The file prediction dosen't seem to work very well when I use the MPEG-2 templates, is it just me? or is it that it dosen't work for svcds?

thanks for the help.

As long as you use one of the KVCD templates, it should work. It was heavily tested with the KVCD matrix and GOP. Not with standard TMPEG parameters.

-kwag

would it cause a big difference in the prediction if the source file is a divx animation file?

It really shouldn't matter, because you're sampling the complete source at a number of intervals, which is proportional to the playing time of the material. So the longer the material, the more snapshots are taken. That is, as long as you follow the formula: MPEG size = ((Total frames/MovieTimeInMinutes)/24) * MPEG sample file size

-kwag

So would the length of the file affect the accuracy of the prediction? since the file I'm encoding is 23 min 31 sec.

The rule is very simple: The more samples you take, the higher the accuracy. BTW, if you want to put a 23 minute video on a CD-R, and only that video will go in the CD-R, hell! increase your CQ to 100 and encode You'll fit that video in one CD with room to spare

-kwag

As i mentioned before, the file I'm encoding is only 20 min long, and I would like the sampler to make a longer sample, so as to get a more accurate result, would the following script be correct?

Sampler(samples=90, length=24)
## MPEG size = ((Total frames/MovieTimeInMinutes)/24) * MPEG sample file size ##

so what equation should I use?