Re: [hatari-devel] Audio/video sync regression in dev-builds

[Nicolas Pomarède <npomarede@xxxxxxxxxxxx>]

Le 10/05/2022 à 18:20, Nicolas Pomarède a écrit :
> Le 10/05/2022 à 18:06, Anders Eriksson a écrit :
> > > For the 1st one, I can create the same wrong sync with devel version,
> > > but if I test with 2.3.1 too, then I also have the same bad sync around
> > > 8min.
> > > Do you confirm sync goes bad too with 2.3.1 ? If so, does it work better
> > > with hatari 2.2 or 2.1 ? (it's possible this drift between audio / video
> > > has always existed more or less and becomes noticable only with large
> > > video).
> > > In the end, it could be related to how fps is stored inside avi file and
> > > maybe there's a bad rounding (we have maybe 0.6 sec of shifting between
> > > audio and video after 8 min, which would be a 0.1 % error, not that much
> > > :( )
> >
> > Hi,
> >
> > back in August I did the (what I think is correct) recording with 
> > 2.3.1. But just to be sure I re-did it now. And it works as expected, 
> > the sync is correct at the Pong part.  So I can't re-create bad sync 
> > with 2.3.1.
> > When I recorded the Youtube thing it was done with macOS 10.4 x64, and 
> > the test today with macOS 12 AARCH64 in Rosetta 2 mode (no AARCH64 
> > binary available of 2.3.1).
>
> ok, that's strange because with 2.3.1 I have the sync problem here. I 
> will do a recording again, maybe I did sthg wrong.
>
> Just to be sure, can you tell which command parameters you use when 
> running hatari to do a recording ?

Hi

For the 1st issue, regarding the audio/video sync problem after several 
minutes of recording, it should now be fixed in current dev version. 
Doing a recording of OVR demo did not show the sync bug when playing it 
later.

Troed's build for macOS should be updated on his site, let me know if it 
works for you too.

This kind of error was quite complicated to track, because we're talking 
about a 0.015 % difference when generating emulated sound freq between 
hatari 2.3 and hatari dev/2.4, which makes it very difficult to see any 
obvious change unless you run the demo for several minutes :(
Had to do a lot of traces to see the problem, but at least it's fixed now :)


For the 2nd issue with STE DMA sound being not in sync too, this is in 
fact similar to the same problem there was with your "More or less zero" 
demo. It was fixed back then, but the new code in hatari dev doesn't 
take into account that the STE has a slightly different clock for the 
YM2149 compared to the CPU (on STF they're the same).
This should be fixed soon I think.

Nicolas