Request option to shorten stems for downstem notes with flags (using alternate flag)

• Apr 13, 2016 - 11:43
S5 - Suggestion

Wrong stems.mscz
Should look different.


1) Critical priority is reserved for crashes and corruption. That is not the case here.

2) Go to: Style -> General -> Notes -> Shorten Stems -> Adjust the "Progression" to 0.00 sp. (instead 0.25s sp)
Result: Wrong stems1.mscz
And for the measure 1, ability to edit the stem length.

Status (old) active needs info

Please use the forum for questions first. From the score I can gather that you think/feel it should look different, but I don't see a real problem with it.
A priority of critical is meant for bugs that cause either crashes and/or data loss, please see… on information on how to make the bug report better understandable for the developers.

Anyhow, if in your score you zoom in enough and select the stem of the single 1/8th note, you can clearly see it has the exact same length as the stem from the beamed notes and the 1/4th note.

Both the beam and the flag are added right below that red line, so to me those are aligned.

Do you have any references as to how it should look and why you feel that way?

If a score writer is not able to create in such simple case correct stems by default, it is a bug. Read Elaine Goulds "Behind bars".

Status (old) needs info active

If you move the tem of that last beamed note and the flag of the next note to the right, you see that the stems are not aligned, 3 different length stems.
If you flip the stems, it looks about right (but the beamed stem is still slightly to short, not touching the far edge of the beam)

Status (old) active needs info

Can someone explain in more than two words what the expect result should be? Also, add a good title for this issue, what's wrong with stems exactly?

If not, I will close this issue for lack of information.

Title Wrong stems Inconsistent stem lenght on downstem notes between beamed, flaged and normal notes

Don't the images speak for themselves?

I think that's by design. When a note is outside the staff and has its stem pointing away from the staff, its stem is deliberately shorter than normal (and conversely, if the stem direction is flipped, you will see it is deliberately longer than normal). But for notes with flags, there wouldn't be room for the flag if the stem was shorter.

Status (old) needs info by design

I think Isaac describes the situation correctly. Stems are generally shortened for notes extending above/below the staffis some, but you simply can't shorten flagged notes by the same amount you shorten notes without flags. Notes with flags need stems long enough for the flag.

So this is indeed by design, and it's a correct design. See for instance Gould p. 14-16. She suggests the possibility of shortening flags as well for notes with shortened stems, but also clearly shows the possibility of keeping longer stems as a valid option. And an examination of the published literature shows it to be how many publishers choose to do it. See for instance the Schirmer edition of Bach WTC I, the very first prelude or fugue. The middle voice includes mostly flagged notes extending above or below the staff, and they clearly use full length flags and corresponding stem lengths, whereas the unflagged notes clearly use stem shortening, resulting in different stem lengths for the same note depending on whether it is flagged or not

1. Look again:
* middle voice, on nearly all the tied notes, the flagged dotted eighth stem is longer than the quarter following; Very notable so in measure 8
* 2nd page, 3rd system, 1st measure, the half notes F in the bottom staff have shorter stems (not even equally short!)
Both situations have same pitched notes above/below staff with varying stem lengths.

You've told the devs to look at Behind Bars for an example; they did and explained to you they're even following your reference.
You claim Marc's example is incorrect, while I (who most certainly know very little about the matter and the 'correct' way to handle stem lengths), can clearly pinpoint at least 2 examples in less than a minute in the sample you provided.

That most definitely warrants for this issue to not be considered a bug.

2. Tastes and Colo(u)rs…
At the very best you could convert your ugliness issue into a feature request for an 'equal stem length' option.

Exactly. The G. Henle Verlag edition shows the same. Virtually every measure, starting with the first, contains a flagged eighth tied to an unflagged note (so obviously the same pitch), and the stem lengths clearly differ any time the unflagged note has its stem shortened (which happens very often, starting with the "E" in measure 1). That's the *very first measure* on the *very first page* of the *very first score* I chose to look at to find examples. You don't have to look very hard to find countless more.

This should be be enough to settle the matter conclusively - the behavior is not a bug nor is it disugsting; it is normal proper notation. If you wish to notate your music in a different way, you are welcome to manually adjust your stem lengths, but by default we are doing it right.

The edition is 125 years old. So I had a look to more modern editions.
Bärenreiter Don Giovanni have the same long stems
Beethoven Sonatas Edition Peters, same long stems

And I did not find situations like stemend move in a different direction than pitch.


