Lousy performance during opening a file
Hello, together,
it lasts about 55 seconds to open the attached file where the task manager shows 60% CPU usage and 500 MB main storage usage. I am using MuseScore 2.3.2 (Rev. 4592407), but the problem occurs also with 2.3.1; any
other files are opened within some seconds. Any idea why?
Regards
Attachment | Size |
---|---|
_Op.65-No.59.mscz | 163.68 KB |
Comments
This file is fairly large, but seems to consist mostly of a few thousand "Fine" instructions all stacked right on top of each other. Not sure how that happened, but delete them all (eg, right click one, Select / All Similar Elements, Delete) and it should get much better. Although on my system, trying to doing that brings the computer to its knees...
Try this version:
Op.65-No.59_fix.mscz
Your measure 32 had thousands of overlapping 'Fine" markers, which I replaced with a single one.
Regards.
In reply to Try this version:… by Jm6stringer
Wow, this helped! Thank you very much!
Btw: How did you find the problem - or how would I have been able to find it?
In reply to Wow, this helped! Thank you… by Dorforganist
More importantly: how did that happen?
In reply to Wow, this helped! Thank you… by Dorforganist
To find it.
a. Look for a slur that looks a little darker than the rest. Click it and it will probably still be black. Press delete and repeat the process several times until you're tired of messing with it.
b. Open the .mscz using a zip program and extract the .mscx from it. Open the .mscx with an xml editor and scroll down until you see the spanner count is ridiculously high for the score. Slurs are not the only things counted as spanners but seem to be the only things that this happens to. A ridiculously high count is dependent on the size of the score, but 20,000 spanners in almost any song is unusual. You will also see hundreds or thousands of spanner definitions at some point in the .mscx with nothing between them. I usually scroll down near the end of the song and start looking for a spanner and find a very high count.
In reply to To find it… by mike320
The later is what I did (open the file in a ZIP program, then open the MSCX file within it in Wordpad). Scrolled down a little and pretty soon I wasn't seeing normal-looking score stuff. but instead the same few lines over and over and over and over.
But while it is indeed true that slurs seems to be the "usual" cause of this sort of problem, in this case it wasn't slurs but "Fine" markings. Which aren't even spanners. Another odd thing about these particular markings is that they were attahced to the bottom staff rather than the top. Normally these get automatically attached to the top staff as per standard practice, but I guess there are probably ways of defeating that.
Would indeed be interesting to know how this all might have happened.
In reply to The later is what I did … by Marc Sabatella
The common point between these few involved files in recent years (whether slurs or the "Fine" repeat/jump sign here) seems to be that their source is the XML format. And here, with an encoding "PDFtoMusicPro" (Myriad)
In reply to The common point between… by cadiz1
Yes, you are right - this file has been scanned from a PDF using PDFtoMusicPro. Maybe I'll give a hint to the developers.
Thanks for help!
In reply to Yes, you are right - this… by Dorforganist
I tried to reproduce problem and have scanned the PDF again, but there are no corrupted bars nor a lot of "Fine" in it. Seems they once have been added during editing it in MuseScore (very often in the last months).
In reply to I tried to reproduce problem… by Dorforganist
And I'm guessing it has to do with the unorthodox staff assignment. Probably it keeps getting copied onto itself because somewhere in the code we aren't expecting to see these elements anywhere but the top staff.