3dbeing Posted January 9, 2007 Share Posted January 9, 2007 I have had some problems with very small pieces in shattered sim... it goes like this if I add torque drag, some of the pieces end up with position, normal and velocity attributes all = -1.#IND ... if I take the drag out, no problem... I'm doing some simple tests and it seems it may have to do with low rotational stiffnes combined with torque drag.. my values are density 5000, rotational stiffness .15 and torque drag .0005 (which is the smallest number I could get to work with 10 substeps, but at 20 substeps it freaks out) has anyone else had to deal with this or know the source of the -1.#IND value in the attributes.. Thanks, Quote Link to comment Share on other sites More sharing options...
3dbeing Posted January 17, 2007 Author Share Posted January 17, 2007 it seems that below a certain mass, combined with timestep, I'm guessing the floating point blows up, somthing like, the equations take the object into the 11th dimension. But I'm sure sidefx didn't program dynamix with the M theory in mind... or did they? Quote Link to comment Share on other sites More sharing options...
old school Posted January 17, 2007 Share Posted January 17, 2007 Check to see if the small pieces have valid volumes associated with them. Both mass and collisions depend on the volume data associated with each RBD object. Quote Link to comment Share on other sites More sharing options...
3dbeing Posted January 17, 2007 Author Share Posted January 17, 2007 yes, there was volume, but the mass was... well, very small, I have since added a mass checker to my volume creator, this issue will not creep up again. Thanks. 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.