On Thu, 27 Jun 2002, Mike Melanson wrote:
> On Fri, 28 Jun 2002, Bill Fink wrote:
> > It works fine, and even the lip sync is quite good, which is doubly
> > remarkable since the PPC OSS driver doesn't have realtime support and
> > thus must use OSS softsync.
> Interesting. Just for fun, run 'md5sum' on the file so I know we
> have the same file. My md5sum for the Phantom Menace "A" trailer is:
No, my md5sum is 5725cdb4c5e58fbae891e9f70d7f00e0. I got the menace_480.mov
BTW you can try out the OSS softsync (which relies on the system clock)
by setting audio.oss_sync_method to softsync in the .xine/config. This
works remarkably well on my home dual 500 MHz G4 which has an Awacs/Screamer
sound chip, staying in good audio/video sync for an extended period. But
on an 867 MHz G4 at work with has a Tumbler sound chip, the audio and video
drift out of sync after just a couple of minutes, which is not that
surprising using OSS softsync. I'm just glad that my home system works
so well in spite of needing to use OSS softsync.
> > Since I don't have the bugs, does that mean mine isn't working right? :-)
> I wonder if two wrongs have made a right here, between the
> QT demuxer sync bug and the PPC A/V sync problems (?). That file gives x86
> users problems, both on xine and MPlayer. The problem has been identified
> and is the process of being solved. The brngiton_002.mov trailer is known
> to work well for the x86 people. Get it from here:
> and see if it plays properly. It's another SVQ1/ima4 file.
That one works fine also including lip sync, although the video quality
isn't as good, but that's to be expected since it's 'SVQ1' video @ 320x172,
whereas the menace_480.mov is 480x216. For the record, the audio on
bringiton_002.mov is 'ima4' audio @ 11025 Hz, 16 bits, 1 channel.