How to access Piano roll editor in Musescore 4?

• Dec 15, 2022 - 14:07

Is is possible to open up the piano roll editor window? I can't seem to find the option
I use it to fine-tune dynamics, durations and the positions of specific notes.

Comments

• Dec 30, 2022 - 00:53

Wow! I just posted a message about the fact that soundfont features and effects missing from Musescore 4. I used pianoroll features heavily in my Buxtehude pieces that I transcribed to MuseScore 3!

I cannot recommend this software, because features did not get transferred to it from the previous version, and scores don't work at all right.

Please pay cose attention to the two fugues.

https://www.youtube.com/watch?v=Ith5gfo8NrQ

Reply

• Jul 13, 2023 - 03:05

4.1 is here, no piano roll editor.

I have been advised that my "nuances and customizations" are better off if I stay with 3.6.2 and sacrifice the beautiful score engravings (which, frankly I can barely see--I do music for listening.)

I'm afraid I'm a complainer, but really, I cannot use the 4.x software, in its current condition. I've already been told I have to sacrifice the nice reverb that was in 3.6.2 if I want those beautiful engraved notes, and do extra work if I want the reverb. When working with pipe organ music, I need the reverb to work in real time, because of the nuances.

I hope all this changes. I'm not giving up Linux to use Musescore 4.

Reply

• May 14, 2024 - 16:51

LIKE: "I do music for listening."

I can't think of any other use for music :)

To me, MuseScore is better than Finale only because it gives me control of the playback --- on a note-by-note basis via "piano roll." They took away the heart of this MUSIC GENERATING software! And now, it's only a MUSIC PRINTING software. Can someone tell me how, in MuseScore 4, one can do what piano roll allowed you to do in MuseScore 3?

As a result, none of the songs written with MuseScore 4 sound good when I play them back. For a piano piece, for instance, the left hand accompaniment plays as loud as the right hand (theme.)

Reply

• May 24, 2024 - 11:15

I don't think the main team realizes how much more important the piano roll was than musesounds is [arguably].

While I often write scores and then export them into a DAW for the final creation, I loved using the piano roll to inform my writing using the interpretation I'd expect to hear/desire, to make mock-ups without the need for a DAW or separate score, and setting up my scores for midi export in a much more straightforward/easy-to-manage process than what is gained from DAWs.

Musesounds has basically become a smarter VST with better native integration to elements you add into the score. It leaves little to no room for user control interpretation of sheet music outside of making a separate score which it 10x more complicated than working in a piano roll with at least being able to easily drag/ customize velocity, note start-time, note length, and perhaps fine pitch. [Without affecting the score]

The even more frustrating part is that different styles will call for different interpretation anyways, so investing more effort into making musesounds an even better/consistent tool that will likely serve as a middle ground for most popular styles feels pointless from the outside. It still seems better to put future development effort into returning convenient playback customization the hands of the user (using midi as the middle-man). This will open a lot more flexibility for playback / midi export... especially with plugins... meaning the effort used to 'fix' musesounds can be easily addressed in more streamlined ways anyways. And then users will be able to more easily control vst playback (when dealing with things like legato). It seems almost pointless to add VST support without piano roll capabilities when you think about why you use them over soundfonts...

I understand there could be serious things holding them back, such as not wanting to go with the previous editor and other things... but it's not even seemingly a remote priority and when people try to address it the don't get anything.

It doesn't need to be a DAW, but being able to conveniently access/ change velocity, note start-time, note length, and perhaps fine pitch on a grid without affecting the score, again, would dramatically improve the experience with playback in general (musesounds, VSTs, soundfonts) and midi export.

Sorry for the rant/vent. I just wanted to make some points and am somewhat happy to see I'm not the only one who feels this way. I'm hoping the main team is already looking at it, though. They're doing an amazing job. It was just a huge blow to my workflow to not have that (when so many other parts of MS4 are amazing/improved from MS3)

Reply

• May 24, 2024 - 19:38

I've indeed seen some of these efforts. Aside from that one response saying they'll contact Mark Mckay when they're ready to add a PR, it's seemed pretty one sided. So perhaps a dead end indeed.

Now I'm wondering if it would be possible to build a pseudo PRE using plugin creation... I'm not sure if it's possible to affect note playback without affecting the score, but I guess it's worth looking into. Or maybe something else can be faked using the Jack implementation that may come in 4.4.

Reply

• May 24, 2024 - 13:45

My guess (or hope) is that MS5 will be a re-joining of the fork that seems to have taken place, bringing together the best of MS3 (functionality, especially is MS3.7) and MS4 (playback and UI). Hopefully the piano roll editor will re-appear during the lifetime of MS4.

For me, lack of plugin access to note start time and note length renders MS4 a non-starter since it cripples the guitar "Let Ring" plugin.

Reply

• May 24, 2024 - 19:43

I think with MS4 running on a different version of qml, that wouldn't be possible, right?

And I guess that answered my question from above... Didn't realize there is no plugin access to note start time / note length...

I'm thinking using JACK's ability to play/pause to scores at the same time, a pseudo PRE may be made/ used to make modifications to a duplicate score behind the scenes as if you're only affecting playback (using a faked PR) but we'll have to cross that bridge when we get there

Reply

• Jun 13, 2024 - 07:09

I agree that there's not a chance of that a chance of that happening soon, with musesounds being a large part of that... but that's mainly due to the team not wanting to rebuild the same/ a similar system. It's not inevitably different due to conceptual architecture.

What would likely be needed to make this work is the equivalent of having two projects open (one for notation and one for mock-up interpretation) where the notation one is 'master' and the playback one in 'servant.' And instead of editing the 'servant' score directly, a piano roll is the middle man.
As far as I know, the old PRE simply made use of smaller divisions like 64th notes in order to change note length 'under the hood.' Instead of having a separate virtual score, the playback engine was directly accessible. So this simulated those micro adjustments that we get in most other PR. So the quickest way to get a piano roll would be some kind of score sync, with an editor to handle those... which is basically what was in 3.6, except the current score being directly edited under the notation (in the same way) instead of a separate score.

The problems I see that prevent this from happening is that
1. That's a lot of work when there other tasks determined as high priority to fix, edit, and implement
2. The team is not done rebuilding the playback engine how they eventually want to (which may be what you meant by musesounds making it inevitably different)
3. They don't think it wise to spend time/resources on something that will be replaced or revamped with the new engine

I'd prefer a temporary editor, and one that doesn't even edit the main score, but I doubt anyone else would make it so I'd have to try to figure that out myself. I've seen a user do some cool things with JACK to sync two instances of MS (which may be getting merged into 4.4) so I might be able to figure out a clunky workaround.
Probably not very soon though

Reply

• Jul 17, 2024 - 04:28

Very sad ... a terrible waste of a used-to-be great platform! Without something like Piano Roll, where fine-tuning can be made to every note independently, MuseScore 4 is a non-starter for me. I have downloaded it but have never used it. Gave up after a dozen attempts. No more can I tweak a piece of music to play back the way that I, the composer, want to hear it played. It appears (based on replies from MuseScore program designers) the software architecture of MS-4 may not have note-by-note adjustment capability. Why build something new if it's not better than the old one in every way?

Reply

• Jul 17, 2024 - 06:48

Seems like by fine-tuning focus into Notation / engraving and first perfecting that, they're trying to become equivalent to leading competitors in that specific way. (Also maybe for publication purposes / legitimacy?)

Seems like they eventually want to become a partial DAW à la Dorico or something but we may see MS6 come out before that is serious as it seems as though they really want to perfect notation. [And it would be an insane amount of work to be fair.]

That sucks for us who use MS3 as a production step and want that fine control before moving to the DAW... Or those of us who want to demo something as a style intends but not as literally as what's on the page.... but it's possible that it may be coming much sooner. MS5?

https://github.com/musescore/MuseScore/issues/19490
This at least shows great potential for things to be worked in step by step. But for now MS evolution is better for things that aren't solely for notation / with literal playback.
https://musescore.org/en/node/362360

Reply

• Jul 18, 2024 - 00:11

It’s not true that the architecture doesn’t support tweaking . It’s just that a full-featured facility to allow it - and allow it in a far better way than ever before - takes time to design and build. And as noted, relatively few scores rely on this feature, so the focus on was on features that virtually everyone uses a - fantastic engraving and default playback, etc. Making everyone wait for those amazing improvements while a years-long next generation piano roll editor was designed and developed would have been irresponsible and a huge slap in the face to the vast majority of users who would never find a user for a piano roll editor anyhow.

Reply

• Jul 18, 2024 - 23:36

"It’s just that a full-featured facility to allow it - and allow it in a far better way than ever before - takes time to design and build."

That's good for the future, and for people who will want to use the piano roll in all its glory (especially newer users) but older users don't even really need a fully featured/functional PRE. With the first priority being that the attack time (offset note position) and duration can be modified without editing the score [visually], and second probably just being half as use-able as it was in MU3 as far as input goes, that would hold the need for an official PR over until a well thought-out one can be implemented.

There are very few people who want a "years-long next generation piano roll editor" implemented ASAP. The problem was more so the fact that what was there has been lost, rendering 1. personal-taste playback adjustments, 2. as-needed basis edits [for specific styles], 3. use-special case adjustments (for Soundfonts, and now VSTs) virtually impossible without making a separate score for audio (and it's extremely difficult, still). Soundfonts and VSTs are crippled, and Musesounds sounds great in a few contexts, but is too inflexible to use for sound.

It does, however, compete well with Note Performer, which was probably more or less the goal. Which is respectable as musescore is a notation software in the first place, not a production suite.

Many who want the PRE merely want the afforded ability and only slight convenience which was taken away... not a shiny new system that everyone outside of this 'niche' would enjoy. (Referring to "vast majority of users who would never find a use for a piano roll editor.") It's understandable that that may be too much work or too low priority at the moment, but if the people who want to contribute cannot even do so, it just seems neglected. Something temporary (even for years), rudimentary, and even less visually pleasing would be welcomed so long as the function is fulfilled.

Reply

• Jul 18, 2024 - 23:54

But as I explained, putting in a half-baked system (eg, something as limited as what was in MU3) would just create a maintenance nightmare, as scores customized using the half-baked approach would need to be supported going forward, and that code would need to be left in place and made to interact with the more sophisticated code to come, etc. Not to mention, any time and effort spent on such a system is time and effort that could be spent on things that the vast majority of users would find useful. Sometimes in the software development world one has to make difficult choices in the best interest of users as a whole even if it inconveniences a small handful temporarily. But, it's not even much of an inconvenience, because of course if you value the piano roll editor over actually good basic playback or engraving, you still have the option of using MU3. So really there was not much downside here.

