Please bring back channel access in the mixer

• Jul 26, 2024 - 23:00

I miss the way instrument channels were handled before:
If you had a violin, you could click on an arrow to get equal access to the pizz and tremolo sounds.
Now, it's all done automatically behind the desk as it were.

In many of my scores, I used and abused trumpet, bass, and synth parts to create multiple guitar and keyboard lines. Once I discovered this method, I never used Change Instrument again.
Many instruments without a growing mixer, perfect.
Please bring that back.

Comments

• Aug 1, 2024 - 16:29

I forgot to mention:
The old way also gave the user independent control over volume and panning for each sound.
Pizz volume might be too low for instance.
The full control was essential for the use I had for it.

Reply

• Aug 1, 2024 - 18:46

It's already the case that you can switch back and forth between sounds with the mixer continuing to expand - unlike in MU3, existing channels are now reused when switching back to any instrument used previously.

That said, a way to subgroup the various channels for a single instrument would be useful, and a much better idea that reintroducing the old hack of having pre-defined subchannels for certain instruments but not others and having to "abuse" them just to get basic functionality.

Reply

• Aug 1, 2024 - 19:47

So maybe I've missed something; it's certainly possible, likely even.
Can one in 4+ add any instrument desired, and then have vol and pan adjustment of it, to any instrument, without using specific text like 'pizz' 'tremolo' or similar?
How might one boost the pizz volume for instance?
If so, can you briefly describe how?
Thanks
I'll look into this.

Reply

• Aug 1, 2024 - 19:52

Well, there is still text involved - the instrument change - but you aren’t limited to hacking trumpets or violins just because they are among the few instruments to provide us channel options. It works in MU3 too, but as you noted, it has the undesirable side effect of adding a new channel for each and every change rather than reusing existing channels when switching back to an instrument you used before.

Reply

• Aug 2, 2024 - 01:40

I did miss that; I was more focused on the part about abusing the system to avoid the limitations of instrument change when actually changing instruments.

I'm not convinced mixer channels really are the right way to handle custom balancing of individual sounds. After all, it is just an accident of the 30-year general General MIDI standard that we happen to have separate patches for arco and pizz but not for staccato or any of a bunch of other articulations that we might also wish to be able to control the effect of. What we really need are more general controls over the behavior of all of those, not further setting in stone accidents of history like the way General MIDI happens to define sounds.

But also, it shouldn't really be necessary to do that sort of customization int he first place- the defaults should simply work correctly, which is to say, accurately reflecting human performance. You should only need overrides where you want to not reflect human performance. For that unusual special purpose, the velcotiy overrides do the trick for soundfonts, and coming "soonish", automation lanes for the more general case.

Reply

• Aug 2, 2024 - 01:49

Could be; I don't recall. Anyhow, again, if you wish non-reality, the velcotiy controls already provide that. So I'd really rather not see hacks like this 90's-style GM-iusm come back. Improved controls that are actually useful across a range of articulations would be far preferable, and that's on the radar as well.

Reply

• Aug 2, 2024 - 01:51

Basically, I use all sorts of samples of Mellotrons, ARPs, and other synths all over the place.
If I needed two presets from Mellotron and four presets from the same ARP, bam channel mapping, done.
Two mixer channels on the surface, six (eight?) when needed.
How exactly would you suggest achieving that?

Reply

• Aug 2, 2024 - 04:49

As I said, use instrument changes. You’ll have one channel per instrument, not one channel for each instance of the instrument as you would have using instrument changes in MU3.

And yes, as I also already said, someday it would be nice to be able to subgroup those channe,so so they could be grouped the way the old hack in MU3 worked when you abused it like you did. That was a nice accidental side effect of that terrible hack. So, someday it will be nice to have that side effect as an actual feature instead of an accidental side effect of a terrible hack. But meanwhile, at least instrument changes don’t constantly proliferate channels like they did in MU3 - it’s a huge improvement for your use case.

