Global stem customization.

• Apr 17, 2019 - 15:46

I find the stem lengths generated by ms to be fractionally too short to my taste especially where beams are attached. Usually I need to adjust every beam height and angle in a score. This is particularly troublesome where crotchets have slightly longer stems than a pair of beamed quavers of the same pitch.

Another situation I find needs manual adjustment is where there is a rising (or falling) sequence of beamed notes with same interval exists between consecutive notes, yet the beam doesn't follow the same angle as the note heads - the stems shorten to the right.

Are there any global controls that will allow:
1) a small delta to be added to the calculated length of all stems?
2) a way to make beams follow the contour of a rising or falling group?

I guess the answer to these questions is no. So what I'm really asking for is more global controls to influence the beam placement algorithm.


Comments

In reply to by Jojo-Schmitz

That setting make almost no difference. I see only one or two isolated crotchets with no beams being affected. Here are a couple of screen grabs that illustrate to point:
1) compare the stem lengths of crotchet E in 1.Fag with the next pair of quaver Cs. The stem lengths are more or less the same yet the notes are a 3rd apart. Compare that with 2.Fag where I've lowered the quaver beam, which is about where I'd like it.
2) look at how the notes in the upper part rise sequentially, yet only the last quaver seems to have the right length stem. I would have to adjust that so that the beam is parallel to the line of the note heads.
3) if you look at the lower line, the beam is horizontal, yet overall the group of notes rises and certainly the last three of the four rise. I would want the beam to rise slightly rather than be horizontal.

The shorten stems settings are for a very specific case - unbeamed note whose stems extend outside the staff (eg, in multivoice settings, upstem notes above the staff or downstem notes below the staff).

Aside from that, the engraving rules for stems of unbeamed notes are pretty universal - it's always the span of an octave (3.5sp), pretty much without exception. You really shouldn't be deliberately violating those standards often if at all. But for beamed notes, there is a lot of subjectivity, so indeed there may be cases where you'd might reasonably want to deviate from the defaults. Since beaming is so complex, I can't see that a global option to length all stems would be of much value. An option to prefer following the slope could be interesting, but realize this too is quite non=standard - it's not how professionally engraved music is normally produced. Beams are supposed to stay flatter than that. In fact, it's quite common to follow a rule where beams never slop more than 1sp no matter how large the intervals. Not saying we would never provide options to customize the algorithms, but in general, we're more likely to provide options to make it easier to get closer to how a professional engraver would do it, as opposed to options that make it easier to get further.

In reply to by Marc Sabatella

If you provide options to get closer to what professional engravers do then I would be very happy. But you have to abandon a notion of "a" universal standard. There are many standards - or house styles. Some of the finest examples for classical music are provided by Bärenreiter, Henle and HH Edition. They tend to use longer stems and tend to avoid the wedge-shaped appearance of some of the examples I gave. It is true that cost expediency has caused some engravers not to fight the impositions of the typesetting s/w. But that's not a good basis for establishing a standard. When I look closely at some of the Bärenreiter engraving I see that they do have flatter beams - as you suggest - but what allows this to look much more elegant and visually acceptable is their use of longer stems. I don't see any examples where beamed F quavers in treble clef don't span the entire stave as one of my examples showed. If I want to put this in a nutshell: it's the overall stem lengths of beamed groups I find often too short; and the inconsistency between stem lengths of notes with beams and those without them particularly when such examples they are adjacent to each other and only one pitch is involved.

In reply to by Marc Sabatella

Easter got in the way. I've done a bit more research on this. I fired up my old copy of Finale (Finale 2000 - yes it's been that long). That version seems very long in the tooth compared with musescore. However beam customization did show up some interesting options, which can be seen here:
https://usermanuals.finalemusic.com/Finale2012Mac/Content/Finale/IDD_BE…
I am definitely no suggesting that musescore should adopt similar options but I throw this in as an example of how beam deployment styles can vary. I found a concrete example from Bärenreiter (Schubert Octet), which is set using Sibelius that illustrates one of the aspects of beam placement I would love to see some global option to address. I've copied the bar in question as musecore would set it using the default template, which specified shorten stems. Note in particular how the 2nd beat quaver stems are shortened in musecore. Bärenreiter seems to have chosen not to shorten stems - they are the full 3.5sp. But also note that the beamed groups use the same stem length. I prefer to uncheck the shorten stem option in musescore but that exaggerated the difference between the beamed groups and the non-beamed notes.

I've included scans that show some examples of beaming that differs for mscore. I have to say you'll have a hard job diving an algorithm: all off these do largely what mscore does but every now and then they do something different. Out of the examples given: Baerenreiter use Sibelius and are highly regarded. Prariedawg is a small specialist American publisher that uses Finale. Novello and Musica Rara are older examples - not necessarily supportive of my case.

I do wonder whether the matter of stem lengths of beamed groups is down to manual intervention in Baerenreiter's case. I know a typesetter that works with one of their engravers; they seem to spend an inordinate amount of time making small adjustments by eye to achieve the desired visual effect - that's your let out clause.

What I would ask in summary is not what Finale does which is to provide a number of esoteric style options, but rather for a few tweaks to the mscore algorithm what allow some generic influence. For example offsets to stem lengths and maybe some intermediate control over flattened beams and non-flattened beams.

Attachment Size
beams-example.mscz 3.41 KB
baerentriter.jpg 943.06 KB
prariedawg.jpg 475.62 KB
novello.jpg 815.22 KB
musicarara.jpg 1.03 MB

In reply to by richardm999

Actually we do provide quite a few of those same options. What we don't have is "maximum slop", but it seems that's the opposite of what you want anyhow. Well, I guess if you look at it as, we're already enforcing a maximum slope and you want to increase it, that makes sense, but unfortunately that's not quite how the code works. Totally would be worth someone's time to do a major cleanup and reorganization fo this code to make such things easier, but it's a very big job.

Meanwhile, though, I don't see what you mean in the Baerenreiter example. i thought you wanted the beam to follow the angle of the notes more literally like in an ascending arpeggiated passage C-E-G-C ? As I said, that's something that is pretty universally frowned upon as it leads to very steep beam angles. Maybe you really meant, honor the slope of the outer notes in a mixed-direction passage, like measure 22, but still keeping things relatively flat? That's more reasonable indeed. But as you surmise, this sort of stuff is usually adjusted by hand, as it is incredibly subjective. Not just from one editor to another, but between passages in a single work depending on what else is going on. For example, look at bar 42 - here they are using a flat beam even though it's virtually the same situation as 22. You'd have to ask the engraver why.

FYI, the shorten stem option is only relevant for notes whose stems extent above/below the staff - eg, notes on ledger lines in multiple voice situations. Doesn't appear to come into play in any of these examples.

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