[ODE] release time

David Walters hidden.asbestos at googlemail.com
Sun Feb 11 07:06:02 MST 2007


> I've investigated that heightfield thing again, and I agree with the
> author of the optimization patch: as long as there is no good testcase
> that demonstrates the bug, there is not much to do. I contemplated
> reverting, but it's probably best to keep it as is, and see if a
> reproducible test case will emerge.

Hi, I agree with Bram that leaving the heightfield code as it is for
0.8 would probably be the best course of action - the steps forward
that have been made are beneficial and although there may be teething
problems they have only manifested themselves in one reported case.
Discounting formatting, _architecturally_ the code has changed so much
that it isn't really feasible to switch back and forth.

I have also thought that perhaps the problem doesn't lie directly with
the heightfield code, because as it relies heavily on the ray and
plane colliders, and given the huge change log list earlier, it may
actually be some complex to discover, simple to fix problem that has
crept in.  For example, I suspect that the ray - cylinder collider
isn't 100% correct - and if that is what Dmitry's truck is using for
it's wheels then this may have started to become a problem?

Finally to add, Dmitry Medvedev has written a quick mail to me to say
that he will try and create a test case for this problem when he is
able to spend some time on it - which he cannot commit to any
particular time frame for..

Anyway, looking forward to 0.8 :-)

Regards,
Dave


More information about the ODE mailing list