wind pressures, etc... and I gave up on finding anything that was doing it in real-time. Am I wrong? Or is there a SIGGRAPH paper I haven't found yet :P

pistachiopro 3y ago
Have you looked into Small Step XPBD (introduced in the paper "Small Steps in Physics Simulation ")? It's the most efficient nonlinear dynamics integrator/solver I've come across, and luckily one of the simplest, too! (Keeping in mind that simplest nonlinear solver is a relative metric.) I've been able to simulate very stiff materials like bone by updating the sim at 6000 steps/second. The exact number of steps/second you'll need will depend on both the desired spatial resolution and physical accuracy of your sails, but I wouldn't be surprised if 6000 steps/second is more than sufficient. And computers are fast enough where you could simulate quite a few large, detailed sails at that rate in realtime.
- guhidalg 3y ago
  Thanks for that link, I had not seen this paper before. I will add this to the backlog of things to try!