[Ffmpeg-devel] [PATCH] 2 % faster wma_decode_block and init simplification
Mon Mar 6 00:20:08 CET 2006
On Sun, Mar 05, 2006 at 11:41:49PM +0000, Benjamin Larsson wrote:
> >3. every change which does not lead to binary identical decoding must be
> > tested with some testsamples (max difference, mean squared error, PSNR)
> > i know from messing with the mp3 decoder that tiny changes (32bit vs. 64bit
> > multiplies at the wrong spot) do lead to audible issues
> > (basically expect anything which is non optional and causes >+-1 max error
> > to be rejected)
> It's difficult to get binary identical decoding when dealing with floats.
> I tested the attached patch with a sample and got a RMS value of
> 3.0519e-07. (http://www.underbit.com/resources/mpeg/audio/compliance/)
> So this patch cause a max error >+-1 even though it to me looks totally ok.
> And here I just replace the pow calculations with tables.
hmm, just so that we understand each other correctly, with max error >+-1
i meant that there is a error larger then +-1 in one 16bit output sample
now from what you say above its not 100% clear if thats the case or not
but an error of 3.0519e-07 RMS seems negligible to me, i assume that
dissapears if you change the tables to double? not that i think that would
be a good idea ...
except the precission question iam fine with the patch
More information about the ffmpeg-devel