Reply

• Aug 6, 2024 - 11:22

Thoughts from Casper gave me a new thought:
https://github.com/musescore/MuseScore/issues/12827

Maybe a new approach is needed. My posts have fallen on reluctant ears in the past; I'm clearly not explaining just how useful this feature was, and how handicapping its absence is.

Slightly less relevant, but I clearly was enjoying this feature for a while now:
https://musescore.org/en/node/278497
https://musescore.org/en/node/278059

I still would like an add channel button, or add track if you will, an arrow which allows you to switch sounds without new clefs or transposition unless wanted, and then collapse them. Then add a staff text which could switch to it.
This would also fix my issue of not having direct access to pizz, arco, mute sounds anymore; I could just add my own if needed.
I think I just need to stop referring to it as a 'channel' and come up with a new term for it. 😉

Reply

• Aug 30, 2024 - 16:57

This is a big one for me as well.

I write a lot of music for accordion, and needed a way to have register switches actually sound differently in playback. (For those who don't know - accordions can switch between different combinations of reed ranks, effectively changing the octave/timbre of the sound that is produced.)

So, I found a custom soundfont that had different sounds for all the different accordion registers. I created a custom instrument with a channel for each of the registers, and then set each channel to use the correct sound. This means that in scores, I am able to use stave text properties to simply switch back and forth between different channels whenever I want - it works perfectly!

However, as far as I can tell, this won't be possible in MS4. I really want to make the move from MS3, but this is holding me back! Is there any way to achieve this setup in MS4?

Reply

• Aug 30, 2024 - 17:21

Use instrument changes instead of staff text. This gives you new mixer channels you can set the sound of, and the previous channels are reused where possible so you don’t have more channels than needed.

Only gotcha is your instrument change needs to be to a different instrument in order to get the new channel. So, don’t change accordion to accordion; change to piano or whatever. Doesn’t matter since you’re going to change the sound anyhow.

Reply

• Aug 30, 2024 - 17:52

Okay, thanks Marc - so whereas before I created one custom instrument with multiple channels, now I would need have several different instruments, one for each sound.

Are there any plans to bring back channels, or something like them, to allow for the kind of switching sounds we can do in MS3? They always seemed like such a slick way to do it, compared to instrument changes - which felt a lot more clunky to me, so I avoided them where possible. I'll give it a try in MS4 though, see how it works - if that's the only workaround, then I suspect I'll need to get used to it!

Reply

• Aug 30, 2024 - 17:53

Oh also - am I right in thinking that this means I'll have to set up an accordion RH and accordion LH in two separate instruments, in order for them to independently switch between different sounds? If so, it does feel like quite a clunky workaround for something that used to be very straightforward - in MS3 channels could obviously be switched separately on each staff within an instrument.

Reply

• Aug 30, 2024 - 20:54

It's true that you'd need separate instruments to have separate playback for the staves. I wasn't aware MU3 let you have different sounds for different staves other than by using different voices, but I guess you found a way? Anyhow, it's certainly possible that if someone requests a feature sand gives a good clear explanation of the real use case for it, a method of supporting that use case would be designed. Hopefully something simpler than the old hack of needing to create custom instrument definitions etc.

Reply

• Sep 5, 2024 - 15:20

Oh, I get it it - you have to start with an instrument that is pre-defined to already use multiple channels,
whereas piano and most other two-stave instruments don't. So more of a hack than an actual intended feature that you can apply directly. There is ongoing design work in MU4 toward supporting this sort use case more directly without relying on such hacks, so that will be better than the channel hacks for this purpose.

Reply

• Sep 5, 2024 - 17:21

Indeed, that hack is the only way to handle a certain set of real world problems in Mu3. in Mu4, most of those same real world use cases are handled much more elegantly - instrument changes, playing technique annotations, sound flags, etc. But for the very few use cases that are not already handled much better in Mu4, yes, it will be good to eventually come up with an equally elegant non-hacky solution to those as well. Which is why it is so important that people are able to describe their unique use cases in detail so we can see what better solutions could be devised to handle them if they aren't already handled better by the existing new features.

