Hi there, On Wed, 11 Jun 2003, Steven M. Schultz wrote:
> > If you use 24000:1001 and pulldown instead of 30000:1001, the encoder will > > keep switching between top_field_first = true/false to generate the > > pulldown sequence. > > Now it's my turn to ask why would I want to do a pulldown when I have > ~30fps interlaced with is what I wanted all along? Ah.. I thought you have a progressive source and wanted to make a dvd compliant stream from it.. I misread your original post. > > > You have the full 444 chroma info at the ppm stage so isn't it best > > to subsample it directly to 420_mpeg2 from the ppm? Otherwise you go > > through 444 -> 420_jpeg -> 420_mpeg2 -- looks more lossy to me. > > Hmmm, possible but I have a feeling that the jpeg->mpeg2 conversion's > "loss" is so small you'd need Golden Eyeballs to notice it ;) Certainly > it'd be a lot less of a loss than the 444 -> xxx conversion. I haven't done any tests. As you say the difference could be small in practice, but in "theory" any subsampling change is lossy ;) Subsampling is basically scaling so the quality also depends on the software -- y4mscaler is indeed a good scaler. Cheers, Selva ------------------------------------------------------- This SF.NET email is sponsored by: eBay Great deals on office technology -- on eBay now! Click here: http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5 _______________________________________________ Mjpeg-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/mjpeg-users
