Velocity atrocity
A overview of Note velocity in MuseScore and MuseScore Studio
• Note velocity in MuseScore Studio 4.4.x functions only with MSBasic sound libraries, and not with MuseSounds. This point (and the following points in this section) are not rants or complaints; they merely describe how things are.)
• The Offset velocity type is a discarded relic. In MuseScore Studio there's now only an absolute/"user" velocity type.
• In MS3, dynamic markings "shelved" Offset velocity values effectively to higher or lower values (depending for example, if the dynamic was ff or pp. In contrast, in MS3, dynamic markings had no effect on User velocities—I believe Offset velocities were introduced to create a system where note velocity and dynamics could reasonably interact, and it's not hard to envision the challenge inherent therein.
The crux of this post
If you open or listen to the following score in MuseScore 3.6 or 3.7 you'll hear both User and Offset note velocities intact and functioning, as expected.
https://musescore.com/user/35880724/scores/22748065
Open the very same score in MS 4.4.x and look in Properties>General>Playback>Notes>Velocity. This shows that MuseScore Studio is not just ignoring the velocity values of the legacy score, it actual discards them (on Open) by setting ALL velocity values to User type 64. Thus the resulting score is expressively "flattened." And far worse, the velocity flattening becomes permanent to the score on Save, without as much as a warning—And THAT is a complaint, of the highest order!
https://musescore.com/user/35880724/scores/22748074
The future of velocity
Development has long promised that its planned "note volume" system will far outshine the limitations of the the old approaches used in MS3. And we can hope that will be great, especially now that we're 2+ years in the waiting.
Users' data should be of the utmost importance
In the interim, as there's no replacement system intact, why pull the rug on the old system?
When a user opens a legacy score that contains User velocities why not simply leave them be, thereby preserving a legacy score's velocity data within MuseScore Studio files? Earlier versions of MS4 did precisely that and managed to function by simply ignoring note velocity when using MuseSounds.
That allowed users to open a velocity textured legacy score in MS4.4x, choose a MSBasic library to hear note volumes according to the note velocity values as they existed in the legacy score. Alternately they could proceed without the effect/benefit of note velocity by choosing a MuseSounds library. But in either case, MuseScore Studio 4.3x and earlier didn't alter the user's velocity data.
In closing
I hope that the new system—when it arrives—will be capable of sensibly conveying note velocity values in MIDI exports.
When scoring for sound—and whenever MuseScore falls short of key functionality readily available in my DAW—I will still want-and-need to export to MIDI to transport my compositions and arrangements to a more fully fledged working environment.
Godspeed on these matters!
Comments
This might be more effective if you post audio files of part of an orchestra piece. One from MU3 and one from MU4. Also, the one from MU4 should actually be two separate files. One with Muse basic sounds, and one with Muse sounds.
In reply to This might be more effective… by bobjp
Sorry. I don't understand the reasoning in your advice.
I think the submitted scores suffice in providing a simple demonstration. So I'm unclear. Are you saying you're unable to hear the difference between the two scores when listening online ... or perhaps when listening to the first score in MuseScore and then the same score opened in MuseScore Studio?
In reply to Sorry. I don't understand… by scorster
Of course I can hear the difference. What I am saying is that a solo scale is not a very good real world example. What would be meaningful to me would be to have someone start with a short orchestra score in MU3. Produce an audio version without velocity changes. Just dynamics and hairpins. Then a second version with velocity changes. Then the same score in MU4 with muse sounds.
I never messed with velocity back when I used MU3. Every one of my compositions, to me, sound better in MU4 then they ever did in MU3. Would they sound even better if I could adjust velocity? No idea.
In reply to Of course I can hear the… by bobjp
Didn't you use any pp, p, mp, mf, f type markings in your MS3 scores?
In reply to Didn't you use any pp, p, mp… by yonah_ag
Of course. But the OP was about individual note velocity.
In reply to Of course. But the OP was… by bobjp
The pp ... ff dynamics have a (customisable) velocity associated with them. The OP just wants to refine this setting further. How would you like your scores to be flattened to mf throughout, ignoring any other dynamic markings that you have?
In reply to The pp ... ff dynamics have… by yonah_ag
OK, I have made a mistake. The OP starts out complaining about Muse sounds, but is ultimately about Basic sounds. Which I have no interest in. My mistake. It is a long post interrupted by scores from .com that might have made me miss some of the text. But not his fault. mostly mine.
