Hello,

Can anyone tell me the significance of setting B spoilage in CQ settings from 0 to 20? Under Tok, it was initially set to 0, and if this was set to 20 the sample file size decreases thus giving me a higher CQ value, but then again the prediction is no longer accurate. For example, my target sample file size i 12,878,060 and the predicted CQ will have a sample size of 12,775,262 but after encoding and muxing, the total file size exceeded the 80min CD-R max size for almost 30 to 40MB.

Anyway, just want to know which one will have a better movie quality improvement, B=0 or B=20.

Thanks,
Dredj

That's a complicate question that I already ask a long time ago . The answer from Kwag was that the spoilage just "bow" the CQ curve and let think that you have a little sample but in fact the result is less good. And you have to raise a little the CQ to obtain the same result at spoilage=20 than with spoilage=0 (so finally, the sample size will be the same )

I did some new test this we in fact, and found Kwag was right.

It seems that all information is in this thread but I never manage to read it all

http://www.kvcd.net/forum/viewtopic.php?t=2073