Poor/Errant conversions of MS3 Offset velocities to MuseScore Studio 4.3
On opening a MS3 score that contains Offset velocities MuseScore Studio 4.3 converts the notes with Offset = 0 to velocity 64. Fair enough ... but a negative Offset value becomes the velocity!
For example, if the Offset value in MS3 was -5 I'd expect MS4 to convert the velocity to be 59 (that's MS4's default absolute velocity of 64, minus 5) but MS4 converts it to -5. Not surprisingly the result sounds horrible.
Not surprisingly the result sounds horrible.
On the "positive" side ... a positive offset velocity of x converts to x
but sounds like velocity = 0 + x
https://musescore.com/user/35880724/scores/8742996
Comments
Isn't the default 80?
Edit: not for mp, but for mf
In reply to Isn't the default 80? by Jojo-Schmitz
Well, it's weirder than that.
On closer examination I see that (years ago) I set mf to 60 ... obviously to soften the sound. Still Offset 0 gets imported to MS4 as 64.
But the real concern is the how a -x Offset converts to absolute velocity = -x. I know that Offset velocities were dumped in MS4, and that velocities convert to Absolute. the part that seems unreasonable is the conversion of the incoming offset velocity to = -x (absolute).
scorster
So, it is already known that velocities don't work well in MU4 with Basic sounds. And not really at all with Muse Sounds.
I loaded your score into MU3. Am I supposed to hear any volume difference between notes? I notice that in the inspector, all notes are offset. Is the negative number supposed to be louder or softer? I'm just trying to understand because I've never used velocities. This is how I learn.
In reply to So, it is already known that… by bobjp
Bobjp wrote > it is already known that velocities don't work well in MU4 with Basic sounds.(
My understanding is that absolute velocities work as expected in MuseScore 4 with MSBasic sounds. And I notice that all play at the same volume if their Offset = 0 ... of if all Offsets are specific number.
Bobjp wrote > Am I supposed to hear any volume difference between notes? I notice that in the inspector, all notes are offset.
Yes, in MS3 notes with Offset = -x are slightly quieter.
In my MS3 example score all notes have their velocity type set to Offset. Those unchanged are at Offset = 0. In MS3 Offset velocities allow MS3 to interact with dynamics.
In reply to Bobjp wrote > it is already… by scorster
Here's an MS3 test score with offset velocities and various dynamic levels:
https://musescore.com/user/35880724/scores/17039644/s/XqixwW
In reply to So, it is already known that… by bobjp
For a long time, MuseScore has relied on velocity as the adjustable parameter for controlling volume. It intuitively made sense, as the harder/faster one strikes a piano key, the louder is the resulting sound. In reality, though, changes in timbre were ignored, as the simpler soundfont samples were not layered with timbral changes. Using MIDI velocity numbers for volume (instead of using an expression/volume controller) allowed for user offsets to be applied (in the Inspector) to the "Velocity type" for any desired sections of a score. This differed from the "User" (absolute MIDI) velocity.
@MarcSabatella provided a good explanation for its use here:
https://github.com/musescore/MuseScore/issues/10990#issuecomment-127797…
and:
https://github.com/musescore/MuseScore/issues/10990#issuecomment-127805…
Indeed, this is a known issue, report on GitHub but apparently not a priority. The whole velocity system is being revamped for future updates and I guess no one wanted to develop interim solutions. A mistake in my opinion, but given how seldom this seems to come up, I guess not one that has affected to make people.
In reply to Indeed, this is a known… by Marc Sabatella
Thanks Marc.
It's a painfully long time waiting for the revamp of the note velocity system. 1.5 years already.
Can you post a link to the Github report(s)?
Much appreciated!
scorster
In reply to Thanks Marc. It's a… by scorster
A quick Google search yielded https://github.com/musescore/MuseScore/issues/14768
In reply to A quick Google search… by Marc Sabatella
Thanks!! I've continued the discussion there.
scorster