Lance R. Marrou (marrou++at++vsl.ist.ucf.edu)
Fri, 4 Nov 1994 00:37:40 -0500 (EST)
> Use pfSinCos; it's fast. According to the current author
> dog/flight, it beats some table lookup code.
I don't see how this is possible, but it definitely would be more accurate.
I would think it would take longer for the function call to pfSinCos (without
actually doing the SinCos calcs) than the simple table addressing. Anyways,
its irrelevant for my case since I need more accuracy than lookup tables.
> > BTW, am I right in remembering the degree fidelity in GL and thus Performer
> > is 1/10 of a degree (thus the .05 for smooth transitions between tenths
> > of degrees)?
>
> Hmmm... just because the old integer rotate(3G) still exists?
Actually, I do not quite remember where I read this. I could have swore it
was in Performer somewhere (maybe the 1.0 manuals). But it seems logical that
I was thinking about the old rotate function. In fact, some of our old code
here was based on integer tenths of degrees! Either way, I see nothing
referencing this low fidelity in the new manuals or man pages (except the
old rotate, of course). However, for those who are interested, I think the
current man page on pfDCSRot is rather vague about where the 1/1000 degree
accuracy occurs (I assume the the pfDCSRot uses the faster, though less
accurate pfSinCos function). It took a few minutes and several man page
calls to figure it out. :) (though that in itself made me learn more by
reading more man pages).
>
> The gfx pipe's accuracy is effectively that of the single
> precision float that is used in the load/multmatrix(3G) or
> rot(3G) and should be better than 1/1000 degree. pfSinCos has
> similar accuracy.
Ok, in case you are interested, I have implemented the direct matrix
calculation for the off-origin rotation. It is definitely faster, though
I do not have specific empirical data on it. Now, I need to increase the
resolution on the angle checking (from the 0.05f to 0.0005f - note the f :)
Thanks again for the quick response. Bob, you catching all this? ;)
_______________________________________________________
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 - 17:50:38 PDT