Help with compound meter
I have a mezzo in 12/8 over a piano 4/4. I created a practice file to familiarize myself with the Musescore process. Now, the first bar she sings she comes in the 3rd beat with 3 eighths followed by a quadruplet - - - ....
For the quad, I entered a dotted quarter and notes>tuplets>quad and it worked fine.
However...on the score when I try this I cannot get a dotted quarter to divide, instead I get a quarter tied to an eighth.
This resulted in an oddity: a quadruplet of 16ths followed by a blank space where an eighth rest should be.
Yet, in the next bar, which begins with a quad, it worked fine because I was provided with a dotted quarter to tuplet. I cannot get the 4th quarter in the previous measure to dot.
Anyone know what happened?
Score provided -
| Attachment | Size |
|---|---|
| Safe in their Alabaster Chambers.mscz | 23.16 KB |
Comments
BTW - the a message precedes the file declaring it corrupt. I click ignore and it shows up fine.
???
In reply to BTW - the a message precedes… by penne vodka
See: https://musescore.org/en/node/54721
Edit With the above method. For the others (5/6/34/35/36), insert some measures and rewrite them
In reply to See: https://musescore.org… by Shoichi
Thank you.
This is what I had to do. I rewrote m5. In 12/8 I originally had a dotted half rest to begin. the second half of the measure had three 8ths followed by the 4quads.
(see score)
I could not get a dotted quarter to quad in the last beat, so I 1) wrote in 4 dotted quarters. 2)wrote the quad on the last beat 3)three 8ths for the third beat 4) then I overwrote the first two dotted quarters with a dotted half rest.
It shouldn't be this way. I noticed that Muse is inclined to go with a half rest and quarter rather than a dotted half rest. In other words I might have been safer with a half rest to begin the measure rather than a dotted half as it seemed in following measures. IF I put a dotted half rest in the second half of mea6, THE FOLLOWING measure 7 will be corrupted with a variety of rests out of nowhere.
The error message says mm35-38 are corrupt. I don't know how, I'm still bogged down in mm5-7.
This new file with the Bis-1 extension: is it mine? Is this the the one I continue to work with? Would i be better off starting over with a new file?
In reply to Thank you… by penne vodka
I'm not following you. Above is the * Bis.mscz file that I attached after some corrections with the cited item.
Now there is one bis 1 + which is your attachment which I deleted the corrupt measures (selecting them and pressing Del).
But you could use the file https://musescore.org/en/node/276281#comment-853038 and correct only:
Bar 5 Stave 1 incomplete. Expected: 24/24; Found: 154/192
Bar 6 Stave 1 incomplete. Expected: 24/24; Found: 286/384
Bar 34 Stave 1 incomplete. Expected: 24/24; Found: 2/3
Bar 35 Stave 1 incomplete. Expected: 24/24; Found: 2/3
Bar 36 Stave 1 incomplete. Expected: 24/24; Found: 2/3
In reply to I'm not following you. Above… by Shoichi
The bis_1 was incorrect. There were 128th rests all over. I did what you said and read the link you gave me and rewrote those measures, but the same thing is happening as explained above.
I think I should start over. I'm only 7 bars in. I'll see if it happens again. Anything of interest I will report back.
Thanks again. L'apprezzo molto.
In reply to The bis_1 was incorrect… by penne vodka
Since you are starting over, I'll tell you that whatever you did sounded complicated and far more work than MuseScore requires, even if you are trying to emulate poly-rhythm using triplets.
You can use poly-rhythm by inserting a time signature, then holding ctrl while dragging a time signature to the point you want to start using that time signature. Set up your time signatures, before entering notes for any instrument. If you want to change time signatures again, change the time signatures before any instruments have notes entered in that measure. If you do this, poly-rhythm works fine. There are certain exceptions to this, but if you follow this, you will have no problems.
One thing to consider, all measures after the time change in poly-rhythm must be empty or you will get an error. The other thing to consider is that you cannot copy from or paste to a measure affected by a poly-rhythm time signature. Each note has to be entered by hand. Everything else works normally, so you can put triplets in the 3/4 measures for example.
In reply to Since you are starting over,… by mike320
The info Shoichi kindly sent was valuable....some over my head at this point. I have more arrows in my quiver, but I still don't know what happened. I figured it to be a piece of cake with MuseScore.
The time sigs are set at the opening and NEVER change. The last two files above show what I wanted. In fact, to be clear, I added an example below of what transpired.
I may be better off writing the part in 4/4 and inserting many triplets in her part. That should be easier, no? And it will still carry out my intentions.
