[Ffmpeg-devel] Fixed vs. Floating Point AAC

Mike Melanson mike
Fri Mar 10 02:50:36 CET 2006

Rich Felker wrote:
> The evidence is clear that even on present-day cpus, integer
> arithmetic is often faster than floating point. Floating point fanboys
> will cite cycle counts or isolated benchmarks, ignoring the overhead
> of converting to/from float and ignoring the gains from using
> vectorized integer operations with small operands. I don't want to

	All right, that's it-- if I ever do write this new AAC decoder 
implementation, I'm using floats just to make you lose even more sleep 
over the sub-optimality of the world at large. Double precision floats, 
no less. And written in C++ using templated, polymorphic, inherited 
overloads, just to add insult to injury.

	Benchmark that!

	-Mike Melanson

More information about the ffmpeg-devel mailing list