Reply

• Jul 19, 2024 - 07:27

Naturally, MS evolution/3 can still be used, which is good. Minus a few things (which I see will be addressed soon on the To-Do, like the arrow keys moving the selection instead of the play-position), musescore 4 does make the writing process a lot easier. Especially when dealing with parts. Otherwise, yes, MS3 still is a great option for nice scores and actually good basic control over playback. MU4 does work better/more conveniently in many ways so actually good 'out of the box' playback and engraving might be worth not having the Piano Roll for simple tasks that aren't meant for production/mock ups. I guess I'll continue to use both as I have been. I'm sure others do as well. Can't wait to fully upgrade though, hahaha

As far as the reasons not to do it, such as the code maintenance being too much of a nightmare/distraction, that's understandable. It's natural that there's only so much that can be done and an order that things most efficiently get done in. Of course I don't know how much work getting the parameters for offset note position and duration to work with musesounds (or with MuseScore outside of musesounds) and perhaps it's not worth it at the moment. MS3's PRE was undeniably a bit clunky and reintroducing that 'as is' may be even more morbid for maintenance but specific elements of playback that it gave access to seem more accessible than rebuilding the whole thing. Especially if the method was essentially shifting a note over using the smallest subdivision.

"So really there was not much downside here" The downside is MuseSounds and VSTs which would be immensely more useful for production with that extra control are not yet able to do so. VSTs will likely go through a DAW anyways, but without Musesounds functioning as a VST there's no way to easily make those modifications yet. I digress. I know musescore is not meant for production even if it incidentally helped out in the past, but the frustration and struggle likely comes from having to choose between what's been improved and what worked in the past. At least midi import/ export is still a thing, eh. We'll keep going back and forth for now, it seems.

Reply

• Jul 19, 2024 - 13:31

By "no downside", I mean, compared to how things were before the release of MU4. That is, before the release of Mu4, you couldn't use Muse Sounds or VST, and nothing has changed in that respect. Also, had the release of MU4 been delayed until the necessary architectural changes were made to support a "real" piano roll editor, we'd all still be waiting. So you'd still be no better off.

Reply

• Jul 19, 2024 - 14:21

That is almost my whole point. It's like most or at least some of the key workflow improvements don't exist. Either or. Just a matter of perspective perhaps. It's like living in the snowy mountains having used an older 4wd vehicle, and finding out the new one you got which is lighter, (otherwise) safer, and more powerful only comes in rwd. Now you have 2 context sensitive vehicles. 2 is better than 1 eh? Have fun I guess. I'm trying to haha.

The last bit of the point is that it's not a "real" PRE that is desired as much as the flexibility that comes with it... Once again, addressable by two parameters/properties: [playback] Note Position Offset [aka the attack] and [playback] note duration. To be edited in a similar manner to "Velocity" and "Tuning (cents)."

With access to those two, it might be doable for users to make a faux/makeshift PRE in the form of a plugin. But if those two parameters are part of the reason it would be so difficult to maintain, then that's very understandable. We'll continue waiting. Personally, I'm hoping that those two properties can make it in with the addition of automation. I'm guessing that would probably make more sense on the dev side.

Even if it still seems niche for now, the conversation around the features at least shows it's not necessarily in low demand. Although the count of 10 people on this post is not objectively a lot, [hahaha] I've seen other posts asking for it as well. I respect the priority for some of the other features that are coming up, but you devs won't know what users don't talk about. In debating further, I'm just trying to pinpoint what exactly is being dismissed here... until enough users chime in for it to become a priority or until most tasks that needs to be done first are eventually done.

Reply

• Jul 19, 2024 - 14:58

The reason I am saying the existing PRE is too limited to seriously consider for inclusion in MU4 is precisely because it is not sufficiently flexible. Yes, you can tweak note on and off times, but not across bar lines, not in a way that would move notes in and out of tuplets, etc. A different system would need to be designed and implemented to support that and more, and then trying to support both the new and old systems is the sort of nightmare that the developers rightly want to avoid. It’s just not worth trying to support the old limited form of offsets. To do it right will require a significant redesign of the internal data structures used to represent scores, which the team has stated repeatedly they absolutely want to do - but it’s a huge change that will take time.

Reply

• Jul 19, 2024 - 22:43

I could actually see that making it into MuseSounds sooner since MuseSounds has deeper integration with the whole software than soundfonts and VSTs which are crippled. But I guess we can't be sure that'll be looked into, given it was a plugin in the first place. Don't know if you've already requested it on github or not.

That seems like it could be done without a refactor of the playback system, though. Good luck!

Reply

• Jul 20, 2024 - 23:45

@yonah_ag wrote:

"Lack of duration tweaking is a complete MS4 show-stopper for me since it cripples automated, realistic guitar sustain as per the TAB Ring plugin. So I'm sticking with 3.7 ..."

Ditto. MuseScore 4's deficiencies (and UI) force me to stick with MuseScore 3.7 Evolution. MS 4's vastly improved—though largely problematic—sounds are enticing, but for my work they do not bring enough to the table.

Once I've finessed a score's guitar durations and velocities it's truly disappointing to hear the same score in MuseScore 4 because:

a) the MuseScore's nylon guitar isn't that great, and like other MuseScore guitar sounds, its emphasizes overplayed notes and string buzz in an attempt to sound realistic. I achieve the sounds I want with Native Instruments' sampled Picked Nylon or PianoTeq's modeled guitars ... and these options provide various parameters for modifying the sound, where as the MuseSounds guitar do not.

b) timbre aside, to my ear realism depends on varying volume from note to note AND durations (in Piano Roll Editor speak those are the Len property values that MS4 ignores.) MIDI sustain (aka Let Ring, aka Laissez Vibrer) isn't nimble enough to avoid excessive sustain. The TAB Ring plugin is the answer.
The lack of note velocity is simply death to expression. That is not disputable. Why else would velocity be in the original MIDI spec?

c) there are just too many problems with MuseSounds: The detaché violin is saddled with baked-in portamento and reverb. The choir ends sustained notes with a "Huh." The acoustic guitars ... meh.

I look forward to a time with these issues are resolved and I can embrace MuseScore 4. And regarding MS4's user interface, I'm well capable of learning new systems, but also I know awkward when I see it. Too many of the properties I frequently used are buried in draws that won't stay open.

I'd love to take advantage of MS4's many improvements and wonderful new features but I'm dissuaded by the UI and sound quality, primarily in my main areas of interest: guitar and strings.

Reply

• Jul 23, 2024 - 01:29

How do I download MS-3.7??

Following the instructions on the Jojo-Schmitz wiki page, I clicked "Actions" and found myself on this Github page, with a list of 1047 "workflow runs" today ...
https://github.com/Jojo-Schmitz/MuseScore/actions

The top-most 3.x workflow with a green tick happens to be Update Italian translations #2734.

In the Artifacts section, I found MuseScore_x64_10040217710 (as well as the 32-bit version.) So I downloaded both versions. So far so good, but when I unzipped either one, there's no .7z file called MuseScore-3.7!

Have I chosen the wrong Workflow Run? I did try other workflow runs but got the same result. NO MS 3.7. Can anyone kindly point me in the right direction? Thanks.

Attachment Size
Screenshot 2024-07-22 093142.png 43.9 KB

Reply

• Jul 23, 2024 - 07:21

Just downloaded the latest build and here's how mine looked. I don't think it should be a hidden file.

This may be weird, but try downloading it on a different device and seeing if that changes anything. Then rezip it yourself and put it on a flash drive or in a cloud drive service like google drive or something. I'm suspecting you may have an issue with chrome or an antivirus (or it may be as simple as your file viewing settings [hidden items, extensions, etc].

Hope you figure it out

Attachment Size
Screenshot 2024-07-23 021600.png 21.29 KB

Reply

• Jul 21, 2024 - 14:07

Like yonah_ag I also use MuseScore versions 3.6.2 and 3.7 side-by-side without issue.

But to the best of my knowledge 3.6.2 and 3.7 do not share preferences or palettes, and they each have their own plugin repository. So you just have to address those details manually if you want that level of consistency between the two apps.

Reply

• Jul 23, 2024 - 09:20

Jojo-Schmitz wrote:

[MS3.6.2 and MS3.7 Evolution] do share their settings and plugin repo

Ah ... my palettes are different between MS3.6.2 and MS3.7 and that led me to think the settings/preferences were unique to each app.

Likely it means I edited a palette in one version, didn't export and import into the other app.

Interestingly the Plugin Manager shows the same repository in both apps, but the checks therein are unique between apps, and thus my plugin menus are unique.

Reply

• Jul 23, 2024 - 17:17

I haven't extensively examined the palette situation but it seems the main difference is that my custom palette (ones built from scratch) are not synced between MS3.6.2 and MS3.7. Nor is the state of the disclosure triangles on standard Palettes.

Also noticed that the "Open recent" menu is different between MS3.6.2 and MS3.7.

Won't have time to test further today.

Reply

• Jun 13, 2024 - 05:13

It's not on the cards for 4.4 (see https://github.com/orgs/musescore/projects/34).
See https://github.com/search?q=org%3Amusescore+piano*roll&type=issues

But understand that it's not just implementing the piano roll itself that's the problem, it's the fact that the underlying playback engine no longer makes use of the same micro-adjustments to note-starts and ends and other controllers (dynamics etc.) that you used to be able to use (particularly when using Muse Sounds).

Reply

• Jul 17, 2024 - 20:23

For personal reasons, I cannot spend time on the music projects I work with. However, i periodically tune in to the JojoSchmidt Evolution pages and get a copy of the latest one that works. I have absolutely no interest in MuseScore 4, and I am extremely disappointed in what the "company" has done to the project. They have sacrificed all of our other needs for beautiful, engraved musical scores. Sorry, but I have very low vision, I love Linux, and I want to hear what I'm working on without having to bolt on all kinds of bits and pieces to get what I have with MS3 and Evolution. As far as I'm concerned, the "company" has abandoned everybody that put in all those contributions which made MuseScore 3 so great. If I can contribute in some way to Evolution, maybe even monitarily, I will. But no more support for MuseScore 4. I have contributed in the past, but no more.

Reply

• Jul 18, 2024 - 00:08

I am sorry you are feeling that way, but as a fellow user and contributor to the project, I think you are pretty grossly misrepresenting MU4. Sure, there are absolutely fantastic engraving improvements which exist not just for your benefit but for the benefit of all the people who may read your score. But also, there are absolutely incredible playback improvements via Muse Sounds, and huge usability improvements as well, not to mention tons of new features, etc. Saying that they have "sacrificed all of our other needs” is just flat out wrong, and disrepectful. True, the piano roll editor hasn’t yet made it in, because doing a good job of it will be a big undertaking and a bad job of it will just mean maintenance headaches going forward. Plus, it’s a feature used by a very small minority of people, so focusing on the features that are actually used by most makes more sense. That’s exactly how a responsible company manages software - focusing on the needs of the many. That can occasionally meet the few might need to wait a bit longer.

