[ODE] ODE's speed

[Troy Chard]

Adam,
I'm curious, with ODE can we do something like that using fast step?
I only tried faststep once briefly and will be investigating it further soon.

I did some experimenting with boxes as dominoes.

It's a really fun test to check ode's island limitations.
So far (again only tried once or twice) I can only get about 8 of them to
fall before things get slow.  (i'll definitely be revisiting that one with 
fast step :)

Troy



At 07:26 AM 8/16/03 +0100, you wrote:
>Thomas Grimes wrote:
>>I was just wondering why, like for example, when you stack so many boxes 
>>(10 for me) that any more after that the simulation will slow WAY down 
>>and every other box you stack it just goes slower.
>
>See section 11.3 of the ODE manual for the 'why' and section
>8 for a cure.
>
> > keynotes from id software at
>>QuakeCon 2003 today and they mention their physics engine can stack at 
>>least 100 boxes [..]
> > They
>>mention that no other phsyics engine can do this without going at snail 
>>speed.
>
>Hmm, then either they're showboating or, more likely, they
>haven't kept abreast of the technology in this area because of
>the prevailant insular NIH-ism at id.
>
>I'd say that most if not all of the game-targetting physics
>engines (including ODE) offer an iterative solver at this point,
>which would have no speed problem stacking 100 boxes.
>
>--Adam
>--
>Adam D. Moss   . ,,^^   adam@gimp.org   http://www.foxbox.org/   co:3
>"I am NOT a nut!  I am the keeper of the seven universal truths!"
>
>_______________________________________________
>ODE mailing list
>ODE@q12.org
>http://q12.org/mailman/listinfo/ode