When buffering PCRs (so that each TS packet *knows* its PCR), current code only
starts playing from the TS packet with the first PCR (since it doesn't know the
PCRs of anything before that). At the *end* of the file, it tries to (linearly)
estimate from the last two PCRs. One might assume that it should do something
similar at the start (since a PCR can occur quite far in).
Was bug 14628 on berlios.de
Original issue reported on code.google.com by t...@tibsnjoan.co.uk on 26 Sep 2010 at 2:09
Original issue reported on code.google.com by
t...@tibsnjoan.co.uk
on 26 Sep 2010 at 2:09