Reply

• Jul 19, 2024 - 21:10

Hello Marc, thanks for chiming in. Many comments in this thread indeed sound negative, but I'm positive there's no malicious intent to trash MU4. Rather, I think the satisfying experience we've had with MU3 has set high expectations for MU4. So, naturally we'd expect it to expand on the same functionalities that we've come to love about MU3. If Piano Roll Editor was only used by "a very small minority of people", then I posit that these are folks who take the PRODUCTION of their music very seriously.

Take myself for instance: I switched over from a FINALE product because MU3 satisfied my demands perfectly. See, I don't just want a beautiful notation tool (which FINALE did admirably) but just as importantly, I expect playback capabilities that makes music sound like "MUSIC"---i.e., played by humans and not robots. I used Piano Roll Editor extensively to tweak my scores note-by-note, controlling Speed (volume), Duration, and Start & End positions of each note. I want to play back my scores with EMOTION, with that HUMAN TOUCH.

Thanks to MU3 with PRE, I was able to realize a 20+ year dream: Music that I composed over the past 20-30 years finally came alive, as if I were a very good pianist or violinist giving a concert. Here's what I mean if interested, and if time permits. (A 8-minute violin-piano duet. Note also the exchange of comments in the side column.)
https://musescore.com/user/38102135/scores/7451519

Therefore, despite many enhancements in MU4, I'm forced to stay with MU3 until PRE is implemented in a future edition of MuseScore. However, I may simply be unaware of new features in MU4 that achieve the same goal as PRE. If so, is it possible to run a tutorial (similar to your MuseScore 4 Jump Start) or even just a Q&A session with focus on how to tweak a score to "humanize" playback without PRE?

Reply

• Jul 19, 2024 - 23:16

I think it is pretty dismissive of those who don't rely on the piano roll editor to say that only those who do "take the PRODUCTION of their music very seriously".

For the record, many of us take the production of the music extremely seriously. Which is why we spend fasr more time getting the actual notes and markings to be as good as possible - focusing on creating the most engaging and well-developed melodies, interesting harmonies and counterpoint, clear and effective technical and expressive texts, and avoiding awkward doublings, bad voice leading, imbalances due to poor register choices. We also take pain to make the engraving as clear and beautiful as possible so the human musicians who read the score will have an easy time of it too - proper spelling of rhythms and accidentals, appropriate note spacing and positioning of other markings, well-chosen system and page breaks, etc. We also might want the default playback to capture as much of the nuance of real instruments as possible, so we want accurate samples with natural sounding responses to articulation, dynamics, and tempo markings. We might add the occasional invisible tempo marking or dynamic, but if we've done a great job of notating things, that is seldom necessary except for extreme rubato passages.

I do very much look forward to the automation lanes that should be coming in the not-too-distant future for much more natural tweaking of things like volume or tempo fluctuations, and I'm much more likely to make use of that feature than any piano roll editor. But I would submit that 99% of the music I hear where people have spend hours tweaking details in a piano roll editor, would have gained a ton more by paying the same amount of time on all the other sorts of details I mentioned above. And in the case of music that was tweaked in those ways was but still limits itself to use of soundfonts, in every single case without exception I have ever heard, the default playback in MU4 is an order of magnitude better right out of the box than the most heavily tweaked playback in MU3.

So anyhow, I categorically reject the notion that spending more time tweaking in the piano roll than on the any of those other things is a sign one values music production more than others do.

Reply

• Jul 20, 2024 - 03:18

RE: "take the PRODUCTION of their music very seriously"
Just wish to clarify my comment above. By "production of my music", I meant having MuseScore produce it, not musicians. It'd be really nice for some skilled musicians to want to read and play my compositions, but I have to honest: It's not likely! MU and I will probably be the only ones playing my pieces.

I play piano & guitar myself, but I like writing music for string instruments too. (I've posted a Violin-Piano duet and a Guqin piece with string ensemble.) If not for capable software like MU3, I'd still only be hearing them played inside my own head. That's why I praised MU3 for letting me realize a 20-year-old dream, and I thank the MuseScore team for that. At the same time, you can understand my disappointment that those tweaks aren't available in MU4, as you've just confirmed.

I understand a software company has to prioritize its R&D. (I've run a video conferencing software startup years ago.) So I just want to clarify here what I meant by "music production." As far as the subject of this thread goes, I'll just leave it at that, and trust the MU folks to continually bring useful enhancements.

Reply

• Jul 20, 2024 - 03:56

I would say Muse Group takes music production - the playback of the score - extremely seriously. That’s why they invested so heavily in the Muse Sounds libraries - the ones originally released two years ago, new ones added last years, more coming within the next month or so, etc. That’s also they implemented so many new playback-related features like tempo changes, and coming in 4.4, ability to set dynamics per voice,, and coming at some point after that, automation lanes. music production is in fact one fo the two biggest area in which MU4 blows away MU3 (the other being engraving).

So it’s too bad you’re limiting yourself to only the tweaks that are possible in MU3, because MU4 is capable iof making your scores sound so much better in so many other ways - just not that one specific detail. But, certainly your right to stick to what you’re familiar with.

Reply

• Jul 19, 2024 - 23:22

So all that said, I would just add that, no, there is no replacement for the piano roll editor specifically, but the default playback is already much more human-like on many levels, and the response to the markings that you'd also want to add for human musicians - as well as the interface for adding them - is greatly improved as well. With the end result being, it's much easier to get results that are overall better than was ever possible with MU3, even if certain tweaks aren't possible currently until new automation lanes and others are implemented.

Reply

• Jul 20, 2024 - 01:08

Just saw the response although I chimed in below;
"So anyhow, I categorically reject the notion that spending more time tweaking in the piano roll than on the any of those other things is a sign one values music production more than others do."

I don't think that was meant to say all others don't take production seriously, but there are different production oriented workflows where certain aspects are more valued than others. Humanization, in this case (as I and as I assume nimbears means), is largely synonymous to that extreme rubato and capturing stylistic deviations from the score that often come with specific styles. In MS3 Jack was an option for those of us who have VSTi software to use that. While musescore 2/3 was rough out of the box, I was able to find a tutorial (from I think 10 years ago) showing how to set that up and used that since 2017. Before that I was enjoying testing many soundfonts.

"I think it is pretty dismissive of those who don't rely on the piano roll editor to say that only those who do 'take the PRODUCTION of their music very seriously'." Using what comes out of the box means that you want a better process overall but are less concerned with individual works in terms of the production. It's close enough for you. He's basically saying that it's not close enough for his own purposes in that statement in saying "their music." One is looking for swift/convenient realism and automated/approximated expression through the beauty of the actual sound/ written music and the other finds the beauty in the deeply personal interpretation of the written music. One is not objectively more important than another, but depending on what you're looking for can be more important to one's workflow. In this case, the extra ability to add personal human expression. Musesounds is very beautiful. I'm looking forward to what automation can do in that regard when it is developed.

My workflow largely consists of writing in Musescore (because I prefer notation to PR for the composition/ input process) and then exporting the midi to the DAW. I'm sure others do as well. Having the piano roll in MS helped tremendously because if I changed a section of music that appeared in multiple placed, I could make those changes in the MS piano roll and copy/paste those changes with the offset positions and continue to make light tweaks.

This was important because if I made those changes in the DAW piano roll, I can't import that midi back into musescore without affecting the score. So I'd be forced to either do a lot of chopping/comparing midi or starting over in the DAW after a small change.

Other times outside of production when working with a less experienced musician (often other students), mock-ups that performed some of the rubato-esque stylistic cues helped to communicate what was needed to 'sell' the passage in performance. Having to move to a DAW for something that isn't even production related means any changes will have to be chopped up again without the PRE.

As you know this can be even more relevant when creating jazz / RnB style lead sheets for people that aren't necessarily used to those styles and don't understand how notes may be placed/ felt differently. Yes you can try to write out separate lead sheets for those who get it and those who need the interpretations 'spelled out,' but then they aren't really learning anything about interpretation. Phrasing can be simplified to what we write on the lead sheets for those who know how to manipulate it and interpret it creatively, but otherwise a mock=up will need to be done. Musesounds is more than expressive for classical music as it is normally interpreted by the sheet. Other contemporary styles are very well covered too.

And of course with little to no setup as well, it becomes much more accessible to have better sound for many users, which I presume was your main point.

Kind of like how many apple-designed products 'just work' out of the box, but things that are designed on other platforms (depending on the device) often grant more accessible freedom, customization, and capability.

That's not to say that anyone here thinks they care more about generalized production than other users in general, but depending on the purpose of the production, and how much flexibility is desired with tempo, and interpretation, there are styles of production that take into account much more than what's on the page... especially when you use styles that aren't the strong suit of what comes out of the box.

There's no intention to belittle you, musescore, other users. just pointing out possible use cases that are meant with certain statements. Hoping to continue a productive discussion.

Reply

• Jul 20, 2024 - 02:10

The notion that the results I get are “close enough” for me but not for someone else is what comes off as belittling, whether intended or not. I’m happy to assume it wasn’t meant that way, but regardless of how it’s meant, I still reject the premise. I just listed a whole bunch of quite specific reasons why: the results I typically hear from people who take that stance might have nicely tweaked on times and off times, but the results still come off as much worse overall because they aren’t paying as much attention to those other details. And if they are limiting themselves to sound fonts, its still not going to come close to sounding as good as Muse Sounds does out of the box. Which is to say, none of those subtle ontime/offtime details will actually matter.

So, I would say I take the production fo the music very seriously, and that’s why I prefer the sound of music that pays attention to all of those details, and doesn’t focus on ontime/offtime adjustments at the expense of what is ultimately more important. The music held up as examples of the value of the piano roll editor might be close enough for the the people making those claims, but it’s not even close to close enough for me.

Now, in an ideal world, there would be people who care about all of this and give equal attention to all of these details. But such people would still be in the position of having to decide between the limitations of MU3 and the limitations of MU4, and I still suspect that for most of them, those who choose the limitations of MU3 would be producing music that doesn’t sound as good to me as the music of those who choose the limitations of MU4.

Reply

• Jul 20, 2024 - 03:58

Ah, I see. When I say "close enough," I mean for what you're doing, even if you can't change the result much. I don't know how to say that in a way that won't come off as insulting to you.
That is not to say you are writing the same 'level/type/difficulty' of music and others wanting PRE want more detail than you; it's saying that it's a different process that desires more from a specific area that doesn't seem affect your scores. Apologies for the delivery, but the meaning would be belittling if I implied your work was somehow not advanced or expressive enough to warrant that level of control, when I simply mean you don't feel a need to express yourself/work in that way. (And the point being that it still sounds good, of course. Not even trying to make it personal. I'm not saying MSnds sounds bad... just that it doesn't offer the same control. Highlighting the difference in process / workflow)
Edit: Think more like, your 2 seater sports car cannot as easily tow another vehicle or bring 4 kids to school in the morning. Different people have different goals/ needs. That's what I was trying to say.

