I have to have a program change + sysex + program change to make the change on the synth happen.
two program changes? that's unusual and i'm afraid it's not well regarded under the qtractor model, where one track is one channel, is one program change. the latter is always sent at the beginning of the playback sequence and may interfer (or not) with any that is embedded on original midi file. that might well be the cause for your worries. but i'm not sure how that relates to missing sysex.
maybe a copy of one or two of the midi files that expose the suspicious behavior is of help here, for evidence right?
I have to have a program change + sysex + program change to make the change on the synth happen.
two program changes? that's unusual and i'm afraid it's not well regarded under the qtractor model, where one track is one channel, is one program change. the latter is always sent at the beginning of the playback sequence and may interfer (or not) with any that is embedded on original midi file. that might well be the cause for your worries. but i'm not sure how that relates to missing sysex.
maybe a copy of one or two of the midi files that expose the suspicious behavior is of help here, for evidence right?
byee