Corrupted custom beam depending on the platform

• Apr 2, 2010 - 18:48
Type
Functional
Severity
S2 - Critical
Status
closed
Project
MuseScore

I scored "Fligh of the Bumblebee" with r>2920 by copy pasting from a capella file on Windows
Attached is the result.

If I open it with MacOSX (same version) no problem. But if I open it with MuseScore on Ubuntu (same version too). I got the attached PDF. Look measure 39. I believe I have customized these beams by moving the handles but I can't reproduce from scratch.

Attachment Size
Hummelflug.pdf 226.2 KB
bad_measure.jpg 7.35 KB

Comments

• Apr 4, 2010 - 10:43

On my ubuntu laptop, mscore -f sends back
printer DPI 1200,000000(1200) display PDPI 96,000000(96) DPMM 47,244094

On mac or windows
printer DPI 600,000000(72) display PDPI 72,000000(72) DPMM 23,622047

Apparently, the user defined y1 and y2 for beams are stored as relative value and depends on the DPI. If the target platform has a different DPI than the platform where the score has been created, we have a problem.
I didn't test with other "user modified" offset but the same problem might appear. Two solutions : Or we store the source DPI or we store absolute value in a given DPI.

Score portability being a top feature I put this bug on critical.

Reply

• Apr 4, 2010 - 12:45
Status (old) active ⇒ fixed

fixed in r2936. The beam position was saved in pixel values but should be normalized to spatium. I increased the *.msc version number to 1.14 so that values in older files could still be interpreted as pixel.

Reply