[FFmpeg-devel] [RFC] Re: releases and stuff

[Michael Niedermayer]

On Wed, Feb 11, 2009 at 09:48:10PM +0100, Benjamin Larsson wrote:
> Benjamin Larsson wrote:
> > Michael Niedermayer wrote:
> >> On Wed, Feb 11, 2009 at 02:51:43PM +0100, Benjamin Larsson wrote:
> >>   
> >>> Michael Niedermayer wrote:
> >>>     
> >> [...]
> >>   
> >>>>   
> >>>>       
> >>>>> Regarding fixed point tables and stuff, the fft routines might behave 
> >>>>> differently also. To be really sure a complete fixed point version would 
> >>>>> be needed.
> >>>>>     
> >>>>>         
> >>>> how hard is it to write a fixed point fourier transform? and i mean fourier
> >>>> not fast fourier
> >>>>
> >>>>
> >>>>   
> >>>>       
> >>> The ac3 encoder already have a fixedpoint fft and mdct. We could use 
> >>> those where needed. Fixed point generation of the mdct tables should be 
> >>> easy to implement also. Would a solution with a -regtest parameter be 
> >>>     
> >> we have -bitexact alraedy and yes fixed point code should then be used
> >> where possible
> >>   
> > 
> > Ok, I'll see what I can cook up this evening.
> > 
> > MvH
> > Benjamin Larsson
> 
> The patch is not really fixed point as it uses sinf. But it might be
> good enough for our needs. If not then some taylor approximation might
> be appropriate to use.
> 
> Compiled with --enable-gpl --enable-debug=3 --disable-optimizations
> --enable-swscale I get this with 4.3.2 on amd64 with these compile options:
> 
> 
> @@ -205,14 +205,14 @@
>  353368 ./tests/data/a-flac.flac
>  95e54b261530a1bcf6de6fe3b21dc5f6 *./tests/data/flac.vsynth.out.wav
>  stddev:    0.00 PSNR:999.99 bytes:  1058444/  1058444
> -4435d87463cd6c5407bd88cca241ca56 *./tests/data/a-wmav1.asf
> +4cd042a85c9c3f52124d6bcdba8cefdb *./tests/data/a-wmav1.asf
>  106004 ./tests/data/a-wmav1.asf
> -stddev:12251.38 PSNR: 14.56 bytes:  1056812/  1058444
> -stddev: 2105.96 PSNR: 29.85 bytes:  1048620/  1058444
> -24ae726aef89e5d17e315c64182f4c61 *./tests/data/a-wmav2.asf
> +stddev:12251.32 PSNR: 14.56 bytes:  1056812/  1058444
> +stddev: 2105.95 PSNR: 29.85 bytes:  1048620/  1058444
> +36491e6aca4516616e6b0c600330070a *./tests/data/a-wmav2.asf
>  106044 ./tests/data/a-wmav2.asf
> -stddev:12255.80 PSNR: 14.55 bytes:  1056812/  1058444
> -stddev: 2099.27 PSNR: 29.88 bytes:  1048620/  1058444
> +stddev:12255.59 PSNR: 14.55 bytes:  1056812/  1058444
> +stddev: 2099.25 PSNR: 29.88 bytes:  1048620/  1058444
>  89f5b8dd97e0dddbe59af0d44fd229f3 *./tests/data/a-pcm_alaw.wav
>  529256 ./tests/data/a-pcm_alaw.wav
>  0568b0b9a72e31559e150e7e09d301cd *./tests/data/pcm.vsynth.out.wav
> make: *** [codectest] Error 1
> 
> Would be nice if people could test this with other compilers and other
> platforms.

i remember posting some code to do sin/cos with integers using
bisection of angles and normalization of vectors
i dont remember the subj/date or what happened with it but
maybe it could be usefull?


[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The educated differ from the uneducated as much as the living from the
dead. -- Aristotle 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090212/44b2f65e/attachment.pgp>