You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Long story short, MuseScore 4 benefits from having split soundfonts in which each split is forced to bank 0, preset 0 (although percussion seems to work with bank 0 or 128).
I managed to figure out where in the code it makes sense to force these to 0, while still allowing the naming to reflect the original preset. So I have a build from my own fork that does what I need. I'd be happy to contribute it back via a PR, but I assume this should be made optional behavior, and I don't feel as confident about adding my own command line parsing code - I figure you may have a preferred method you'd like to use.
Note I also changed the file naming in a way that suits my purposes. For one thing, I skip the bank because it's no longer relevant. I keep the preset number but use 128 for drumsets, which matches their bank number but conveniently places them at the end of the list when sorted alphanumerically. I also enclose the preset/bank number in curly braces to make sure the split soundfont files appear after the original and any other "full" soundfonts in alphanumeric ordering, and I eliminate the underscores because they don't look so nice in the MuseScore GUI. So, "{000} Grand Piano", etc.
This naming is probably something pretty unique to my usage, but anyhow, this could be made a command line option also, or perhaps some more general solution to allow customization of names could be devised (eg, passing an actual "printf" format string as the argument to the option).
I'm also perfectly happy to just keep using my fork and not worry about updating your repository, but I thought I should at least let you know what I'm up to.
And - thanks for providing this tool!
The text was updated successfully, but these errors were encountered:
Long story short, MuseScore 4 benefits from having split soundfonts in which each split is forced to bank 0, preset 0 (although percussion seems to work with bank 0 or 128).
I managed to figure out where in the code it makes sense to force these to 0, while still allowing the naming to reflect the original preset. So I have a build from my own fork that does what I need. I'd be happy to contribute it back via a PR, but I assume this should be made optional behavior, and I don't feel as confident about adding my own command line parsing code - I figure you may have a preferred method you'd like to use.
Here's my fork: https://github.com/MarcSabatella/soundfont-cmdline-tools
Note I also changed the file naming in a way that suits my purposes. For one thing, I skip the bank because it's no longer relevant. I keep the preset number but use 128 for drumsets, which matches their bank number but conveniently places them at the end of the list when sorted alphanumerically. I also enclose the preset/bank number in curly braces to make sure the split soundfont files appear after the original and any other "full" soundfonts in alphanumeric ordering, and I eliminate the underscores because they don't look so nice in the MuseScore GUI. So, "{000} Grand Piano", etc.
This naming is probably something pretty unique to my usage, but anyhow, this could be made a command line option also, or perhaps some more general solution to allow customization of names could be devised (eg, passing an actual "printf" format string as the argument to the option).
I'm also perfectly happy to just keep using my fork and not worry about updating your repository, but I thought I should at least let you know what I'm up to.
And - thanks for providing this tool!
The text was updated successfully, but these errors were encountered: