toadstorm Posted August 29, 2013 Share Posted August 29, 2013 (edited) I just can't seem to convince the bullet solver not to explode. I have some geometry that I've pre-shattered in SOPs, no clusters or anything, and I'm trying everything I can do run the sim through bullet to no avail. I want to use all the nice glue network constraints that RBD can't do, but I have no idea how to prevent bullet from exploding. I've searched through several threads saying that I need to make sure all my pieces are convex hulls, but I don't know how to iteratively generate those. When I look at the collision geometry generated by the bullet solver, nothing seems to be overlapping, but inevitably a few tiny pieces shoot out of the simulation and then everything explodes within 2 or 3 frames of the simulation starting. I've tried increasing substeps and collision padding, and nothing changes. What can I do to prevent this from happening? I'd even suffer through RBD slowness if I could have fine control over gluing objects together, but I can't find any information on how to do that. I'm attaching a .HIP if anyone wants to take a look and tell me what I'm doing wrong. Thanks in advance. Using Houdini 12.5.469. shatter.zip Edited August 29, 2013 by galakgorr Quote Link to comment Share on other sites More sharing options...
cwhite Posted August 30, 2013 Share Posted August 30, 2013 Looks like the Drag Force may be causing the problems - if you turn it off then everything looks fine. Quote Link to comment Share on other sites More sharing options...
toadstorm Posted August 30, 2013 Author Share Posted August 30, 2013 Looks like the Drag Force may be causing the problems - if you turn it off then everything looks fine. This is so weird. What about the Drag Force would cause such a strong reaction? Quote Link to comment Share on other sites More sharing options...
cwhite Posted August 30, 2013 Share Posted August 30, 2013 I've seen a few cases where Bullet doesn't deal with rapidly-changing forces very well, although that's usually been with much stronger drag forces than the one in your scene. In this case, you probably want to try the Ignore Mass setting, which will cause all of the objects to decelerate at the same rate regardless of mass. In your original scene, it looked like the smaller pieces were tending to be the ones flying off. Quote Link to comment Share on other sites More sharing options...
JuriBryan Posted August 30, 2013 Share Posted August 30, 2013 yeah drag forces are kinda strange in the latest versions... Also in pops, when you add some mass to particles and have a single drag in there even with a scale of 0.05 no particles will move at all. and in Dops, well it just kinda blows stuff up.... Could be useful to create explosions when the drag force is turned on only for a frame or two... Quote Link to comment Share on other sites More sharing options...
whodini Posted October 17, 2013 Share Posted October 17, 2013 I see the same strange issues with the drag force when working in 12.5. I had to scale my entire scene up by 10 to 100 times greater to get more predictable results. Quote Link to comment Share on other sites More sharing options...
rayman Posted October 18, 2013 Share Posted October 18, 2013 You can always use vop force/script solver/rbd state to recreate proper drag behaviour. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.