Reply

• Sep 5, 2024 - 17:26

My main use case is a closed score SATB and there be able to easily change the volime of S, A, T or B, independant from one another, for self-rehearsal.
Possible in Mu4? I won't (be able to) switch to Mu4 untill it is possible and as easy as with Mu3, as the members of my choir rely on this

Reply

• Sep 5, 2024 - 21:20

Not currently possible except via velocity adjustments, which is indeed not quite so convenient - takes maybe 5-10 seconds longer per use, which I guess could add up to a lost couple of minutes over the course of the rehearsal indeed.

But this is definitely one of those use cases where the notion of adding staff text with channel change instructions as a back door into makes no logical sense as a solution and is very clearly just a hack to make do with a system that was designed for an entirely different purpose (switching between arco and pizz).

In this case, the much more elegant solution is to not require any of that hackery, but instead simply add dedicated controls in the mixer to give independent volume (and sound, and pan, etc) per voice and per staff, always on every instrument. I would definitely be much more in favor of seeing a user-friendly feature like this be introduced.

It's conceivable that the automation lanes being designed might be coerced into doing something like that, which would be a somewhat less-hacky-than-MU3 but not-as-elegant-as-mixer compromise.

Reply

• Sep 5, 2024 - 23:16

It's not bold to call a hack a hack - it's just being honest. Yes, it's a known way to get the job done given the constraints of the MU3 system - a way to take a feature that was never designed for that purpose and shoehorn it into a totally unrelated use case. The point is, no one in their right mind would have designed that system this way as a method of solving that particular problem. There are far better ways of solving the problem that don't rely on mis-using a feature that was never intended to be used for this purpose and is totally non-intuitive and would never in a million years occur to someone who wans't specifically shown that method. Isn't it better to have a good solution that will appeal to people to never learned the old MU3 hack? Given a simple mixer control, or the need to mess invisible texts and dialogs and so forth, which do you think is the better user experience?

Reply

• Sep 6, 2024 - 00:46

I can absolutely assured you the channel feature was not designed for that use case. The channel feature existed solely for the purpose of supporting the change to arco/pizz and mute/open for strings and brass, and for many years, that was all it was capable of. It was very much a later afterthought that additional use of the facility was made to support guitar, and then someone had the idea to also retrofit this onto vocal staves and add special invisible S/A and T/B texts to trigger it. But it makes no logical sense for that use case - it was merely a quick and dirty way to get the job done.

For arco & pizz, of course there needs to be a text element there to trigger the sound - what else would human players have to go on? But there certainly didn't need to be separate mixer channel for tnis. That's only due to an accident of how the GVeneral MIDI spec was written back in the 1990's, where arco & pizza were separate "programs" but other articulations like staccato etc were not. So MuseScore dutifully added channels for those programs but not for any other articulation. The fact that arco triggers a channel change but a staccato marking does not is just an accident of history.

But that accident of history in no way suggests that people wanting to control volume of the alto part separately from the soprano part should be forced to resort to the same awkward GM-ism used for arco * pizz nut not staccato. Worse, it certainly shouldn't haven't required people to find and add invisible text elements to their scores to trigger all this.

It would have been far better had someone sat down and actually designed a more logical system for this. That is what we have the opportunity to do now.

BTW, what I am describing regarding the history is not a matter of opinion. I was involved in these design decisions. I'm not just guessing or giving "opinions" on what they were designed for. i can with absolutely 100% certainly assure that the original purpose of the channel model was to support the way General MIDI models arco/pizz and mute/open, and had nothing whatsoever to do with independent control over alto and soprano volume. That wasn't retrofitted onto the system until well over a decade later.

Reply

• Sep 6, 2024 - 15:06

FWIW, in your use case, personally, I'd be using explode to create separate staves and hiding them. Then you would have the combined staves for display, the exploded ones for playback. It's basically the same approach that people generally use now to have separate control over combine flute parts on a staff (or clarinet, or trumpet, etc), or to have separate parts when instrument in the score uses chords instead of multiple voices, which is quite common. And this is all the same (except easier to set up) as one would handle those cases in MU3.

