Hi all I had a chance to test Alsa 1.0.0pre3 over the weekend against a suite of programs I use. I noticed a small but very annoying quirk with the OSS midi emulation. The applications were jazz++ 4.1.3 and playmidi 2.4. While the symptoms were slightly different I suspect there might be one underlying cause. Output is to the external MPU401 on an ES1370 card. Test midi files used were created under jazz++ 4.1.3.
Under Jazz++, any "hold on" controller event which occurs within the first bar played gets turned off too quickly (assuming playback starts at beat one of a bar). In fact, even if there are no subsequent "hold off" events within the entire midi file the "hold on" is still turned "off". The turning off happens very soon after the "hold on" is transmitted - the "hold on" has some effect but it's very short lived. "hold on" and "hold off" events work exactly as expected in all other bars of a piece - it's only in the bar at which playback starts (not necessarily the first bar of the midi file) that there's a problem. If you start playback at bar 4 then the problem's in bar 4. If you start at bar 2 the problem's in bar 2 and bar 4 plays perfectly. The effect is similar even if playback starts in the middle of a bar but then obviously the effect can sometimes carry over into the next bar. I have not done extensive tests to quantify timings in the case of starting mid-bar. Under playmidi the problem is a bit different. It seems that the midi events in the first beat of the first bar of the midi file played are sent too quickly - it's as if the timing info sent back to playmidi runs too fast for the first beat or so. I didn't get a chance to get exact timings on these things - observations were by ear. If it would help I can try to measure things a little more objectively. My rather nieve throught is that the timing events sent by the ALSA emulation are running fast for the first bar or so of playback. Whereas this affects all events sent by playmidi, it only affects controller events from jazz++ (maybe because of the odd way jazz++ does its timing). Of course there may be more to it than this - if it was simply timing then an isolated "hold on" would still remain in effect during jazz++ playback, but tests show it is turned off "by magic". Any ideas as to what could be causing these problems? The existance of this problem is the one thing stopping me from moving to ALSA permanently. Currently I have to use OSS for midi work which is a bit of a pain. Regards jonathan -- * Jonathan Woithe [EMAIL PROTECTED] * * http://www.physics.adelaide.edu.au/~jwoithe * ***-----------------------------------------------------------------------*** ** "Time is an illusion; lunchtime doubly so" ** * "...you wouldn't recognize a subtle plan if it painted itself purple and * * danced naked on a harpsichord singing 'subtle plans are here again'" * ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Alsa-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/alsa-devel
