On Fri, 29 Feb 2008, Ben Wheeler wrote: > On Thu, Feb 28, 2008 at 09:58:52AM +0000, Adam Davison wrote: > > If this really is the problem then we could avoid it by caching the > > first few k of decoded audio at the cue point in memory. > > I've long thought this would be a good idea; it's exactly what CDJs do, > so that playback from cue is instantaneous while the CD is seeking to > the right place.
Ahh...I was wondering about that very behavior Adam described. Last Friday I spent a few hours practicing with mixxx, with waveform-simple to avoid issues... everything worked fine, got two hours in without a hitch. I still have the cue-jump-latency symptom. I did compile for the profiler, but haven't done the obvious cue-cue-cue-cue-... to get profiler data yet. I'll do that Sunday or Monday. I still don't know why I had the gtk-window-no-update-callback problem, but it's not returned since I turned waveform off. Now that I have what seems to be a baseline of good behaviour, I will turn waveform on in a controlled manner and see if that makes it happen. Then I'll profile and see where it's spending it's time during that audio pause after cue business. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Mixxx-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mixxx-devel