I've used VST's in MS2/3 with Jack so soundfonts notwithstanding, one could get an impressive sound out of the older MS with VSTs. Albeit, with potentially much more time spent as you said (which is indeed a notable trade-off that musesounds covers well). But after setting it up initially it becomes much easier. (And musesounds is amazing for freeware which is objectively much more important for most given how much money is spent on good sounding VSTi's.)

One of the differences in my workflow in yours which greatly changes this narrative is that (especially in the context) I wrote in musescore with the intention to export the MIDI to a DAW anyways. Using MuseScore's PRE allowed me to edit the composition/notation easily and reimport that into the DAW without worrying about needing making those changes again. So perhaps since I'm going to spend that time anyways making how I want it, I don't see the full benefit of that saved time that only helps with specific projects or parts of certain projects [and creates more time going back in forth in other parts of my workflow].

For me the PRE was just a tool to visualize the offsets/duration edits I was making to notes already there and rarely used for note entry, so I get why you found the PRE unwieldy if using it like a PRE in a DAW.

Of course if you're composing all within musescore for a quick result than can still be used for professional production works, musesounds can definitely deliver. But like many tools that save time, it can only do so properly in so many contexts. If you're going to use a DAW anyways or you're using different stylistic contexts, that saved time and the impressive results aren't truly applicable still.

But for orchestral works I do often use MuseSounds in this manner: https://www.youtube.com/watch?v=HYEGwQ-JQrA
I took note of this video when I saw, as it was essentially what I was doing already but showed some extra tricks. And to be fair, I do export musesounds playback as audio to use in the DAW as well sometimes. But that means I'm no longer writing with notation in mind most of the time. Which brings us back to interpretation and the PRE.

Reply

• Jul 20, 2024 - 14:07

What I mean is that I think my standards are quite high - and that's precisely why I virtually never find the results of playback from MU3 - hand-tweaked or not - remotely as satisfying as the playback from Mu4. i don't care if things have been hand-tweaked or not - I care if they sound good. I'm interested in using any and all tools available to achieve that result. I don't care how good the hammer in MU3 is, if you don't also have a screwdriver, wrench, drill, utility knife, and other tools, you're still not going to achieve results as good.

So it's not about the music or whether it is 'advanced" or not - it's a recognition that there are multiple aspects to making a score sound good, and focusing all attention on one of them at the expense of all others may produce results you personally find "close enough", but me and probably the vast majority of listeners will probably still find those results inferior to what is possible by focusing that same attention on other aspects of the music. And that's OK, everyone has different tastes. It doesn't have to be a contest between whose tastes are better.

Reply

• Jul 20, 2024 - 15:29

Nobody said your standards are not high, Marc. Only saying you don't care about control or expressing your tastes through rubato styled expression. Using your tools analogy, Musescore 3 actually gives you a whole toolbox to work with [allows you to bring in new tools and take your work out to different workshops], and MU4 is saying, "you don't need these tools, send the designs to us and we'll build it for you." The result will definitely be good, but if you desire something different than what Musesounds outputs it simply won't cut it. You'll still need your tools to refine those parts [which we don't have conveniently accessible]. Note: I did not say better, I said different. Just because you didn't choose to add on the right extensions to your tools for your purposes previously doesn't mean it was the same for others. You didn't feel sculpting was possible with MS3. Fine. Others could hammer at that stone and find beauty in something you didn't. Fair. That's your assessment of good art/ music. I personally went out and bought a chisel and other tools, so I was able to make more than what the hammer of MS3 provided out of the box. IMO more easily than I can with the MS4 'CNC router.'

"So it's not about the music or whether it is 'advanced" or not" - I quite literally said that; it's a matter of different goals and different output. Unless you uploaded your own audio to MS.com you weren't going to get the realism you desired anyways. But that's why I used the Piano Roll to export to a DAW. I've already footed the bill for my VSTs which may not produce good music as quickly as Musesounds going from score to production, but I've been able to get *my own* desired result a whole lot quicker... And not sounding worse, thanks to having cinematic studios instruments (most conveniently) among other libraries to produce professional sounds.
This was in MS2 https://www.youtube.com/watch?v=Qj92FiHBRi8 and in Musescore 4 [only the guitar] functions largely the same way but just more accessibly in not having to use JACK. Musesounds still requires a lot of haggling to get better results for my own purposes. Which I'd be better off doing in a DAW with Amplesound Guitars Instruments. "recognition that there are multiple aspects to making a score sound good" Using a DAW, Piano Roll, Midi CC Automation, Keyswitches, and paid VSTs does grant one those other aspects. Not only to sound 'good' by general standards but to create a performance YOU desire. If you do desire a specific one. It's like being involved with gigs/bands that you prefer to play with.

Using MS3's piano roll I could compose in notation and edit the Piano Roll in musescore with convenient copy/paste of the notation with those same tasteful changes. If you watched any of the video linked previosuly, you'd hear "5 different conductors, 5 different performances." It doesn't cut it for others' preferences, and becomes a distraction... which is why it currently necessary for some to use both MS4 and 3.

You have been the only one taking taking/implying statements as indications of someone's tastes, tools, or writing as being objectively better. I'm saying that they are only better in/for certain contexts/ purposes. Musescore is not a DAW, does not work as conveniently as one, and I don't expect it to. For my own desired results, I still get better results from the tools I paid for outside of musescore more quickly than with Musesounds when doing anything that requires more detailed tempo automation, where I use rubato, and where I go for a human performance/expression outside of just realism/ parameters that create results someone else would desire in Musesounds. It's a no brainer that paid tools still may work more efficiently when using them for all the fine adjustment abilities they grant you as intended.

Musescore was simply a wonderful step for me to express those ideas as notation and then make it sound how I want without ruining that. Which I can't easily do in MS4 and still be able to edit those ideas. Marc Sabatella, people don't use all tools the same ways.

For the record - if a slightly more flexible version of MS4's parts system could be backported to 3.7, I'll shut up

Reply

• Jul 20, 2024 - 18:50

I found this vid in like 2016 and took a shot at getting it set up in 2017 and at the very least, some things sounded better than soundfonts. At the time I eventually found midi export to be the better option and that is still the case in MS4, hence why I want the Piano Roll back lol.

I loved transcribing / writing in the cheesy 80's style and using the Korg M1 VST to capture some of that old style and in MS2/MS3. It allowed me to transcribe and practice the lines I was taking with a little more warmth than soundfonts provided. JACK was really helpful for that. Not the most convenient though so I only used it sometimes, and had soundfonts to take place while arranging.

Hypothetically you could make amazing arrangments with MS's Piano Roll and those VST's assuming you had automation done in your DAW synced up with musescore via JACK. MIDI export just made more sense as VST's often have more flexibility within the PR of the DAW. Musescore's PR was great for composing and not having to do any tweaks in the DAW's PR though. Great for orchestral stuff

Reply

• Jul 20, 2024 - 21:54

When you say math modeling, do you mean similar to how SWAM tech uses math? Is it lighter on RAM? I'm curious because I expect it to eat up some CPU, but less samples can mean less problems.
It sounds amazing. Watching a YT video of the functions and seeing the lua scripting capabilities makes me want it even though I don't know much about classical guitar hahaha. It sounds gorgeous! What are you using to host it?

Reply

• Jul 20, 2024 - 23:36

Claasical Guitar is a recent addition to Pianoteq and the only reason that it interests me. However, the pianos sound just as good and also offer a lot of customisation. I don't know the details but the programmers at Modartt model physical instruments mathematically. This means no samples and lots of customisation such as attack, decay, sustain, eq, reverb, fret noise, amplification.

It can be used in standalone mode, (e.g. to playback MIDI files), or as a VST plugin. I use it with MS MIDI Out for real-time playback of scores. They say that it uses more CPU than samples but my 13 year old dual-core PC manages no bother with CPU at around 25% and no latency.

They offer a fully featured free trial with occasional notes missing so you can get a really good idea of how it works and sounds.

This shows my MS setup with Pianoteq:
https://musescore.org/en/node/358470

Reply

• Jul 21, 2024 - 14:19

It is false to say I don't care about control or expressing my tastes through rubato-styled expression. I do care, but I achieve these results through more "normal" means.

It is also false to say that MU4 provide no such tools - obviously it does, via the normal mechanisms of tempo markings. Similar for dynamics and articulation.

As for me being the only one dealing in statements of someone's tastes being objectively better, well, this discussion started with me disagreeing with the implication of a statement I found dismissive. And continued when you stated that I "want a better process overall but are less concerned with individual works in terms of the production" and that I am only to obtain results that are "close enough for (me)" but close enough for others. That is again, dismissive, and I'd prefer that rather than try to defend these sorts of statements, you simply acknowledge they are inappropriate. You could, for instance, say instead, "we all value production, but some people care only about tweaked ontime/offtime values while others care more about that long list of super important things you already mentioned that others choose to ignore".

Which again, is my point - that these sorts of adjustments are but one of many things that contribute to a truly musical playback, and many of the other things that are important are possible only in MU4, not MU3. Plus MU4 automates many of the things that previously required manual tweaks., so even many of the things that people used the piano roll for are just lo longer necessary.

So you have to decide between performances that have these tweaks but lack almost every other thing that makes the playback sound good, or have those other things but lack those tweaks for the things that aren't handled automatically. I personally feel that in almost every single case I've ever heard where someone chose the former and posts the result as evidence of the superiority of MU3, the results would very clearly have been far superior overall with less effort in MU4. And I suspect 95% of listeners doing a blind listening comparison would agree. Possibly even the people making those hand-tweaked MU3 versions would agree if they actually did the experiment themselves (and put in an honest effort to use appropriate techniques to optimize the MU4 results).

So I still would encourage people who purport to care about the "production" of their music to at least consider the possibility that they can actually get results that most people will find superior by using MU4 even though it lacks the piano roll editor, and that they might end up agreeing.

Reply

• Jul 22, 2024 - 04:20

"we all value production, but some people care only about tweaked ontime/offtime values while others care more about that long list of super important things you already mentioned that others choose to ignore"
correction:
"Some people want to use that super long list of things you already mentioned along with tweaked ontime/offtime values." No need for that much passive aggression. People value different things. They may ignore it because they tried and didn't like how it sounded/what it did to the score, or it was far too tedious for their purposes.

Please don't twist my words or meanings, marc. I'm being as direct as possible.
If the other things you mentioned were gone, I'd either continue to do them in my DAW, or I'd be here talking about how maybe having to use 5 different invisible tempo markings to change accel/rit. should be addressed on the to do list. Or I may discuss the clunky parts system of MS3 that is greatly improved/revamped in MS4.

