[Bf-committers] optimising the code

Stephen Swaney bf-committers@blender.org
Sat, 29 May 2004 10:42:47 -0500

Pardon my numerical ignorance, but are float arithmetic
operations faster than doubles?

Most modern computers have some sort of hardware
floating point unit for doing double precision math.
I rather suspect that float operations are done by
first converting to doubles, pushing them thru the
hardware FPU and converting back to floats.

Perhaps someone who is more of a hardware weenie can

Stephen Swaney