[ODE] XODE - should angles be in degrees or radians?

[DjArcas]

well, everyone else on the list has put their oar in, so I will.

It's supposed to be a human readable format, so put it in degrees, which are
nicely human readable.

Now, onto my simple car problem... ;)

Adam

----- Original Message ----- 
From: "Jon Watte" <hplus-ode at mindcontrol.org>
To: "Bob Dowland" <Bob.Dowland at blue52.co.uk>; <ode at q12.org>
Sent: Friday, March 19, 2004 5:00 PM
Subject: RE: [ODE] XODE - should angles be in degrees or radians?


>
> > But, not being able to store things like right angles precisely,
> > would seem like a bit of a showstopper for a geom/phys file format.
>
> I believe this argument is a fallacy. The reason is that ODE, internally,
> uses radians, so even if you can "store" a 90 degree angle precisely, as
> soon as you "load" that into ODE, it's no longer exact.
>
> I think that replicating exactly what's inside the scene is more
important.
> I think basics for a file format should include being able to stop a
scene,
> take a checkpoint, play out the scene, then restore the checkpoint and
play
> out the scene again, and get no divergence.
>
> The only way this is going to happen is if you don't introduce arbitrary
> conversion between internal formats and the stored format. You can print a
> floating-point number in a format that will exactly replicate after load
> for all values. However, you can't make a conversion to degrees, and then
> back to radians, exactly replicate after load for all values.
>
> Thus, it seems to me that the "right angle" argument is moot (because you
> lose that right angle as soon as you actually load the file) while the
> "perfect persistency" argument seems to have some weight.
>
> Cheers,
>
> / h+
>
>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>