On 03/07/06, Chris Cannam <[EMAIL PROTECTED]> wrote: > On Sunday 02 Jul 2006 16:08, Vince Negri wrote: > > Solution: when sending an MMC stop command, I set a flag to ignore > > the next incoming MTC quarter frame (since it might be "stale"). See > > attached patch. > > Sounds OK to me -- though are you sure it'll only ever be necessary to > eat one quarter frame? >
The quarter-frames arrive about 10ms apart. For eating one frame to go wrong, the multitrack has to emit two quarter frames in the interval between the AlsaDriver being told to stop and the STOP MMC command reaching the multitrack - in other words, the STOP command has to take over 10ms to propogate. 10ms is about 30 bytes's worth, so in a sensible setup (where the MTC'ed multitrack has a whole midi device to itself) it shouldn't be an issue; but on a loaded MIDI bus, or a slow machine, I suppose it could go wrong. However, you don't want to over eat, since that will hurt sync-up time. Probably a good compromise is to gear up to eat 3 quarter frames, but set the "eat counter" back to zero on a resetPlayback() or a play start. Vince Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Rosegarden-devel mailing list [email protected] - use the link below to unsubscribe https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