BTW - rubato is often done with a soloist while an accompanist is still playing 'in time' and following shortly thereafter. This is not achievable with tempo markings in my understanding, though I may be wrong. [Meaning they are playing in separate tempos more or less (it's less) that eventually meet back up]

Similarly, adding 50 invisible markings is simply a less efficient way to make changes than editing midi. I look forward to that changing in MU4 with the advent of automation, but believe me, I already used these in MU3 along with the piano. The difference is that in MS4, they directly affect playback with musesounds; but for me this is pointless because Musesounds cannot do what I want it to do while respecting the notation output I desire. I need to make 2 separate files. Which is so pointless that I just use the DAW after I export the basic midi... and only use the notation full when I'm not producing, or when I expect to finish a project quickly and don't need edits afterwards. I still use MS4 as production software when I'm taking little samples of musesounds to bring the wav/mp3 export into my DAW. I throw out that perhaps 60% the time, the audio export's does not match how I would notate for real musicians and expect the same result.

All that to say I do use MS4, Musesounds, those techniques you talk about, and MS3. And a DAW and proprietary VSTi's. MS4 currently cannot reliably do both notation and production at the same time for my purposes.

So no. I am not ignoring the other things; I am using them and they are not enough for both, or not worth using when I can export to something else that does it better because it's meant for that. We all value production, but we value different aspects of it and we value it in different contexts.

"[You say that] want a better process overall but are less concerned with individual works in terms of the production." I say that I want everything that you want but want even more. You are concerned with less/different features than I am and value the fluid notation process/different features more. Perhaps that's a better way to say it? Words next to the slash depend on context. Some of these things I paid for outside of musescore. Which is why my bias against the priority of some elements of musesounds makes you think I don't care. No. I want it all. I'm not comfortable enough with C++ yet to really do what I want, but we're getting there slowly but surely hahaha. I have some ideas with JS which I believe would work with MS3, but not 4 yet.
"So you have to decide between performances that have these tweaks but lack almost every other thing that makes the playback sound good, or have those other things but lack those tweaks for the things that aren't handled automatically" If they're not handled how I want them I will fix them anyways. What's the point. My stance is that Musesounds would be much more convenient to use if it read midi data instead of only notation. Musesounds doesn't read the notation how I'd want it to.

I've been saying I use all these tricks:
https://www.youtube.com/watch?v=0vOY-JSkrok
https://www.youtube.com/watch?v=HYEGwQ-JQrA

And you completely ignore these, Mr Sabatella. You may not have time to skim through these videos and understand the concepts of how getting the sound you want out of musesounds in MS4 can differ with the desired notation, but if you actually watch these, your rebuttal will be from a different angle than you've been taking because you will understand more.

"So I still would encourage people who purport to care about the "production" of their music to at least consider the possibility that they can actually get results that most people will find superior by using MU4 even though it lacks the piano roll editor, and that they might end up agreeing."

I can't agree if I've been considering/using these methods far before you suggested it. Even nimbears was using those invisible markings... I didn't download the score to see the extent, but I believe it. (I believe you also accuse him of not using these [which I believe he did], because you think his score would sound better with the improved playback samples/engine over his piano roll adjustments with soundfonts. I argue that it would probably blow me away even with both [which is not yet possible in MS4]. But I appreciate the idea/sound of them in different ways otherwise.)

Reply

• Jul 22, 2024 - 05:42

Yes, Speedmeteor101, I did use expression markings extensively in that violin/piano duet that I picked as example for this discussion. As I explained, though, I hid them all to avoid the clutter for the benefit of human eyes, as only MuseScore needed to "see” them. Tomorrow, I'll have some spare time, so I will unhide these markings and link the score again, so you all can see. You will laugh at what painstaking effort I spent just to ensure each bar or segment was played in exactly THAT way, the way I would have played it. No way MuseSound could have guessed how I wanted my music to sound like.

Reply

• Jul 22, 2024 - 05:47

It wasn't passive aggressive - it was deliberately chosen wording to make a point, to help you see why the original statement to which I was responding came off as dismissive. Glad you agree it comes off that way now that the tables are turned. In case that point isn't totally clear, here was the original statement:

If Piano Roll Editor was only used by "a very small minority of people", then I posit that these are folks who take the PRODUCTION of their music very seriously.

If I had said,"If MuseScore 4 is used by a large number of people, then I posit that these are folks who DON'T beat their wives", then it would be pretty clear what I was insinuating, would it not? The statement to which you are now objecting was just more of the same.

Anyhow, that is again all beside the point. I don't think I've heard your music in particular, but I have heard dozens of other scores posted by people who are trying to argue that their hand-tweaked MU3 versions are better than what they'd get from MU4. To date, other than the very special case of guitar "let ring" playback support which I will readily agree is an unfortunate omission from MuseScore, not a single one of the examples people have posted and I have checked out and compared to MU4 comes even remotely close to sounding as good as a corresponding MU4 version - not to me or to anyone else that I have played them for in blind tests. And that is my point Not saying there is zero value in the piano roll editor, or that there exist no cases where it can produce music that sounds better than what is possible in MU4. Hust that I think in general, many people are not putting it in context and are shooting themselves in the foot as a result of not looking at the big picture of all the things that go into great music production.

Reply

• Jul 22, 2024 - 07:08

I'll grant that I probably gave that statement a 'pardon' due to perceiving a dismissal for the need of a piano roll on your part- when in reality/ objectivity I do agree with your understanding that statement and to a large degree that whole sentiment that the PR in MS3 alone would be hard pressed making music that allows the listener to feel the expression being added. With worse sounds in using soundfonts or the inability to program much about the midi (which seems impossible without JACK even with VST) the expressive touches would likely need to be more exaggerated to be appreciated by an active listener audience.

Perhaps I could upload an example of what was done with MS3. Maybe I'll screen record a bit. Or maybe I'll make an article comparing 3 methods of using MS3midi+DAW, MS4 focused on audio +DAW, and MS4 proper notation+daw and highlight the strengths and weaknesses I find in the different workflows.

To be honest I think we actually agree on everything but 2 things, but our opposing angles in addressing exaggerated the disagreement. The context of the responses had me previously interpret that you think the piano roll couldn't be 'needed' or convenient by/ for anyone, when you actually seem to communicate that you doubt users are actually utilizing MS4 (lacking a PR) to its fullest potential before flaming it. (Which I obviously tend to overlook as I want the PR hahaha and am also frustrated). I do find that process heavily inconvenient personally, "but don't knock it till ya try it" is a phrase for a reason.

Perhaps we may just agree to disagree about the priority and some conveniences/use-cases but overall I do see your perspective.

Lastly, even though I've butted heads with you quite a bit in this thread, I would like to say that I sincerely respect you and love your work/ efforts for the MS community... And miss the days when I would listen through 'Reunion' as a kid opening up MS... Idk if it was even 1.0 yet maybe 0.9.something. I'm no longer a pianist really but, I started learning it at one point as I thought it was a beautiful tune. Thank you for a passionate but sincere discussion, Marc Sabatella 🙏 It helps to distract from the pain of losing someone close, yeah.

I've been advertising Musescore since I started to everyone I know, so naturally I'm more elated when I can show others how well it works for me and little tips and tricks to get the most out of it (even for things it's not meant for). Patience is a virtue, however, and now I feel like our perspectives are understood. I actually love MS4 when making sheet music and still am excited for the future... Even if the exciting news is a ways off for me.

Happy Scoring, all :)

Reply

• Jul 22, 2024 - 15:10

I think that's a good summary. Yes, I think everyone agrees that the types of adjustments that the piano roll editor made possible can be valuable as one part of a well-balanced approach to music production. I don't actually particularly care for that specific type of user interface to the controls - a piano roll is way too "techy" and divorced from the notation for the average person or even for me - but indeed, specifics of the user interface aside, some sort of access to those sorts of controls are great to have and are definitely being planned. So there is no need to convince anyone that these types of controls are worth having.

Meanwhile, though, I would also say that there are many other aspects of a well-balanced approach to music production, and many of those are only possible in MU4. So either version makes possible things not possible in the other, and either version limits you in ways the other does not. Any disagreement is over one's subjective impression over how this tradeoff plays out in practice.

My main concern here remains as it always has been with the way some people are trying to use the absence of this one particular feature as evidence that MU4 is inherently inferior and that developers don't care about users. No, it's that after careful consideration of years of accumulated user feedback, it became obvious that the vats majority of users would prefer the tradeoffs made in MU4. Especially considering the fact that "missing" features are constantly being added back, and usually in greatly improved form made possible in part by the very same changes in underlying architecture that required the original version of the feqture to termproarily be sidelined.

And thanks for the comments on Reunion - actually it was 1.2 when it was introduced. So crazy that a piece I knocked off in an evening as a demo to show off as many MuseScore features as possible has such staying power!

Reply

• Jul 19, 2024 - 23:31

" Saying that" .... "is just flat out wrong, and disrespectful."
@Marc Sabatella
I'm very sorry if the discussion/complaints have come across as disrespectful. I'm sure nobody who's said negative sounding things so far intends to do so or belittle the great work that has been put into MU4. I'm sure it's just the frustration of what, to some users, has felt like neglect after a year and a half of minimal communication on the matter. I understand that your perspective is also defending the arduous work that has been put into MU4 so far and respect that. I may have been a bit abrasive in trying to get the answers to the questions that I personally desired and others' words here may be coming across as such. I apologize for that. At the same time, even though a lot of, perhaps unnecessary, words were exchanged, I got to better understand your sentiment as well as the frustration behind the repetitive requests/ reason that nothing can be done atm. Thank you for that. And thank you for your work on Musescore over the years as well as your instructional videos. Quite frankly, I've been using musescore since my childhood and have always been very passionate about it. Perhaps that isn't showing in the best light at the moment, but I really do appreciate all of the work you and other contributors have put into the software.

I do remember your leading response above being, "It’s not true that the architecture doesn’t support tweaking. It’s just that a full-featured facility to allow it - and allow it in a far better way than ever before - takes time to design and build."

Nimbears' response does align well with my sentiments and likely many others'. While I was trying to see why certain alternatives or solutions wouldn't work I've neglected a key part of your statement which Nimbears did not:
"Therefore, despite many enhancements in MU4, I'm forced to stay with MU3 until PRE is implemented in a future edition of MuseScore. However, I may simply be unaware of new features in MU4 that achieve the same goal as PRE. If so, is it possible to run a tutorial (similar to your MuseScore 4 Jump Start) or even just a Q&A session with focus on how to tweak a score to "humanize" playback without PRE?"

This is a fantastic idea and would actually be amazing. I didn't truly consider I may have been overlooking something, and even if it's not the case, we appreciate you taking time to respond to our questions and concerns.

