Oddity in rendered system width from MusicXML file
I report this now from a single very specialized case. I have not yet generalized the phenomenon. It may be rare in the wild.
We have a MusicXML file generated from a single-measure test score where, in the generating application, the system width was set to less than the full available width.
If that score contains two or more parts, and one or more of them are for a transposing instrument, then, when the MusicXML file is loaded into MuseScore, if Concert Pitch is ON, the system width is rendered essentially as in the original score. If Concert Pitch is OFF, the system is rendered at the full available width.
The attached MusicXML file should demonstrate this phenomenon.
Best regards,
Doug
| Attachment | Size |
|---|---|
| MXML_layout_2030.musicxml | 6.93 KB |
Comments
Maybe the Signature Keys?
In reply to Maybe the Signature Keys? by Shoichi
Shoichi,
Yes, that could well be part of the situation.
Ah, yes, your test score shows that nicely.
Thanks.
Doug
In reply to Shoichi, Yes, that could… by Doug Kerr
The presence of a key signature seems to affect the rendered width of a single-measure score by more than the allowance for the space the key signature takes up.
But the effect when we do not have a transposing instruments is a bit different from when we have such in the picture.
We see the effect with no transposing instrument in the attached test scores.
Very interesting.
As far as I can tell this has nothing to do with any quirk of MusicXML or of concert pitch, it's simply a matter of this being the last system of the piece and being just barely over the last system fill threshold that is set by Format / Style / Last system fill threshold. Any change that shrinks the system below that threshold - in this case, pressing Concert Pitch results is the elimination of a key signature, but it could just as well have been shortening the staff name or replacing two quarter notes with a half note - causes the fill to be suppressed.
In reply to As far as I can tell this… by Marc Sabatella
Hi, Marc,
Oh, sure. I had not thought about the last system fill threshold. So I think there is no anomaly here.
With last system fill threshold set to 100%, so there is no "fill", there is still a difference in the horizontal scale of the rendered score with key signature, but the result is certainly reasonable.
Thanks for getting me on track here.
Doug
In reply to Hi, Marc, Oh, sure. I had… by Doug Kerr
Well, sure, the key signature takes up space and is included in the nominal measure width, and that nominal width of the measure is then subject to internal stretch calculations that are applied even before the final full system stretch.