Re[2]: [XviD-devel] xvid + bframes and mplayer/ffmpeg decoding

[Michael Niedermayer]

Hi

On Saturday 17 August 2002 14:44, Christoph Lampert wrote:
> On Sat, 17 Aug 2002, Michael Militzer wrote:
> > However I'm still a bit confused about this problem: our current idct has
> > a peak error of 1 (iirc) so I don't see how zeroes should be introduced
> > out of non zero coefficients, not even for quant 1...
try a matrix with all 0 and a single 3 or -3 (a few of these will have all 0 
output)
note: the h263 dequantizer at quant=1 will convert 1,-1 -> 3,-3 so 3,-3 are 
the smallest IDCT inputs possible

>
> Peak error is difference between  original data and  DCT+iDCT
> reconstructed? So there is no intermediate quantizer step?
hmm, dct isnt specified in the mpeg4 std, so i would guess they mean 
difference between the iDCT and the reference iDCT

[...]

Michael