Reply

• Jul 20, 2024 - 01:51

No worries, all good. And to be clear: I did almost none of the development for MU4 - few bug fixes here or there, porting over some small of code I originally wrote for MU3, etc. So I’m not “defending” anything in that sense. Just providing insight for the benefit of those who seem quick to judge the motivations or intentions of those who did perform that incredibly amazing work.

As I said before, there is not currently anything in MU4 to achieve the same specific result - modification of ontime/offtime/velocity values for individual notes. Well, velocity adjustments are possible but only for sound fonts, and if you want to be stuck with sound fonts,you mighty as well stick to MU3 indeed. What MU4 offers is Muse Sounds, which is a much superior playback technology that already completely automates most of the sort of adjustments you might otherwise have hours on, plus goers far beyond that in recording all sorts of individual samples for different articulation, different note transitions, etc - all completely automatically. A piano roll gives you none of that. A VST within a DAW could, but only after hours and hours of manually tweaking things - and much of that work would need to be redone after each edit.

That’s why I say, even the default out of the box playback is generally better than what most people manage with the piano roll editor plus a DAW. no tweaking required. it would be the shortest tutorial ever: step one, install MuseScore 4, step 2 install Muse Sounds, done.

Now, for even better results, you can also do things like add extra dynamics, and tempo change elements (new for MU4), so that could indeed be the subject of an upcoming Café.

Reply

• Jul 20, 2024 - 02:06

Thanks, Speedmeteor101. Glad that you support the idea. I see us going in circles, with many of us bemoaning the demise of PRE, which in turn causes Marc to waste time and effort to defend the decision. It makes more sense to refocus on positive action and try to move on!

Marc, your latest comments above made me sense that I MUST HAVE overlooked new capabilities in MU4 that simply negates the need for PRE (and in fact does a better job.) More often than I care to admit, I get fixated on "what has worked well for me" and fail to learn "new and better tricks." (The too-comfortable-adding-to-learn-to-multiply syndrome?) I sincerely think a brief "Jump Start session" can be the solution that helps Marc and us, and that is what I would like to propose.

Reply

• Jul 20, 2024 - 04:06

I’ve already listed all the things Muse Sounds does that you could never in a million years hope to emulate with anything as crude as a piano roll editor. Things that make a far bigger difference to the overall impression of realism in the performance of your score than any manual tweak to ontime/offtime could. So that even without those tweaks, the overall effect is just all around better. That is, you don’t need any “new tricks” aside from actually adding the appropriate markings to your score. The score you posted is really pretty bare in terms of including the sort of markings that would allow either MU4 or a human player to bring your piece alive. You have very few dynamic markings, no tempo changes notated, almost no articulations, or expression markings, etc. Adding those would be the necessary first step in getting an expressive performance, whether asking Musecore to play it or a human musician.

Reply

• Jul 20, 2024 - 05:45

RE: "You have almost very few dynamic markings, no tempo changes notated, almost no articulations, or expression markings, etc."

You see hardly any markings because I hid them as much as possible. Since MuseScore is my target audience and performer, only MuseScore needs to see the markings for tempo, dynamics, slurs, etc. When you listen to the playback, you know the markings are there, as you can hear the ritardando/accelerando, crescendo/decrescendo, slurs, pauses, etc. Often, I even redefine dynamics symbols (p, mp, mf, f, etc.) to make MuseScore play at the loudness levels that I want. E.g., default velocity of "pp" = 33, and "p" = 49. But if I want a passage played at velocity 40, I'd change the number in the Inspector Tab to 40, and also edit the marking to show "pp40." Now, since such a symbol means nothing to a real musician, I make it "Invisible" to spare the confusion. (Not that I actually expect anyone else to want to play the piece.)

Suffice it to say, whatever tool you've got, you find ways to work with it. There's a method to any madness. Please don't invalidate my feeling that Piano Roll Editor is actually a pretty good and flexible tool. It serves a specific purpose, and does it very well. In return, I will take your advice to explore and use MU4 to the fullest, and WILL find a way to make it do what I need.

Reply

• Jul 20, 2024 - 14:01

It's a fine tool for doing the very limited sorts of things it is capable of. I'm just saying, you can tweak those values all day long and still never come anywhere near the level of realism that is actually possible using MU4, in most cases.

For the cases where that isn't true - where you can actually get better results with Mu3 than with Mu4 (I've heard almost zero examples of this), then indeed, it will be great to someday have a fully functional piano roll editor in a future release.

I'm not sure what else there is to say about this.

Reply

• Jul 20, 2024 - 13:55

"Adding those would be the necessary first step in getting an expressive performance, whether asking Musecore to play it or a human musician."
I'd disagree it's a necessary step, certainly looking at that score it's in a sufficiently conventional style that it really doesn't need more markings for a human to provide an "expressive performance", any more than the music of Bach does (or any music from a period in history where very few performance directions were typical).
On that basis I don't see why computers shouldn't ultimately be able to do the same, even if we're not there yet.
Most of the tweaks I have to put into MuseScore are because it doesn't do the things that humans do naturally and it sounds unmusical or downright unpleasant to listen to without them.

Reply

• Jul 20, 2024 - 16:44

Thumbs up, Dylan! One of my piano teachers told me all expressive markings (which excl. stuff like staccatos, ties, time signatures ... the obvious "structural" things) are GUIDELINES. She even encouraged me to go against the markings wherever instinct tells me to ... at least to try and see how I like my rendition. Her quote that I'll never forget: "Average musicians PLAY music, inspired musicians INTERPRET it." Well, since computers can't interpret yet, I put them in for MuseScore's benefit (yet hid them for the benefit of humans.)

A little segue if you don't mind: You're so right to say, "I don't see why computers shouldn't ultimately be able to do the same." We've probably all seen A.I. attempts to compose music in the style of Brahms, Mozart, Chopin, and I'm trying to learn a software that promises to have me sing like Elvis or Pavarotti!! (More precisely, I would sing and upload a song in my crappy voice and the s/w "sings" it back to me in Elvis' voice!)

Reply

• Jul 20, 2024 - 17:15

Precisely one of the main reasons I loved the PR. Less is more.

When you're working with good musicians, they might understand the influences and styles of your work from one glance or playthrough and if not, a quick explanation and example often suffice. More chord changes, nit picky dynamics, and markings that control too much often hinder them, their enjoyment of playing your music, and the beauty they are 'allowed to' inject with their own musicianship. When working with less experienced musicians sometimes more markings can help, but it'll often be different than what musesounds needs as everyone has different weaknesses and needs different help. Which is where having the piano roll and providing example/ practice tracks becomes a lot easier.

Those are great points you make that are being overlooked. Realism helps a lot, but once everyone has the same level of realism, the difference between that and what is real becomes more obvious anyways. That's why the ingenuity behind older system-limited video-game composers music is so beautiful and innovative with only 16 and 32 bit sounds. And that's where your goals and vision for your own music shine through no matter what tools you're using no matter how advanced or 'crude.'

The thing about computers is currently that each computer system or AI model represents one super human's tastes or abilities guided by prompt, or in this case, notation. Perhaps far in the future, humans will work more closely with AI devs much like plugin devs began to work closely with orchestras and soloists to make the plugins we have today like ms4... but at the end of the day, they're all still tools to help bring our visions to life. Even with A.I. voice models, you still the confidence, vibrato, and inflections of your desired singer and have to sell that performance with your own voice... and it can cover the tone that that singer had, as well as most of their accent. The tools are getting much better (SUNO and UDIO AI music generators have been making seriously good music that I could unironically listen to...) but they have not yet crossed the bridge of starting with something that has already been born and building on that. More so, providing something new based on the ideas in a prompt. Really cool to listen to, but not very creative. Perhaps still a few years away haha.

You all are providing great discourse and subject matter to think about and consider.

@Nimbears
If you weren't already using the techniques I linked in the vids/post below, I hope you find them helpful

Reply

• Jul 21, 2024 - 13:47

True, a human could of course add their own expression. And sure, a sufficiently clever AI could do the same.
But as a composer you'd lose the sort of influence/control over the results that folks seem to be wanting here. So, for practical purposes as per this discussion, those sorts of markings are necessary to achieve the desired result in this context.

Reply

• Jul 20, 2024 - 04:16

I believe the orchestrational tricks that you and he allude to would be found in these videos:
https://www.youtube.com/watch?v=0vOY-JSkrok
https://www.youtube.com/watch?v=HYEGwQ-JQrA

You may end up finding that you make one score/file for notation and another for desired playback. MuseSounds has gradually improved with every update so this isn't necessary for fixes anymore as much as it is just for personal expression.

Personally, I try to completely finish the notation and then export Midi/audio to a DAW where I can actually mockup/produce. For pieces where I have more straightforward parts /tempo and don't want to bother with a VST, I actually just export the MuseSounds instruments as audio files and import those into the DAW. But if you don't have one, then just make separate scores... at the most complicated level, you can do some in MS 4 export the parts where you need more control to MS 3 evolution (it now open MS4 saves but you can also use XML/Midi), and then combine/complete the audio in audacity or something.

This is what I currently do for now; if it's not what you're already doing this may actually be a bit nicer of a product than doing everything in MS4 or MS3. MS4 has made engraving so much easier and we're here because of MS3's control. Different vehicles for different errands, amiright?

Reply

• Jul 20, 2024 - 18:41

I still get a little twitch when I see a phrase like "humanize" in reference to playback. I go back far enough to when there were discussions about adding timing variations to various parts because real players can't hit notes all at the same time. This made for more realistic recordings. Twitch. I would rather spend time making recordings more "musical".

It seems to me that for those people to whom PRE is so important, they should stay with MU3. No reason to move to MU4. In fact, it seems to me that PRE is a must because playback is so poor in MU3. However, I would never say that PRE isn't needed in MU4. Things like Let Ring are important. It would be nice to, at the very least, be able to control velocity. Muse sounds certainly do not always follow dynamic marks or hairpins. Very high string and ww pitches are not always in tune.

Not to mention that if you don't set up the computer audio device properly, Mu4 might not work correctly at all.

I use notation software to compose. And 95% for playback only. And as a hobby. I have spent some 60 years playing all different kinds of groups. I've lately had to give most of it up. But I know how groups sound from the inside out. So I write for playback. You would think that I would want all the control I could get. After all, it's my music. My vision. My need to put something out there. But for me it isn't about control. It isn't about forcing my music into some Humanization (twitch) box. It isn't about what I want. If I write something and I play it myself, that's one thing. But as soon as I involve computer sounds, it just seems to me that the game has changed. They not only have a voice in what happens, they are the voice. Computer sounds don't blend the same way real players do. There really isn't much way around that. When I write something that doesn't really sound like I want it to, I have some choices. I could spend time trying to force what I wrote into something different. I might succeed. Or I might not. Or I could just write something better. To me, that's part of composing. I want to write something musical. Not Human (twitch). You may be tempted to say that because I write as a hobby, that I'm not picky about the quality of what I write. You would be wrong.

