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

[Nicolas Pomarède <npomarede@xxxxxxxxxxxx>]

Le 19/05/2022 à 16:45, Nicolas Pomarède a écrit :
> 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.

Hi

this issue should be fixed now too.
Different clocks between CPU and YM2149/DMA Audio are now correctly 
handled when computing YM2149's clock changes.

From what I hear with SMFX/DHS "The (fucking) (complex) Art of 
Profanity" demo, all the samples are correctly played now during the 
whole demo

Note that precision is not optimal yet, I see a slight deviation that 
could be improved, but I think it's 99% good enough already and should 
not give hearable artefacts.

Thanks for reporting theses 2 issues, I must admit that after rewriting 
the YM2149 emulation I did not run some precise enough tests/traces to 
show this slight change in number of samples per VBL (less than 0.2 %, 
but hearable after a while)

Nicolas