[FFmpeg-devel] [PATCH] Implement the function cmdutils.c:parse_int_or_die
Wed Feb 20 14:05:46 CET 2008
On Wed, Feb 20, 2008 at 01:43:44PM +0100, Michael Niedermayer wrote:
> Also the normal (outside the academic dreamworld) way to handle float is
> not to assume floating point operations to be exact and keep in mind that
> they are represented by a finite number or mantisse and exponent bits.
You forgot: and use them for things that do not really matter, or need less
than 0.01% of the precision available by the hardware (basically that's
what double is there for, not because someone actually needs the
precision in the end result) or use an algorithm that basically verifies the result.
More information about the ffmpeg-devel