If you are a superior programmer then you can get better audio results
using fixed point, AND you can do it in less time with simpler code. Not
to mention that a floating point unit uses a lot more gates than a fixed
point unit.
For any given number of bits floating point is worse, uses more gates,
and is slower.
Is this for the on-board mixer?
If not then from what I read on the Linux Audio Dev list you guys may
be splitting hairs. The serious audio processing guys on that list
use the host CPU for all the effect work not anthing on the sound
card. A modern athlon blows away all but the really expensive DSPs
--
Richard A. Smith
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