But my curiosity won't leave me be. Besides, if it is a problem, you'd want to know.
In reply to The info Shoichi kindly sent… by penne vodka
What you are doing wrong still doesn't make sense to me. It's probably my fault. It's more important to concentrate on how to do it right.
Any time you create a tuplet, you need to look at the measure to see how much time it takes. For example, in your 4/4 measure, a 3 eighth note triplet takes exactly one beat. You would see 3 beats of other notes in that measure leaving you a single beat for the triplet. To make the triplet, you create something the length of a quarter note and press ctrl+3 to make a triplet. In your 12/8 measure you need to do the same thing. If your Quadruplet lasts 1 & 1/2 beats, make something (usually a rest) that lasts the duration of a dotted quarter note and press ctrl+4. If I were already in note entry I would type
5 . 0 ctrl+4
I would then see 4 eighth rests with the first highlighted so I could start entering 8th notes, or any other notes to complete the quadruplet. As I said before, poly-rhythm doesn't affect this. If you could post a picture of your goal, I, or someone since it's close to my bed time, could explain step by step if you needed it.
In reply to What you are doing wrong… by mike320
I've been doing it precisely that way, using toolbar, not keyboard. I've unintentionally come up with the faulty result when I started over, one which you should see. When you try to dot the 4th beat value, a salad of rests, sometimes including 128ths, are strewn into the next measure! The original corrupt file report indicated "Found 154/192" (!) and the three final blank measures (34-36), which I was nowhere close to, were corrupt as well.
Example B shows the process I've used.
The PDF, if I didn't foul up, show what happens (rest salad) when I try to dot the quarter rest.
In reply to I've been doing it precisely… by penne vodka
I now see the problem. You don't see it until you close and reopen the score. I haven't used different times signatures like this in version 2.3.2 so I guess they broke it when they tried to fix other things, which I don't think they did well.
I think your best bet is to do as you wanted to do and use triplets for the voice staff. You can right click the time signature in that staff and change the display to 12/8. In any measure that has any notes, I use triplets to make the rests and notes look correct. For example, you need a dotted half rest in measure five. You will need to make a half rest, select it, press ctrl+3, then change the rests in it to a dotted half. You will want to hide the triplet bars in these measures. You can select as many as you want at once and, in the inspector, change the bracket and number settings to none. You can then create your quadruplet inside of the triplet and leave the bracket to its default. It's easier, in my opinion, to do several at once sine you can select several measures, right click a bracket and in the pop up menu choose Select>All similar items in range selection.
In reply to I now see the problem. You… by mike320
Last night I looked over the manuscript and it became evident that my intentions are better served in 4/4 with triplets. The idea was to have the mezzo float over and independent of the piano. The triplets mixed in with eighths imply a better rubato feel than the imposition of 12/8.
After all this!
However, I am pleased to have discovered something and learned a few things (including the photo) in the process. Thus far, no problems with triplets in 4/4. I had a feeling it was the combo with local time.
Thanks for your time.
In reply to I now see the problem. You… by mike320
As it was said throughout history, suspect first the idiot behind the wheel. The mistake was so simple I am even surprised you guys did not suspect my faulty technique. I've completed 9 scores now without knowing how to properly enter a dotted note. Whilst trying to do a recreation I stumbled upon the proper way.
Folly demonstrated in attached expose'.
For me this was a learning experience, but for all the commotion I am forbidding myself from eating pasta dishes for 4 weeks.*
Thanks for your patience and assistance.
*(no way!)
In reply to As it was said throughout… by penne vodka
Hmm, glad you learned how you can enter dots directly, but actually, what you did should have worked just as well, just unnecessarily difficult. And in any event, it should never be possible to create a corrupt score. So I'd still like to understand more about how that happened.
And I can now verify that attempting to add a dot in the manner you did will cause corruption, both in 2.3.2 and in 2.1 (and, I would assume, going all the way back to 2.0 and earlier, since most of this code has not changed). The issue has nothing to do with tuplets but everything to do with local time signatures. Attempting to add a dot to a quarter note on beat 10 of 12/8 (when used against 4/4) will cause corruption every time.
So it's a bug indeed - one that's been around a very long time, and you're apparently the first to notice! Thanks for helping track it down. I will file an official issue on this one.
EDIT: filed as #276363: Corruption on duration change of note in measure with local time signature. Note it isn't just about the dot command, it can also happen with other duration change commands (eg, pressing "W", or "6" to turn a quarter note to a half). But it doesn't happen on any duration change in a measure with local time signature, not totally sure what the actual trigger is.
