On Mon, Jan 11, 2016 at 7:23 PM, Ganesh Ajjanagadde <[email protected]> wrote: > This is quite an accurate approximation; testing shows ~ 2ulp error in > the floating point result. Tested with FATE. > > Alternatively, if one wants "full accuracy", one can use powf, or sqrt > instead of sqrtf. With powf, one gets 1 ulp error (theoretically should be 0, > as > 0.75 is exactly representable) on GNU libm, with sqrt, 0 ulp error. > > Signed-off-by: Ganesh Ajjanagadde <[email protected]>
Pushed. Took the liberty of abstracting out the pow implementation we discussed on IRC into a utility function and push that. It's more readable, and next time we'll know clearly that's the way to go. _______________________________________________ ffmpeg-devel mailing list [email protected] http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
