[FFmpeg-devel] [TeX] DCA subband transform

Michael Niedermayer michaelni
Sun Sep 7 14:58:29 CEST 2008


On Sun, Sep 07, 2008 at 06:26:53PM +0600, Alexander E. Patrakov wrote:
> Hello,
> 
> recently there was a discussion whether the DCA subband transform is a
> windowed MDCT or something else. The attached TeX file contains the proof
> that the original obfuscated transform is in fact, up to a constant factor,
> a pseudo-QMF cosine modulation ?lter bank combined with the sign ?ip of
> subband inputs whose number is congruent to 1 or 2 modulo 4.

It would be more interresting to know if every "pseudo"QMF is a windowed *DCT
and if every windowed *DCT is a "pseudo"QMF


> 
> I will use this math in order to write a higher-level simplification of the
> transform code, in particular, I will remove the "noidea" history buffer.

you can just look at our mp2 decoder that implements the same summation 
without the extra buffer.

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

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- 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/20080907/2665b160/attachment.pgp>



More information about the ffmpeg-devel mailing list