Reply

• Sep 6, 2024 - 16:52

Yes, this is indeed what orchestral composers have been living with all along. Not ideal, but assuming you don’t do this until you’re done with any major work on the score, it’s manageable. Explode/implode greatly simplify the process of keeping things in sync if you make larger changes. And it does give you more flexibility in how you present - like if for some particular purpose you want open score format S A T B instead of closed SA TB.

Also, I would say it’s considerably less of a hack in that it doesn’t rely on using any features for something clearly out of the scope of what they are intended for, and these are all core features present in both MU3 & MU4. So less likely to break in future release.

But certainly, a straightforward way to take any instrument and have separate mixer controls per voice would be welcome for lots of potential use cases, and frankly I’m surprised no one seems to have submitted a GitHub issue for this (unless perhaps they did and it was closed prior to 4.0 release when that was the general way of putting things off for a future release instead of using tags & projects like they do now).

Reply

• Sep 6, 2024 - 17:30

...then I am clearly missing something.
I had thought sound flags was it, but apparently not.
Please explain:
Tell me the key words to use and I'll get right on it.
I just don't want it imeediately closed.

bkunda said recently:
"FWIW, we do have a concept for the sound flags feature that will, one day, extend to soundfont and VST. For the initial release though, its application needed to be limited to Muse Sounds for practical purposes."

Between that and you saying sound flags is a non-starter, I am very confused.
Maybe that is the point.

Reply

• Sep 6, 2024 - 17:40

What I am saying is that sound flags solve a very different problem from the one being discussed here. The subject at hand has been\, how to have different volume levels for the soprano and alto parts in a second. Sound flags will not help with that case, whether they can be applied per voice or not, because they don't give you control over anything but the specific samples they control. So for example, if that request were implemented, you could use it to have pizz and arco in different voices, which is occasionally useful. but it wouldn't allow you to control the volume of the two voices independently, nor would it allow you to set one voice to violin and the other to glockenspiel or whatever.

That is why I cannot possibly stress enough how vitally important it is that people are specific about the real world problem they are concerned with, so we can consider how to solve that specific problem. There are multiple different use cases being discussed here, and they may each have different solutions.

So still, we're waiting for as simple clear and direct statement of the exact real world use case you are trying to solve. Sounds like there might be more than one, and perhapoos for one of them, the pervocie sound flags are it. For others, it is quite clear than instrument changes are the solution even though they weren't so ideal in MU3. For other cases, though, it could be a new feature still would be needed to handle it problepy. I simply don't know which real world use case you mean here, but all I can say is, sound flags won't help with the one we've been discussing these last couple of days.

Reply

• Sep 6, 2024 - 17:50

I am referring to my original problem; Jojo's case seemed conveniently similar, but not enough apparently.
I also think the accordion problem brought up could be solved similarly to mine,

Give me a means to generate a mixer channel or several on demand with my choice of sounds and ability to tweak them. Allow the ability to assign one voice or several to this new channel.

Reply

• Sep 6, 2024 - 18:03

Unfortunately you've alluded to quite a few very different unrelated use cases, and given very fdetals on any of them, so I can't be sure what you mean here. Again, I cannot possibly stress enough that it is important to be as specific as possible about your real world use case. Since this thread is now pretty hopelessly confusing with at least half dozen entirely unrelated things being discussed, I would recommend starting a new thread, focusing on one and only one use case, with a specific attached, and a really clear description of the problem you are trying to solve - not the specific feature you happen to think might be a way to solve it. Focus on the *problem and then we can brainstorm the most appropriate solution collaboratively. And once a consensus is reached on the best solution to that one specific real world problem, then we can make sure a good GitHub issue is created with a clear description of the real world use case and the consensus regarding a propsoed solution. Then do the same for any additional unrelated real world problems that might well have different ideal solutions.

Reply

• Sep 6, 2024 - 18:05

I meant the one I started the trhread with; I chose not to re-state it.

