Bullet suitability for simulating system with many gears?

brouhaha
Posts: 3
Joined: Fri Aug 24, 2007 12:17 am

Bullet suitability for simulating system with many gears?

Post by brouhaha »

Is Bullet suitable for simulation of a complex mechanical device with hundreds of spur gears, cams, axles, etc.? I've designed the parts in SolidWorks, and might be able to simulate them in COSMOS, but haven't tried yet. I'd rather export the parts and simulate them with an open source package, so that I can make the simulation publicly available for others to experiment with.

Because of the nature of the device, I can statically determine which components need to be checked for collisions. Is it easy to get Bullet to take advantage of that, to avoid unnecessary collision detection?

Thanks!
Eric
User avatar
Erwin Coumans
Site Admin
Posts: 4221
Joined: Sun Jun 26, 2005 6:43 pm
Location: California, USA

Re: Bullet suitability for simulating system with many gears

Post by Erwin Coumans »

brouhaha wrote:Is Bullet suitable for simulation of a complex mechanical device with hundreds of spur gears, cams, axles, etc.? I've designed the parts in SolidWorks, and might be able to simulate them in COSMOS, but haven't tried yet. I'd rather export the parts and simulate them with an open source package, so that I can make the simulation publicly available for others to experiment with.

Because of the nature of the device, I can statically determine which components need to be checked for collisions. Is it easy to get Bullet to take advantage of that, to avoid unnecessary collision detection?

Thanks!
Eric
It all depends. Bullet has not been designed for such purpose, but it can simulate a small number of gears in real-time under specific conditions. Many settings can be tunes, like number of iterations, collision margin and so on. I would first focus to see if you can get the quality you want from Bullet, and then deal with performance. Yes, you can add your own optimizations, but it might very well be that Bullet's built-in optimizations are sufficient.

Thanks,
Erwin