Re[4]: [XviD-devel] xvid + bframes and mplayer/ffmpeg decoding

[Michael Militzer]

Hi,

ok, I've just updated the CVS mbtransquant code to use your skip decision
criteria. I also did a test if additional checking for qcoeff[i*64+9] gives
some improvement but for my input material there was nearly no difference at
all (iirc ~0,00x dB), so I removed the qcoeff[i*64+9] check again...

bye
Michael

----- Original Message -----
From: "Radoslaw Czyz" <syskin@bigpond.com>
To: <xvid-devel@xvid.org>
Sent: Saturday, August 17, 2002 2:48 PM
Subject: Re[4]: [XviD-devel] xvid + bframes and mplayer/ffmpeg decoding


> >... reactivate coeff thresholding for now
> > (maybe define a TOO_SMALL_LIMIT of 2 or 3) which should fix the problem.
>
> Come on. Just implement my coeff thresholding. I know I should upload
> it to cvs myself but I can't, wincvs sux. I don't wanna destroy
> anything.
>
> For those who don't know:
>
> TOOSMALL_LIMIT = 3
>
> Block skip decision:
> if ( (sum >= TOOSMALL_LIMIT)
>      || (qcoeff[i*64] != 0) || (qcoeff[i*64+1] != 0) ||
>      (qcoeff[i*64+8] != 0) ) { encode_block }
>
> This code gives me psnr improvement of about 0.1 dB (doesn't degrade
> psnr in any case) for one-pass fixed bitrate.
>
> Perhaps checking qcoeff[i*64+9] (DCT 1,1) could improve it even more.
>
> Don't expect me to commit it :P , sorry.
>
> Radek
>
> _______________________________________________
> XviD-devel mailing list
> XviD-devel@xvid.org
> http://list.xvid.org/mailman/listinfo/xvid-devel
>