[FFmpeg-devel] [RFC] remove table_4_3_value with CONFIG_SMALL

Michael Niedermayer michaelni
Fri Oct 16 00:43:55 CEST 2009


On Thu, Oct 15, 2009 at 07:04:41PM +0200, Reimar D?ffinger wrote:
> On Thu, Oct 15, 2009 at 02:48:04PM +0100, M?ns Rullg?rd wrote:
> > Reimar D?ffinger <Reimar.Doeffinger at gmx.de> writes:
> > > On Thu, Oct 15, 2009 at 11:45:31AM +0100, M?ns Rullg?rd wrote:
[...]
> > >> It is at least as nonsensical to have CONFIG_SMALL assume a blazingly
> > >> fast FPU is available.
> > >
> > > Yes, to a degree (though that might need further investigation with
> > > someone who actually uses --enable-small), though you could just as
> > > well say that those device with little memory are likely to have
> > > even slower CPUs, thus (almost) all those CONFIG_SMALL optimizations
> > > are nonsensical (yes, it is an exaggeration, but maybe you get the
> > > idea).
> > 
> > No, I do not get the idea.  Making the code orders of magnitude slower
> > to save a few kB is not acceptable.  Period.
> 
> Probably.
> But you are misrepresent this from the numbers I have (I admit, I have no
> FPU-less system to test on), it saves more than 100 kB while on
> FPU-enabled devices slowing down the function by a factor for (so its
> likely "orders of magnitude slower" on FPU-less), but that function is
> only a small part of overall MP3 decoding.

yes, also one could show that the bitrate puts a limit to the percentage
of coefficients that use this code. That is higher bitrate files are
hit harder by this ...

besides this, one can make the tables smaller by a factor of 4 by spliting
te exponent&3 out and doing a seperate 32x32 multiply
and if its not obvious already, smaller values occur more often so a partial
table with fallback to slow fpu for very rare cases can be done ...

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

If a bugfix only changes things apparently unrelated to the bug with no
further explanation, that is a good sign that the bugfix is wrong.
-------------- 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/20091016/e576c3c1/attachment.pgp>



More information about the ffmpeg-devel mailing list