Search found 324 matches

by Erin Catto
Mon Feb 25, 2008 6:04 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Constraint Solver problem
Replies: 5
Views: 4183

Re: Constraint Solver problem

You can think of Baumgarte stabilization as a spring + damper. However, if used properly, a Baumgarte stabilization can be stiffer than a spring while being stable.
by Erin Catto
Sat Feb 16, 2008 7:05 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: accuracy of collision detection: Bullet vs. Solid
Replies: 17
Views: 12104

Re: accuracy of collision detection: Bullet vs. Solid

Erwin, thanks for the picture.

It seems that the number of local minima decreases as penetration becomes shallow. Perhaps this be exploited. Also during deep penetration the exit direction may not matter much (it won't be pretty in any case).

It is unclear how these observations carry over to 3D.
by Erin Catto
Fri Feb 15, 2008 10:08 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: accuracy of collision detection: Bullet vs. Solid
Replies: 17
Views: 12104

Re: accuracy of collision detection: Bullet vs. Solid

You can read about non-convex optimization on this page: http://www.solver.com/probconvex.htm If you want to learn more about optimization, I recommend this book: http://www.amazon.com/Numerical-Optimization-Operations-Financial-Engineering/dp/0387303030/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=120311272...
by Erin Catto
Fri Feb 15, 2008 7:40 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: accuracy of collision detection: Bullet vs. Solid
Replies: 17
Views: 12104

Re: accuracy of collision detection: Bullet vs. Solid

Please, may we discuss this problem? I'm finding fast enough incremental alghorithm and local vs global convergence is its weakness. What is your question? Like I said, I don't think this algorithm will work. It is basically a separating axis test with a local search. Local searches for the minimum...
by Erin Catto
Thu Feb 14, 2008 4:46 am
Forum: Links, Papers, Libraries, Demos, Movies, Comparisons
Topic: Physics for Phun
Replies: 9
Views: 9155

Re: Physics for Phun

Hi Kenneth, This looks really nice! A had phun playing with it. :) Will this project become open source? Will your team be publishing technical details on the implementation? What are you using for the concave collision? The stacking looks quite stable. Are you using any sort of shock propagation an...
by Erin Catto
Mon Feb 11, 2008 12:27 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Rolling friction model
Replies: 8
Views: 27102

Re: Rolling friction model

Relative damping should work ok.

However, damping wouldn't support static rolling/spinning friction. You may want a ball to stop rolling on mildly flat surfaces.
by Erin Catto
Sun Feb 10, 2008 7:55 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Rolling friction model
Replies: 8
Views: 27102

Re: Rolling friction model

Yeah, you could use relative angular damping.
by Erin Catto
Sun Feb 10, 2008 9:07 am
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Rolling friction model
Replies: 8
Views: 27102

Re: Rolling friction model

How about just using this velocity constraint: omega2 - omega1 = 0 And use a maximum torque for this constraint of: torqueMax = roll_spin_factor * choose_a_length_scale * sum_of_normal_forces. You can safely use this in all cases, even with multiple contact points. So you wouldn't need to detect rol...
by Erin Catto
Thu Feb 07, 2008 6:45 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Problem with large effective mass and SI convergence
Replies: 6
Views: 3385

Re: Problem with large effective mass and SI convergence

You can use a conjugate gradient solver (Google GPCG).

You can also form the associated LCP problem and use a pivoting solver like Lemke's method.
by Erin Catto
Wed Jan 02, 2008 10:53 pm
Forum: Links, Papers, Libraries, Demos, Movies, Comparisons
Topic: Tokamak Open Source
Replies: 7
Views: 8242

Re: Tokamak Open Source

Hi David, I'm sure many people here would be curious to know about the inner workings of Tokamak. I used the algorithm describe in this paper http://www.cs.hku.hk/research/techreps/ ... 005-01.pdf (Collision Detection of Convex Polyhedra Based on Duality Transformation). The algorithm seems sound, b...
by Erin Catto
Thu Dec 20, 2007 9:15 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: The best engine to generate close packings of spheres?
Replies: 11
Views: 5534

Re: The best engine to generate close packings of spheres?

Alessandro, I scanned over this paper. Some of the math is a bit deep and I didn't spend the time to digest it all. How would you compare this method to Bullet's sequential impulse solver? Does the CCP provide you with isotropic friction? Is that the main advantage? The paper states that the solver ...
by Erin Catto
Sat Dec 15, 2007 10:13 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Physics and Lowpassing
Replies: 8
Views: 4104

Re: Physics and Lowpassing

Is there any reason why I couldn't just save the contact points in untransformed object space for each contact, and then comparing against those when getting the new set of points, instead of the feature info? (I mean, other problem than having to compute the inverse transform of the objects for ea...
by Erin Catto
Thu Dec 13, 2007 9:08 pm
Forum: Research and development discussion about Collision Detection and Physics Simulation
Topic: Physics and Lowpassing
Replies: 8
Views: 4104

Re: Physics and Lowpassing

How are you matching contacts for warmstarting? Is the translation disrupting warm starting?

Are you warm starting friction and normal impulses?
by Erin Catto
Thu Dec 06, 2007 7:22 pm
Forum: Links, Papers, Libraries, Demos, Movies, Comparisons
Topic: Position Correction
Replies: 12
Views: 18914

Re: Position Correction

Ahh. Yeah, for cloth constraint ordering can make a big difference, especially if you use a stability factor as in Andrew Megg's cloth presentation.

https://www.cmpevents.com/GD05/a.asp?op ... essID=3931
by Erin Catto
Thu Dec 06, 2007 6:16 pm
Forum: Links, Papers, Libraries, Demos, Movies, Comparisons
Topic: Position Correction
Replies: 12
Views: 18914

Re: Position Correction

When I was studying constraint ordering, I would simulate a stack of boxes or a chain and export the MLCP to a text file and load it into Matlab. It was quite easy to try different orderings and measure convergence. You may get better results, but the best I could do was shave off one iteration. In ...