[Ffmpeg-cvslog] CVS: ffmpeg/libavcodec utils.c, 1.173, 1.174 avcodec.h, 1.455, 1.456

[Mike Melanson]

Alex Beregszaszi wrote:
> Might be an option, but on intel double is just 48bits. However, I dont

	What are you talking about? What CPU uses 48-bit floating point numbers?

> think double precision audio has any sense :)

	I don't think so either, but 64-bit floating point does exist in the wild.
-- 
	-Mike Melanson