At 10:43 15/05/2003, you wrote:
Hi Wiljo, thanks for commenting so quickly. I found http://www.eee.metu.edu.tr/~alatan/PAPER/icip.00.pdf which comments on those to modes, and explains regarding VBR: "In VBR VBV model, data always enters VBV at Rmax, if the buffer is not full. When the buffer is full, no data enters the buffer".Well - actually in MPEG1 0xFFFF is used to mark a stream as variable bitrate. Due to an older version of the standard that I used when writing the first version of PVAStrumento, I kept this nomenclature for MPEG2.... With MPEG2 the standard really does not differentiate between CBR and VBR anymore.... _but_ a value of 0xFFFF for VBV-Delay puts the decoder buffer in a different method of operation (which makes it essentially not look at vbv-delay values).
How do you detect whether a stream is VBR in PVAStrumento? Do you look at bitrate changes between headers - but then again those could be set to CBR as you point out in the helpfile. Does this mean when "Flag Stream as variable Bitrate stream" is set it always sets the vbv_delay to 0xFFFF even for proper CBR files? I take it, it can't have a negative impact for the decoder then (as underflows are allowed but not overflows)?Maybe sth. like that happens, so setting vbv_delay to 0xFFFF will heal the decoder. I don't know.The linux driver ppl. should have a look at this.
You're welcome. I know what you mean, I'm sure you get lots of support mail, that can be quite a drag I guess.Thanks for the flowers ;-) You should mention that (the Helpfile....) to the german-only-speaking users :-/