Bug: Gradual tempo changes like accelerando and ritardando don't seem to affect long notes, causing them to either end early or sustain a bit into the next measure.
This only seems to happen to notes within the period that the tempo is changing — everything is perfectly normal on the measure after the tempo changes. I have attached a file showing the issue using short notes on cellos and long notes on violas (although this seems to apply to any sustaining instrument). It seems that the software calculates the length the note should play based on its duration at the original tempo, without factoring in how that time changes with the tempo. This would also account for why short notes still work fine — any such miscalculation is too small to notice.
| Attachment | Size |
|---|---|
| Tempo Change Issues.mscz | 20.22 KB |
Comments
This is a known issue, being worked on but with no good solution in sight.
In reply to This is a known issue, being… by Marc Sabatella
If you watch the tempo indicator on the upper left corner, you'll notice it doesn't gradually change, but rather changes every quarter note. I just realized that I could use that to my advantage by breaking a problematic half note into two tied quarter notes. The interpretation is musically the same, but the software ends up calculating it properly so the note stops in time. Try that in the demo file I sent and you'll hear the difference. Even just breaking up the whole notes into a dotted half note tied to (and followed by) a quarter note fixed the issue.
So a potential solution (at least, until something more elegant can be discovered) could be for the software to automatically treat long notes as a chain of tied notes no longer than a quarter, either everywhere or just whenever they are within a tempo change (unless the necessary calculations are memory inefficient).