Not to mention that there are plenty of us that do not have DAW's and the expensive sounds to go with them. This may be all the more reason to have PRE in MU4. No argument from me.

Only this: When I write a horn melody and simple string backup in MU3, I get a certain playback..... quality. I could spend hours tweaking the score. I could use all kinds of hidden expressive marks. I could tweak note lengths and phrasing in PRE. None of which will fix the fact that basic sounds are junk. Or I could open the OE score in MU4 and it at least sounds Musical. Perfect? Maybe not.

Reply

• Jul 20, 2024 - 19:29

@bobjp
Great points. If you don't call it humanization, do you call it 'feeling?' I think I've been referring to it a rubato-ish 'expression.' Like I said, some people just don't care for that sort of control. I don't think that has any bearing on the quality of music you make. I just want it in what I make. Once I started messing around in FL Studio, Reaper, ended up having to use logic and pro tools for projects and seeing how other film-scorer's worked... I began to love programmed music that still felt like it could breathe. Obv, that's why I ended up buying a DAW and instruments. But I think via notation first, so Musescore is always the first step in that unless I'm recording a solo into a synth, or making a little beat or super experimental soundscape.

Sometimes making it feel good is just about having that extra delay groove with notes being swung half between triplets and straight sitting nice and behind the beat. Sometimes it's the 'soloist' 'rushing' the line past accompaniment to accentuate the phrasing and making it breathe as we learn to do (more or less) when playing solo repertoire. Personally, I love to capture that... and doing that in musescore means I can copy/paste/edit that in musescore. I don't care if it's real or fake, I like when it feels good too.

I've definitely been a bit biased against musesounds having already invested in other software to work with it, but as you said... that's perhaps all the more reason to have a PRE. I'll definitely concede that being able to have that good basic sound without much effort is game changing in the freeware side, and once the PRE/automation is added will be game changing in general. But I always got that quality elsewhere and had it setup conveniently with easy midi export/import. I just love the way MU4 revolutionized showing and hiding parts, the pallets, and have a hard time choosing so end up clunkily using both.

To me, it's basically like noteperformer currently (which makes sense, given sibelius is also notation software). It just works and sounds nice 'out of the box.' Sure. I do see it as a key part to the journey of getting MuseScore rightly recognized in the publishing world and the snobbier sort of notation world... but of course, I don't prefer it coming at the cost of a key part of my production work flow (when the UI, pallet, and part updates are so good).

While I may not fully understand the history behind why "humanize" make you t- twitch, your response was vocalized wonderfully and I appreciate that and the points you made

Reply

• Jul 21, 2024 - 00:33

If it is just me and my guitar doing some tunes, well then I am the master. I decide how things will go. How fast, loud, expressive. I am the human. But as soon as another person joins in? Well, things start to change. The goal, of course, is for the result to be better. If it is better, that's great. But in any case, I'm no longer in complete control. Now I like playing in a band or an orchestra. But the larger the group, the less control I have. I'm not some kind of control freak. I'm just saying what happens. Alone I humanize my way. The second person that joins in, humanizes their own way. Each member of an orchestra adds their own twist. Yes, they work together, but in their own way. That's what makes a live performance so great to hear.

So here I come along and try to match that on a computer. I try to humanize 30 something instruments. But it's all me, not 30 different people.

I know that part of my problem is that I have no idea how to use PRE. Which is why I've stayed out of this thread. I also have now idea how to use a DAW. Again, what am I doing here? It's just that I had pretty much given up composing because I don't play piano and only had paper to work on. Then in 2007 my music teacher wife bought Sibelius. I had been a music ed major, so I knew notation. I loaded it up and was completely blown away. I remember the string bass sound. You could hear the sound of the bow moving on the string. Holy Cow.

But my point is that I can't Humanize my computer music. I can try to make it more musical. Semantics? I think there is a difference.

Reply

• Jul 21, 2024 - 01:46

I think that's really not much different in some ways. The real question is when someone says, "humanize," humanize how? What are you implying in saying that word? In this case it's those musical nuances that come with personal interpretation. But I do actually kind of agree with what you're saying about musicality vs humanization. "Humanization" probably isn't 'humanization' until you can fool someone that a human is playing it live... on some level. Even virtual instruments played by a musician on a keyboard (which may have been the implication earlier). That often only happens in major movie productions where half of the layered sounds you hear are indeed real humans hahaha.

When I play gigs or get to offer compositions/arrangements for some performances where I welcome the interpretations, I usually know who is playing them. Perhaps my form of 'controlling' the 'playback' (for a lack of better terms) is like choosing what I share with whom. Who can I play this different bass line groove with, who do I want to play this composition... am I curious about their interpretation and style and will I appreciate it? If not, they don't need to see/hear it. (Of course, unless there is another value in sharing it with them such as their own enjoyment if they really will enjoy.)

After those initial decisions, I love to use music to communicate with others. I don't feel like I need to control things, but I like what I like hahaha. And if I don't like it, I wanna fix it. I don't want to control what I already like. If I don't have access to the people I'd love to hear play it live, I'll produce it myself. In a way where the listening experience is enjoyable for me. The extra control makes the computers like-able. I do like to listen to some pure, raw, midi data, limited Sf2 compositions, but there is a specific character to what they're doing that transcends realism that makes me like it as a recording. They take advantage of the limitations, exploit them, and compose with/for them. Otherwise I'd rather hear it played or produced well.

I love that experience you recount about your wife bringing home Sibelius and it blowing you away. Technology's improvement in realism/accessibility is always amazing to see. Kind of like A.I. (when not abused as it is). The semantics can go both ways, but those points you made are still valid points I actually agree with and resonate with deeply.

Reply

• Jul 21, 2024 - 01:09

Re bobjp, "None of which will fix the fact that basic sounds are junk".

That's where MIDI Out is so useful: the score output can be sent in real-time to a VST3 for vastly improved playback – and not necessarily expensive. I haven't heard a classical guitar in MS2, MS3 or MS4 that comes close to Pianteq's classical guitar.

This is without any PRE tweaking at all: it's out-of-the-box plug and play: MS > LoopMIDI > VST3.

As for humanization, I also twitch. I had a play with a plugin to create less than perfect timing and the results were awful. If this was humanization then it was more like a beginner human making the sound less musical. Good humanization (more musicality?) needs sophisticated AI programs or, better still, real humans.

Reply

• Jul 21, 2024 - 02:11

Another note on humanization:
Let's take starting with a great sound but still sounding a bit like playing on a keyboard as a starting point.
https://www.youtube.com/watch?v=l0d-t8HRIgA

But when someone else is able to take the same virtual instrument and play it in a way where I think it's someone else's real playing, that's definitely humanization (in my book).
I'd close my eyes and listen to this pianist: https://www.youtube.com/watch?v=b_WlGzh3Qik
and think that this guitarist is actually playing it: https://www.youtube.com/watch?v=uEG9manmZWg
Same applies to if this is programmed into a piano roll. For the record if it can be played into the Piano Roll, it can be input into it. Just takes a little study.

Even if we're talking about something as crappy and as unrealistic as soundfonts, if you can make programmed/typed/clicked in notes sound as though you played them on a keyboard (or with the same inflections that the real instrumentalist would've used), that is humanization as I understand it.

Food for thought. And to push the discussion forward.

Reply

• Jul 21, 2024 - 05:18

yonah_ag,
I know you are all about guitar. I have played most of my life. I built from scratch the guitar I play. But, alas, I am but a simple folk singer. I can read tab and notation a little if I have to. I'd be more interested in a strum pattern plugin.
I'm glad you have something that works for you.

Reply

• Jul 21, 2024 - 06:03

Looking forward to more of your thoughts :) I get the different wavelength you're thinking with. I just liek to play devil's advocate.

And although it's directed for yonah,
I highly suggest ample guitars for this and it comes as VST3 among others, meaning it can be loaded in to MS4
https://www.youtube.com/watch?v=CVvVTgwmSkA
As you can see around 8 minutes into the video above, it does have a chord strumming mode where you can use one set of notes to choose chords, another range to choose different strums (up/down for example) to play the currently selected chords, and then abother to play a whole strum pattern you can punch in. I used that feature a few times, and the first couple times it took a while to set up, but in the same contexts later was easy enough to load in and use. It is a little over $100, but very much so worth it if you'll use it often. Great realism (same brand used in the youtube videos I linked before)

https://www.amplesound.net/en/pro-pd.asp?id=8

Reply

• Jul 21, 2024 - 14:58

I only suggested amplesound since bobjp wanted a strumming plugin (assuming he talking about a VSTi).
It definitely needs a little bit of help to sell the last bit of realism (pianoteq felt more natural in the videos), but once you're used to the system, you can buy a bunch of their instruments on a bundle and know how you like to work with them and your process for making em sound good. (Whether electric, acoustic, medievalish folk, ethnic woodwinds and or anything else they offer).

Aside from that plus (which is a minus for your savings in a way smh) the other plus was just the strumming system. Pianoteq seems great for that real guitar sound.

Reply

• Jul 21, 2024 - 16:24

I think that the Amplesound Alhambra would be a nice addition to Pianoteq as the underlying guitar sound is sufficiently different: almost like a typical cedar top vs. a typical spruce top. This would give me some more choice in output sound. I still need to be able to upload the custom VST playback to musescore.com so that it can be heard on web playback.

Reply

• Jul 21, 2024 - 16:57

speedmeteor101,
Thanks for the suggestions. I'm just not interested in downloading yet another sound. Why? Don't I care about the quality of my work? Don't I want the best possible output? I have a number of vst's. The sounds are very even. They come with various controls that help them sound real. They require more effort than I am willing to invest just to make them still sound boring.
You want to make your scores sound like a human is playing them. I want to make my scores sound like music. Until the ability to record sound came along, the only way to hear Beethoven was to go to a live concert. You could go to a concert in one town, and then go to one in another town later. You might hear the same piece, but it would be performed differently. Both are human interpretations of the same music. Is one better? Which one is correct? Does it matter?

As near as I can figure out, this might be my mind set. And thinking this much gives me a headache.

I walk out on stage and sit on a stool in front of a microphone. I can't really see the audience because of the lights on me. There was scattered applause when I walked out, so I'm guessing there may be a few people out there. So now it's just me and my guitar. No band or backup singers. Just me and my guitar. So now my goal is to make music. Sure, I'm human. I can't help but humanize what I play. If I use the guitar I built, I play a way that I know that guitar sounds best. If I walk out there with a different guitar, I can't quite do the same things. An electric guitar changes things again. But the goal is still the same. I have to make the best music I can with what I have.

