p.vaz Posted October 16, 2015 Posted October 16, 2015 Hi, For a car simulation, I've made some tests with JointSuspensions, but I've realized that it's not possible to keep real mass values for the bodies without getting instability. The physics car demos have still "fake" values like gravity of -19.6 or a car mass of 4Kg to get better mouvement. It would be great if you improve this. Regards, Philippe
silent Posted October 19, 2015 Posted October 19, 2015 Hi Philippe, That's why we used strange physics settings in City Traffic demo :( We are considering to add 3rd party physics engine integration later this year (or at the beginning of the 2016). Thank you! How to submit a good bug report --- FTP server for test scenes and user uploads: ftp://files.unigine.com user: upload password: 6xYkd6vLYWjpW6SN
esger.abbink_ Posted October 21, 2015 Posted October 21, 2015 Probably is not prudent at this time for you to disclose, but could you elaborate on which engine(s) is being considered? This could be very relevant for us as well.
ulf.schroeter Posted October 21, 2015 Posted October 21, 2015 I think Nvidia PhysX was named as potential candidate (see this post). Nevertheless I would doubt that usage of the Nvidia physics engine would solve above stated stability issues when using real-world mass, as also PhysX internally only using float precision for physics simulation. Maybe outdated, but for quite a long time only Havok physics engine provided optional double precision support for improved stability and more real-world physics simulation setups.
Recommended Posts