Lance R. Marrou (marrou++at++vsl.ist.ucf.edu)
Tue, 8 Nov 1994 18:18:52 -0500 (EST)
> Does anyone know why it is that angles are expressed
> in degrees, as opposed to radians, in Performer apps?
Because GL (and the geometry hardware) was originally designed to work
with floating point degrees rather than radians. Not sure why, though,
I never found a reference to it in the original papers.
> I'm sure there is a resonable explaination for this
> design decision (performance, implementation?), but
> it forces the user to convert degrees to radians and
> visa versa when intercomunicating angular information
> with non-Performer apps. This process of having to
> re-express angular information is inelegant and
> performance intensive. Is there a way to get Performer
> to accept radians instead?
Not directly, but it depends on what you are needing it for. For DCSes,
you can create your own matrix (without having to use the pfDCSRot()
function). You can also do the same thing for the channel view. i.e.
use pfChanViewMat() rather than pfChanView(). Thus, you never really
need to specify degrees to Performer.
Otherwise, switch to degrees and only convert when you communicate with
other processes! You can use pfSinCos to get a better speed from the
angle functions in degrees (though lose a little accuracy, but not much).
It depends on your needs.
>
> While I'm on this issue, why is it that colors are
> sometimes expressed in the 0-255 range and other
> times normalized from 0 to 1? The man pages are
> not always explicit as to which range a particular
> Performer color is to be expressed in either.
Same reason as above. If you use byte values from 0-255, the GL pipe
must convert them to a floating point value from 0 to 1 to render correctly
in RGB mode. Of course, it is not this way in color map mode.
By the way, I got this information from the papers by James Clark and also
I think Kurt Akeley mentions it in his papers. If any of this is wrong, I
hope someone corrects me. :)
>
> Kent Watsen
> DCS Corporation
^^^
Dynamic Coordinate System Corporation? Interesting name....
I wonder if there is an SCS Corp. out there? ;)
> Simulation Branch
> 703.683.8430 x369
_______________________________________________________
IST __ E-mail: marrou++at++vsl.ist.ucf.edu
Visual / / ______ /\____ ______ ______
Systems / / / _ / / __ // ____// ____/
Lab / /__ / /_/ / / / / // /___ / __/_ R. Marrou
________/____//____/\\/_/ /_//_____//_____/____________
This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:50:39 PDT