Fortunately, I survive. But now I'm sitting in front of my computer. I have a score open using Muse sounds. It's just me and my computer. No lights. No audience (thank goodness). Just me and my computer. But the goal is the same. I want to make music. I have a section where flute and oboe are playing a little duet. Because recorded sounds don't always play well together there is an odd overtone when the two instruments play a certain combination of notes. It is the fault of the oboe. When I change the note to a different one, the overtone goes away. But I really wanted the first note. What do I do? I'm the composer. I'm in control. What I want should be what ends up happening. Right? I could buy and use all manner of expensive stuff and make what I want to happen be the result. Or I could do what I did when I walked out on stage with a borrowed guitar. I make music. I use what I have, and make music. That is my goal.

But there's more. On some level, I'm not sure I'm ever in control. I think that when most people sit down to write some music, they have something in mind. They have a little melody or chord progression that they want to expand upon. Normally, I start with a blank score in front of me. And normally my mind is just as blank. And then it happens. I start putting notes on the page. I know it sounds odd, but the music seems to be writing itself. I'm not in control. I'm being used. After a while, the music and I work together to.....make music.

And now I have to go lay down.

Reply

• Jul 21, 2024 - 20:55

@bobjp
Beautiful recollection of borrowing an instrument on stage. I now see it's a different mindset altogether. You go with the flow with a great deal of patience. On the other hand, I will if I have to, but if I'm borrowing someone's rosin and it's not grippy enough for how I like my bow, I'll find some alcohol, even in sanitizer, to get that. If someone's amp sounds bad, I will turn down and play acoustically as much as possible just to improve the tone to how I want it. Assuming there's no eq on it haha. When sometimes, the simplest way to make music is to lean into that equipment that you're not used to and embrace it for what it is. It's quite profound. I've got some pages to take from your book.

Still, while some of my desired control does inevitably make the machines seem more 'human' in interpretation, I do it not for the humanity, but that feel. The breath. The life. The swing. The phrasing that says so much about the player in a single line. Even if the player is a printer motor running at different voltages sounding stiff and inhuman, there's a charm and a natural subtle swing to hearing all the innerworkings, imperfections, and limitations of a printer that's been programmed to play music. And it feels good. Not necessarily human, but 'good.' It's actually music that abuses the limitations of phase-overlapping midi/ synth notes between the oboe and flute, or any two types of waves that shows that creativity that shows... "This is definitely a creation to listen to" vs "I would love to hear this band/group/orchestra play this or this person arrange this."

It's for my own enjoyment of the music. The commentary about humanization was merely in response to your twitching hahaha

On another note, I assumed you meant VST plugin, not a notation plugin. My bad! It was just a good tool for emulating strumming and had a lot of other bonuses. If you're not looking for that, I misunderstood before, but understand now hahaha.

Reply

• Jul 21, 2024 - 23:52

speed meteor101

I read a story once. I have no idea how true it is.
Clapton and Van Halen when playing some kind of show. Everyone had done their sound checks. Clapton was last. He set down his guitar and as he left the stage, Van Halen walked up and asked him if he could play his guitar. Clapton said yes. So Van Halen picks up the guitar that Clapton had just put down and he started to play some of his own style music. Everyone was surprised. It didn't sound so much like Van Halen as it did Clapton.

As you pointed out, the equipment matters. This story backs it up. And that is my fear. I'm afraid I will add so much processing to my score that my music will get swallowed up and lost. Oh sure, the notes may still be there. The sound might be better. The phrasing will be smother. Everything might be technically correct. But will it be music? My music? Is signal processing, music?

So I write a few measures of something and play it back in Muse sounds. I might write some more, I might modify what I wrote earlier. Or not. I playback what I wrote. This is my music. This is what I hope for. If it wasn't what I hopped for, I wouldn't have written it. If I play it back with another font, it is no longer what I wrote. If I run it through various processors, it is no longer what I wrote. For me, simpler is better, and cheaper. Me and my guitar making music.

So I work with MU4. I have no use for MU3. It isn't even on my computer. I work with muse sounds. Even though they are full of warts and scars and broken promises. Can I force them to do what I want? For me music isn't about forcing anything. It's about bringing something to life. Letting it happen.

Reply

• Jul 22, 2024 - 00:17

Re, "I read a story once."

You are very fortunate that MS4's sounds just happen to be what you want – "your music". The sound is produced from digital samples (which you didn't make) using digital processing (with software which you didn't write) on a computer then converted from digital to analogue through signal processing software for your speakers (which you probably didn't engineer) to play. Is this digitally processed stream of numbers over which you had so little control still your music? Clearly yes. Fortunate indeed!

For me, picking up my curvy shaped box of wood with strings stretched along a neck and over a sound hole is where my music starts. Is it any good? In truth, not really but it's my music. So, what I want from music software is an easy way of writing down this music and playing it back with hi fidelity to how I imagine myself playing it if I were really any good. If that means a bit of tweaking then I'm prepared to spend some time on it. It doesn't need to be forced, just carefully birthed into life.

I'm glad to hear that MS4 suits you so well but for me it's not up to the job yet.

Reply

• Jul 22, 2024 - 02:28

Good points from both bobjp and yonah!
Thanks for providing the other perspective, yonah and thanks for your thoughts bobjp.

I'm not so much worried about equipment. I just want to make it easier for music in my head to come back to my ears... Or make it easier to hear sounds that I find pleasant from other people or tools.

I love your organic writing process, bobjp. You come in with no expectations and find the beauty in the sounds that you find. And make the ones that sound the best to your ears with what you have. That's beautiful. Sometimes that's also my process. Sometimes I'm trying to bring an idea to life and am trying to recreate that in real life (or computers). No matter whether the product is inspired by my head and I mess with 1 million parameters, or whether I let the free tools I have and their limitations inspire me, it is still indeed my music. I made it. I edited it. And that organized blend or cluster of sounds, sounds good to me and soothes/excites my soul. Others as well, if I'm fortunate. I don't think your process or music needs to include what's in mine. That's what makes all people and their ideas unique and beautiful.

Yonah's points are definitely strong. Nothing we make/ have is truly only our own creation. Something else made the thing you used to make the thing, and the ideas that inspired you and transformed through you into something new, unique, yet similar enough to fit into that genre.

The way I understand it, once you feel too much resistance to the process of making music to where it may not feel as if flowing freely, you no longer consider yourself to be making your music. If you have to fiddle with too many buttons and change too many things along the way... If you feel like you're killing that first brain baby of an idea that made it onto the score before you considered editing it... You don't seem to want that.

I welcome the challenge and the resistance... I want to push my creativity and problem solving beyond making music, and explore/modify the processes, the softwares, and my own musical skills as a human. Every challenge I overcome makes the previous an easy part of making my own music the next time I come across it. I'm not where I want to be yet, but it's fun to me. Not everyone's cup of tea, but it's mine. And yours is yours. If you don't want to use other tools, of course, don't use them! We'll continue to pursue and perfect our own in the meantime :)

Reply

• Jul 22, 2024 - 16:22

Indeed. I did not make the tools I use to make music on my computer. I have done the world a favor in that way. But here is an interesting thought:
I bought planks of rosewood and mahogany. I cut the rosewood into thin sheets and bent them by hand over a hot iron into the curvy shape I needed. I cut and shaped the mahogany into a neck that fit my hand. I ordered high quality spruce for the top. I decided where the sound hole should be. I designed the placement of the internal bracing. I made the rosette and binding of wood because there would be no plastic on my guitar. Even the tuning knobs. As well as my own shape wooden pickguard. And a corian nut and saddle. As a result, I have a perfectly average sounding guitar. But in so many ways, it is truly mine.

It isn't that I'm happy with MU4. It would be much easier if I could just get some notes on paper and let the players fight it out. But I doubt that anyone will ever play my stuff. But that isn't why I write.

And one more thought. Even if I spend time running my stuff through all kinds of things. Tweak every note. Set up everything just the way I want it. Even so, as soon as I play it on a different system, it sounds different. I didn't set the trumpet so loud right there. Why can't I hear the horn as well? What happened to the second violin part? Things I would notice, but maybe others wouldn't.

Reply

• Jul 22, 2024 - 17:03

Re bobjp: "But here is an interesting thought"

I am honestly impressed. I have no idea how to make a guitar but it sounds like a labour of love. I would say that to create a "perfectly average sounding guitar" is a huge achievement and, as you say, it truly is yours. How long did it take you?

Reply

• Jul 22, 2024 - 21:53

Including the inlay, about 80 hours. I actually took a stab at trying to sell them. But it was just a crash and burn. I did sell one. That person promptly moved away and I never heard from them again. I made two ukuleles that I gave to some people I knew as gifts. Both of them have since passed away. I have arthritis slowly taking away the use of my fingers. I love making music. That's why I'm working more on my computer.
MU4 is far from what I wish it was. But it is what I have.

Reply

• Jul 23, 2024 - 00:16

@bobjp
Everyone has a level with which they can feel something is their own. I'm glad you didn't feel a need to find/ cut down natural tropical hardwood hahaha. What you did is already an amazing achievement, but I'd say even people who assemble a guitar building kit with limited knowledge are possibly making something that's their own to them. No need to fiddle with things you already feel are out of your control and not worth pursuing. Whether that's crafting your own guitar from scratch, tweaking fine details of a score, or building software.

Being your own luthier is absolutely amazing and a testament to skill. Even if you call it average sounding, to make your own tool that can successfully express your own music as you want is marvelous. Even if that's only to a degree, and I'm sure you far outdo that. And As a musician and a craftsman. From your mindsets and philosophies, I'm sure your music has a lot of soul to it. I'd love to hear it. Sad to hear arthritis slowly robbing that form you, but it's great that you're still creating with what you have. I hope that never changes for you or for any of us.

Reply

• Jul 21, 2024 - 15:05

"A strum pattern plugin: now that's an interesting idea!"

Many guitar VST libraries now include strum patterns.

Just last week I tried my first VST guitar strumming pattern project in Logic with Native Instruments Acoustic Sunburst 2 via Kontakt.

This short accompaniment-only composition uses just a single strum pattern ... and indeed it's an appreciative nod to Joni Mitchell:

    https://audio.com/audius/audio/mitchell-canyon

Here's another version of the recording where I solo the strum track at various points:

    https://audio.com/audius/audio/mitchell-canyon-soloed-strum-ab

The project required only minimal inout, as you can see below (in the image of Logic's Piano Roll Editor.)

• Keyswitches are highlighted in light blue. ( I used a single keyswitch to select the library's "Sweet Memories" strum, and I stayed with that pattern throughout. I only needed a single keyswitch to accomplish this; the redundancy was so pasted sections would always function as expected.)

• The green bars beneath show velocities.

• Nothing else was involved in generating the strums in this project.
     Mitchell Canyon MIDI Editor window.png

When I get a chance I'll see if I can recreate the same playback in MS4.

If I ever write a melody for this I'd like to reduce the volume of the strum's string slap. Fortunately there's probably a setting for that in Kontakt!

scorster

Reply

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