Yes, as I mentioned, there are multiple valid solutions to the problem of stem length on notes that extend outside the staff. And there is no getting around the fact that different stem lengths are often required for a note depending on whether it has flag, beam, or neither. Consider also the situation of multiple beams - make those sixteenths or thirty-seconds in your example above and it becomes clear there is and can be no general rule that says the stem lengths must follow the direction of the line in all cases - it's just not always viable. You won't find such a rule in Gould, and again, it is trivially easy to find counterexamples in publisher literature from *any* century. There simply is no "right" and "wrong" about it, just different engravers with different personal tastes.

So there is no way the defaults can be exactly what everyone expects. Different highly respected professional engravers have different opinions, and indeed, they often change their approach from measure to measure according to other considerations. The best we can do is provide reasonable defaults and some ways of overriding those defaults. Which is exactly what we do. Our defaults clearly match what some highly respected engravers do, and we provide multiple overrides in the form of settings to control stem shortening and beam thickness / spacing, plus of course the ability to manually adjust stem lengths and beam position however you like.

Muse Score behave different than your SchirmerScore stem direction up.

Looking at Elaine Gould (p 62 in the German version, your page number +-3 to 62) showed this inconsistent behaviour. Wrong Stems 3.mscz Why is there a difference, because of the numbers of lines.

Giuliani LArghetto Op 50 nr 17.mscz Look at the stems of the quarter in the Bassline, they are to short compared to the stems of the eighth with beams and other stems of quaters.

Looking at

The steming have a bug. Make in the Giuliani-File 5 more line in the system and look how the stems of the quaters in the bassline are changing.

Title Inconsistent stem lenght on downstem notes between beamed, flaged and normal notes Request option to use short flags for notes with shortened stems
Status (old) by design active

Different lengths of stems due to different staff lines is a separate matter. Feel free to file a separate report if you see a problem there. I am not seeing it in 2.0.3, however - stem lengths are the same regardless of number of staff lines. That was a bug I remember fixing after the release of 2.0.2

For for stem lengths on a standard staff, I spent considerable time looking at different editions to see if there really is anything "wrong" with what we are are doing. What I found supports what I said: different editros do things differently, and there really is no one single correct rule. THe majority of the time, I *do* see notes with flags having longer stems than notes without flags. This is true for the Schirmer editor of WTC I mentioned, it is *also* true for the Peters editions of Beethoven Sonatas you mentioned. In fact it was true for around 75% of the editions I consulted.

So again, if you prefer something diffierent from what MuseScore does, simply change the default settings I mentioned, or use manual overrides. One thing you might personally prefer is a longer "Shortest stem" setting, or a smaller "Progression" setting. It will make the music look different from how most editions do things so I am not in favor of changing the default, but it might achieve more of the look you apparently prefer.

Meanwhile, I do consider it a vaid feature request to add an option to use short flags for notes with shortened stems, so that it is easier to achieve that less-common look if you desire it. So I have changed this issue accordingly.

Here, btw, is an image shoing the Peters edition of the Beethoven sonata does in fact use longer stems when necessary for flags:


See the top staff - there are two D's, one with flag and one without, and the one with flag has a longer stem. Again, this is consistent with about 75% of the dozen or so editions I consulted while onvestigating this matter. There really can be no question this is common, standard behavior.

Title Request option to use short flags for notes with shortened stems Request option to shorten stems for downstem notes with flags (using alternate flag)

And here is Schirmer WTC - with stem direction up:


Other editions do shorten the stems in the upstem case - just as we do - because the flag is opposite the notehead and thus won't collide. So we actually do "better" than Schirmer and Peters in the upstem cases, if your measure of "better" is, more consistent stem lengths. It's really only the *downstem* notes where we don't currently shorten stems for flags notes (nor do the majority of publishers), but we could conceivably add an option to do that. I saw only a very small minority of editions doing this, but it certainly is possible to do amd doesn't look bad.

Beams are a more complex matter. For one thing, most of the time when you have beamed notes, they aren't all the same pitch, so the stem length depends on the *other* pitches. For another thing, there may well be more than one beam, in which case you cannot avoid the need for longer stems.

That said, it is true we could conceivably shorten the stems a little more for the very special case of single beam, all same pitch, for notes extending above/below the staff, so that the stem length for beamed notes is a closer match for notes with no beam or flag. I would propose tying this to the same option we I am proposing for flagged notes. So the default is a more standard stem length (which is to say, varies according to beam or flag) but if you enable the "consistent stem length" option, then we use shorter flags for downstem flagged notes *and* shorten stem lengths in the special case of single beamed notes of the same pitch. It's very much a corner case, and again there is little consistency in how different editors handle this, but it does make some sense.

Meanwhile, if you like more consistent stem lengths, you can disable the stem shortening entirely, or reduce the extent of it.