And yet, I'm not sure there is great advantage to using MU4 with complete velocity control, if it is only for Basic sounds. No reason to upgrade. I have always said that Staying with the version that works best for someone is the best plan.
In reply to OK, I have made a mistake… by bobjp
Seems that way to me too: not much point upgrading unless you want Muse Sounds. For some of us the loss of certain features makes MS4 particularly unattractive. Maybe MS5 will combine the best of 3 and 4.
Let me explain further.
I do understand the value of velocity adjustments and PRE. They have their place.
But I might spend a vast amount of time tweaking most every note, Every phrase. Every dynamic. Everything. So that the score sounds just like I want it to. Composer intent.
The problem is that the score will only sound that way within MuseScore and on the system that created it. The rendered mp3 or wav will sound different. And it will sound different on a different system. Headphones, or computer speakers, or car stereo, or blue tooth speaker, and on and on.
Sure these things are helpful when working with tablature. Which I don't.
And then there is the idea of making playback sound more human. Factor in some bad timing or some kind of mistake. To me the whole thing is obviously artificial as it is. Why try to make it worse.
MU3 sounds need all the help they can get. I was never so happy as to leave them behind. But who knows. This is why I would like some sample music. My own experience might be too limited. I do have some music that I worked on in MU3 years ago. To me, the results of the same score, without changing anything, were far superior in MU4. Not perfect, to be sure.
In reply to Let me explain further. I do… by bobjp
Whilst different playback systems will sound different, this will be minor in comparison to throwing away all the dynamics. Playback of CDs sounds different on different hi-fi systems but we still get an insight into the emotion of the music - composer's intent. The same piece with the dynamics squashed could sound lifeless and boring in comparison.
Tablature and PRE are not relevant to what scorster is demonstrating here. This is simply a matter of dynamics: note velocity in the inspector, and relates to standard notation just as much.
Your comments on human playback are completely off topic. Scorster is not advocating bad timing, he is talking about dynamics. Your attempt to conflate so called 'humanisation' with tweaking dynamics is to miss the point entirely.
In reply to Whilst different playback… by yonah_ag
"throwing away all the dynamics" is quite an inaccurate statement.
Actually, all my comments are "off topic" because Scorester isn't really interested in reaction from users on this topic. We all already know the many short comings of MU4.
I was merely interested in a real life example of what "throwing away all the dynamics" would sound like in an orchestra score. Considering that velocity control is hardly all the dynamics. I know that is off topic. Sorry.
In reply to "throwing away all the… by bobjp
In context we are talking about velocity dynamics, not all dynamics, and that's what my comment applied to. However, velocity control is a significant part of dynamics, whether applied via, pp, p, mp, mf etc. or by individual note velocity setting.
In reply to "throwing away all the… by bobjp
Your off topic was not about orchestral scores but rather dragging in bad timing, which has nothing to do with dynamics.
"Velocity atrocity" says it all. What a shambles.
I always get hooked somehow on this topic and I probably shouldn't because all this has already been said many times in one way or another.
This time Scorester gave a very clear and illustrative example and expressed a hope that the situation will be improved in the future. Personally, I hope the same.
While I can understand Bobjp reaction if it came out of a frustration of repeated objections, I still think he repeatedly chose a wrong ways to react. I don't want to offend anyone and I apologize in advance but... Bobjp comments leave the impression that they come from a person who has no much interest in classical music, especially romantic music and has not developed an ear for a good interpretation that can arouse strong emotions in the listeners. 'To play a wrong note is insignificant; to play without passion is inexcusable' said Beethoven. Music played by a master of interpretation sounds like a completely different universe than the same music played robotically. The heck, this is exactly what is called musicality. In many videos on YouTube, we can see how much pianists, together with their teachers who lead them, practice the fine details of interpretation, like small rubato and dynamic changes (especially for music of Chopin).
To be fair, I believe that Bobjp is aware of all this, only he thinks that this is not important for a computer playback. But even a computer playback can sound much better when it's tweaked by a person who developed an ear for interpretation details. Thus it would be very useful for us all to have tools which could make this tedious process easier. Then maybe more people would dare to go into this process.
In reply to I always get hooked somehow… by hstanekovic
I think that's a fair summary.
As you say, the process can be somewhat tedious so having an efficient set of tools for this would really help. I don't want to go down the DAW route so hopefully, (with the improvements to engraving and sound production now in place with MS4), the next development phase will include such a set of tools.
https://musescore.org/en/node/373861
I believe this is part of the conversation.