To summarize: (quoting)
In many of my scores, I used and abused trumpet, bass, and synth parts to create multiple guitar and keyboard lines. Once I discovered this method, I never used Change Instrument again.
Many instruments without a growing mixer, perfect.
Please bring that back.

Reply

• Sep 6, 2024 - 18:07

You started the thread to two (entirely unrelated use cases, neither of which were accompanied by a score or any sort of clear description of the actual problem. Please, don't assume we are mind readers. Please follow the steps I outlined previously. I very much would like to help here but we need the sort of clarity I am asking for here.

Reply

• Sep 6, 2024 - 18:15

Quoting, again, the one that is most important to me:
In many of my scores, I used and abused trumpet, bass, and synth parts to create multiple guitar and keyboard lines. Once I discovered this method, I never used Change Instrument again.
Many instruments without a growing mixer, perfect.
Please bring that back.

I didn't provide a score; it seemed unnecessary.
Again, for longwindedness, please at least look in the mirror.
You write novellas as opposed to sentences.
I appreciate the potential help, but it does feel frankly pointless sometimes.

Reply

• Sep 6, 2024 - 18:22

I don't know how I can possibly be more clear about this:

A vague description of abusing trumpet parts to create multiple guitar lines is nothing even close to resembling a clearly stated use use we can all understand. There is nothing in that sentence that suggests why instrument changes or playing technique annotations or sound flags aren't already the answer. Again, please, start a new thread with a clear description of the actual poproblem you are trying to solve, with a score attached. Don't expect us to read your mind - be clear and precise and specific.

I also don't know what you mean about longwindededness, I never mentioned anything having anything whatsoever to do with that. I did say the thread has become hopeless long, not because of anyone person - on the contrary, because

And if my explanations are long, it's because it seems people don't understand my shorter ones, so I keep trying and trying and trying again to make myself understood. I am nothing if not persistent. I truly want to help. I am telling the exact steps to follow in order to increase your chances of success (from near zero to near certainty, I'm thinking). It's up to you to take that next step.

I look forward to seeing your new thread with your one clear description of the real problem and score attached so we can start a new discussion focused on that and only that.

Reply

• Sep 6, 2024 - 15:03

The reason to care is that by creating a well-formed feature request for a non-hack solution to the problem at hand, there is a much greater chance of it being implemented. I too would love to see a solution to this - not because I have any need for it personally, but because I can the value in it for a variety of use cases. Like, for instance, someone wanting to have two flutes on a single staff, and be able to choose the Flute 1 and Flute 2 sounds for the two voices. Of course, that still wouldn't solve the case of someone combining the two parts using chords instead of voices - and the same is true for choral music in MU3. But it would still be useful.

To my knowledge, there is no such request for general purpose per-voice mixer controls ioon GitHub. Seems that step on to getting a solution implemented would be for someone to write up such a request. And again, best chances of success would be to focus on the real-world problem and the best possible solution to that problem, not a request to simple reintroduce older hacks.

I'd definitely support such a request, and I think given the flute 1 / flute 2 use case, so would the core team.

Reply

• Sep 6, 2024 - 15:24

Sound flags are entirely different from mixer controls. They deal with entirely different use cases. Sound flags are nice for selecting between alternate sample sets within a single instrument in Muse Sounds (eg, the "non vibrato" and standard samples for strings), but they allow selecting totally different instruments (eg, setting your voice 1 & 2 to use flute and saxophone saxophones respectively), nor do they help with setting levels or panning, and they don't apply at all to soundfonts, only Muse Sounds.

So it's a fine feature request, but not at all related to the use case we were just discussing about choir rehearsal. That's why it's import to focus on the problem to solve and the best design to handle that.

Reply

• Sep 1, 2024 - 13:39

I don't know your situation, but I have 3.7 and 4.4 happily running side-by-side on my computer.
The only potential conflict would be that 3.7 would share/alter any other 3x settings if you have any.
Once 3.7 got so up-to-speed, I happily moved on from 3.6.2, so it wasn't an issue.

Reply

Do you still have an unanswered question? Please log in first to post your question.