[ODE] Previous timestep significance? (part 1)

[Adam Moravanszky [Novodex]]

Hi,

> All these made me consider the solution, I've bumped into, as
satisfactory.
> Perhaps my experiments could help you, in case you haven't solved this
> problem already. Have you? :)

:-)  To be honest I have not worried about this problem a lot so far simply
because one needs to use constant time steps if one needs
determinism/reproducibility anyway.  I figured out that the problem goes
away if I get rid of the 1/dt term, but the 1/dt term is otherwise very
practical, so ...

But your results sound interesting, I will try to figure out why your
experiment lead to an improvement.  Just to be clear, you were talking about
ODE's the global ERP + CFM, and not the special ones like dContactSoftERP?

> If I'm right, both of these _are_ solutions. I also believe that the 2-nd
solution
> should also fix the impossibility of completely getting rid of noticeable
bouncing
> when objects bump into each other with some essential velocity. That's why
I'd
> rather prefer the 2-nd solution.

Right, thats exactly what I was trying to say with the "other problem".  And
that this second solution doesn't neccesarily solve the problem of the time
step oscillation -- at least my implementation of it doesn't.  So go and try
it and tell me that you have a better implementation than me that makes the
problem go away.

Regards,
    --  Adam