[ODE] Fwd: Patches (Sept 2006)

[David Walters]

This was my reply, didn't know you wanted this to go to the main list :-)


---------- Forwarded message ----------
From: David Walters <hidden.asbestos at googlemail.com>
Date: 13-Sep-2006 14:04
Subject: Re: [ODE] Patches (Sept 2006)
To: Remi Ricard <remi.ricard at simlog.com>


Hi,

> First of all thanks for committing the patch. I will be easier for me to
> stay in sync.

No problem, thank you for submitting them!


>
>
> Second, It is a little bit off topic but
>
> The patch about the assert for non-zero mass center.
> Should ODE be able to work with a center of mass not centered at the
> frame body ?

I'm not sure - it seems like a useful thing to support but right now
it doesn't work so an assert makes sense (it could be removed at a
later date if support was there for it). Personally I'm not
experienced enough to know whether that would be a good or a bad
feature - it's definitely something worth discussing on the mailing
list.

>
> Lots of people are making the assumption that the center of mass can be
> somewhere else.

Yeah, that's why I think at this time it is very important to actually
alert the user to their misconception rather than let them struggle on
with weird results.

>
> I think the changes to be able to work with a CM not at the frame body
> will be only in the step, quickstep and faststep functions.
> I started looking at that but did not finish. (Some code is still
> commented out in my sandbox)

Well if you want to submit that as a patch I'm sure it'll get looked
at - like I say, there may be some conceptual reason why this would be
a bad thing. I think an open discussion on it is the best way to
discover that!

Regards,
Dave.