In reply to Hmm, glad you learned how… by Marc Sabatella
>>> Attempting to add a dot to a quarter note on beat 10 of 12/8 (when used against 4/4) will cause corruption every time.
I see. I certainly don't know enough to disagree, but when I get to beat 10 the PROPER way, that is duration > dot > enter, I end up with a dotted quarter.
When I did it my old way, the only thing available was a quarter tied to an eighth which I could not tuplet.
In my attachment (guilty!) I showed what I expected when you enter a dotted note in compound meter (12/8,9/8, 6/8). Instead, after entering a dotted quarter to 12/8 you are left with an eighth rest followed by a half rest. Incongruous.
As someone who understands (perhaps develops?) the software I'm sure you know the justifiable reason for this. I'm also certain as a musician you'd agree it would make more sense for the 12/8 bar to breakdown the way I expected.
Perhaps a future issue.
I'm glad to have blundered into a discovery for you (us). If only I can now blunder my way into the next Le Sacre du Printemps.
In reply to >>> Attempting to add a dot… by penne vodka
Yep - entering the dotted note directly works, adding the dot later doesn't (in this specific case of a note near the end of a measure using a local time signature). As I said, this appears to be a very old but only newly discovered bug. And it's not just adding dot that fails - also increasing during using the "W" or command, or simply select a new duration using the toolbar or duration shortcuts (1-9).
BTW, another thing you may not have realized - you don't need to first enter a note and then select it in order to enter a tuplet. Just as you can enter the dot directly (not note first then add dot), you can also enter the tuplet directly (not note first then divide into tuplet). In note input note, to enter the tuplet I believe you intended (quadruplet across a dotted quarter) simply type 5 . Ctrl+4 (or use equivalent toolbar buttons / menu items - quarter note, dot, Notes / Tuplets / Quadruplet). No need to enter any notes whatsoever before this.
If I understand you correctly about the rests, this is a known annoyance. Specifically, after entering a note and MuseScore needs to fill the remaining space with rests, it always calculates them as if you were in 4/4 time - it doesn't understand the rest grouping rules for 12/8. No big deal, you can just enter the correct rests manually, just as you would in any other situation (eg, at the beginning of a measure).
In reply to I've been doing it precisely… by penne vodka
As fr as I can recall there haven't been any recent changes to how local time signatures work - they have La8ways* been somewhat problematic, and continue to be - but there have been changes in how tuplets work. As I recall, there was a pretty serious issue with certain tuplets causing corruption going all the way back to 2.1 that was partially addressed for 2.2 but that introduced other problems that were addressed again for 2.3. As far as I know that finally took care of it; I am not aware of any known corruption issues involving these tuplets. But local time signatures, as mentioned, have always been problematic.
So I am trying to understand if the problem here is something about local time signatures or something about tuplets or something about the combination of the two, and also whether something has changed since a previous release. Could someone please type a series of precise step by step instructions to reproduce the problem from scratch (eg, starting from File / New, or from the default empty score)?
In reply to As fr as I can recall there… by Marc Sabatella
I know this would have worked just fine in version 2.1, but causes corruptions in 2.3.2 so the problem started in a 2.2x or 2.3x release. Changing the last rest in a measure never cause a corruption for me before when using multiple time signatures. I don't currently have version 2.1 on my computer so I can't test it to verify this.
In reply to I know this would have… by mike320
Could be, but as I said, in order to investigate, we'd need precise step-by-step instructions to reproduce the problem. There are too many different files and scenarios in this thread for me to get a clear read.
In reply to As fr as I can recall there… by Marc Sabatella
Marc
FWIW - As I trundle through this piece in 4/4 (without local time)I find no problem with triplets, nor have I in previous scores.(2.3.2). I did suspect it was the combination, as you've suggested, because the problem only occurred with the combo and never when I wrote out the desired measure on an independent staff.
One more thing. I did not understand your last request. Did you want someone (me?) to recreate the process, step by step, more thoroughly than previous? I would not mind if it would help. MuseScore's been awesome.
In reply to Marc… by penne vodka
Yes, I would indeed like it if you could post the steps to recreate the problem. I see some information in a file called "Example_B", but that file has only a single staff, and I am not sure which measure of which of the several other files attached to this thread I would need to try that on. Something simple like this:
1) new score for two instruments, 4/4
2) ctrl+drag 12/8 to first measure of second staff
3) in staff ___, enter ___
4) etc.
In reply to Yes, I would indeed like it… by Marc Sabatella
duh...I was not entering dots properly. Should be duration > dot > enter
I was doing this: duration > enter > out of N > dot > enter dot
what happened is recreated in reply to mark320 with attached