max quantizer & n pass encoding

[tawy]

yo,
i need 2 answers....

1- where the max quantizer, easy to config on 5.02, is on the 5.03 version ? i think it'll be at max 31 and this is why my encoding have a lot of artifacts and crappy things around objects on the encoded movie (it's on regedit, but which key) ???

2- nth pass encoding, as divx.com said, works with the bitrate modulation. how it works ?

i think a website explain it somewhere, but i didn't find it...

thx

[dTb]

Afaik there is no way to adjust min/max quant etc. in DivX 5.0.3, there aren't even any registry keys. I doubt that it's set to max of 31 however, I tend to think that it's most likely the same as the old default of 12. I would say that only the developers know however.
If you're interested in what quants have actually been used try the on screen display option of the ffdshow filter.
nth-pass aims to refine the bitrate distribution with every pass. The bitrate modulation will distribute the bits to obtain better quality on the high or low motion scenes at the expense of the other.

[theReal]

You can still set the min and max quantizer via the commandline of Divx 5.03.

Data Rate
-dr MaxQ,MinQ,RCAveragingPeriod,RCReactionPeriod,RCDownUpRatio
Example
-dr 12,2,2000,10,20


However, I think it's a bad move they removed these settings - I mean, they could've hidden them under "expert's settings" or something, so newbies wouldn't be confused by quantizer numbers.

[dTb]

Have you actually got those options to work via the cli theReal? I don't think it recognises them as they just disappear when you adjust any other settings. I'm yet to come across anyone who has been able to adjust any of the data rate controls.