Re: [FFmpeg-devel] [PATCH] avcodec/mpegaudio_tablegen: speed up table generation

On Wed, Nov 25, 2015 at 9:22 PM, Timothy Gu wrote: > On Wed, Nov 25, 2015 at 6:03 PM Timothy Gu wrote: > >> On Wed, Nov 25, 2015 at 05:17:29PM -0500, Ganesh Ajjanagadde wrote: >> > +double f = value * cbrt_lut[value] * pow(2, (exponent - >> 400) * 0.25 + FRAC_BITS + 5) / IMDCT_SCALAR;

Re: [FFmpeg-devel] [PATCH] avcodec/mpegaudio_tablegen: speed up table generation

On Wed, Nov 25, 2015 at 6:03 PM Timothy Gu wrote: > On Wed, Nov 25, 2015 at 05:17:29PM -0500, Ganesh Ajjanagadde wrote: > > +double f = value * cbrt_lut[value] * pow(2, (exponent - > 400) * 0.25 + FRAC_BITS + 5) / IMDCT_SCALAR; > > While at it, you could change pow(2 to exp2(, which h

Re: [FFmpeg-devel] [PATCH] avcodec/mpegaudio_tablegen: speed up table generation

On Wed, Nov 25, 2015 at 9:03 PM, Timothy Gu wrote: > On Wed, Nov 25, 2015 at 05:17:29PM -0500, Ganesh Ajjanagadde wrote: >> +double f = value * cbrt_lut[value] * pow(2, (exponent - 400) * >> 0.25 + FRAC_BITS + 5) / IMDCT_SCALAR; > > While at it, you could change pow(2 to exp2(, which

Re: [FFmpeg-devel] [PATCH] avcodec/mpegaudio_tablegen: speed up table generation

On Wed, Nov 25, 2015 at 05:17:29PM -0500, Ganesh Ajjanagadde wrote: > +double f = value * cbrt_lut[value] * pow(2, (exponent - 400) * > 0.25 + FRAC_BITS + 5) / IMDCT_SCALAR; While at it, you could change pow(2 to exp2(, which has a libm.h shim and is easily 4 times faster than pow() o