[ODE] solving penetrations "statically"

[Anton Shekhovtsov]

I applied the patch, and noticed these problems:
 * the modification does not help to resolve penetrations: resting bodies
sink into the ground depending on timestep
 * the oscillation of resting bodies becomes bigger and causes them to move
around


----- Original Message -----
From: "Andres James" <andres.james+ode at gmail.com>
To: <ode at q12.org>
Sent: Saturday, June 24, 2006 3:07
Subject: [ODE] solving penetrations "statically"


> Here's a test modification to ODE using Erin's "Split PGS" idea to
> resolve penetrations more gracefully, demo included:
>
> http://andres.ediy.co.nz/files/test_split-sor.zip
>
> It seems to work okay, but I'm not certain my changes are correct...
> Also the final position update is just hacked in, it needs to be
> properly integrated with dxStepBody (as it currently doesn't apply any
> rotational error correction at all).
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode