Hello all,
I’ve attached a project with a bunch of beam angles for groups of two eighths and two sixteenths, somewhat loosely based on Ted Ross’s beam prescriptions. I’m generally pretty happy with my eighth note beam angles (they don’t match Ross’s beam slants 1:1, but they’re close enough for me), but with my current slant and stem shortening settings the beam slants for sixteenth notes are very inconsistent. The attached file (with all Engraving Options set to factory except Beams and Notes > Stems) should help illustrate things, but the short version is that for certain intervals I tend to get either very shallow or very steep slants within that interval type, and I will often get inconsistencies between intervals (usually with fourths having a 1 or 1 1/4 space slant and fifths having a 1/4 space slant).
I’m not married to Ted Ross’s exact prescriptions and I’m not opposed to having softer eighth note beam slants, but I do prefer having some angularity by default over something like Henle’s very flat beams, and I mostly just want some consistency between different intervals in my sixteenths as well as my eighths.
Beam angles.dorico (1.7 MB)
2 Likes
I have similar issues even if I am not necessarily following Ross’s prescriptions.
Whatever setting I change improves something and breaks something else. Beam slants on their own mean nothing without stem lengths, stem shortening rules, etc.
I am actually turning off every auto-shortening of beamed groups so that I do not get extra behaviour that I do not necessarily look for. As you found out, I also can manage to get 8th-note groups quite consistent, but then 16th-note groups are a nightmare. The main problems seem to be around 4th-5th intervals which can be 1-sp slanted in certain situations and 1/2-sp in others. You get one right and you break the other.
I have proposed this before, to have a library of beam slants so that things would just look as one wants. It would then become possible to save a “Ross” library, “UE” library, “French-style” library, etc.
Since this is ‘just’ aesthetic and not truly a functional problem (no crash, no clear bug), I am keeping my hopes low for now. I just try to get as reasonably close as possible. If I see that I am editing more than a handful of beams per stave, then I change something.
Another possible improvement could be to have flow-specific settings because in pieces such as Tchaikovsky’s The Seasons Op 37a one needs certain settings for one movement and other settings for another. I managed to tweak things to get January basically perfect, just to find that February was a mess …
4 Likes
Fourths and fifths are probably the intervals with the most problems, but sixths and sevenths can get pretty weird too. This stretch of intervals is particularly egregious imo.
I think some of it comes from using a slant value of 7/8 for fourths, fifths, sixths, and sevenths in Engraving Options > Beams, and bumping that up to a value of 1 for fourths and 1 1/4 for the latter three intervals does put steeper slants on the fourth and sixth in this stretch.
However, the reason I had fourths through sevenths set to 7/8 is because at any values higher than that for those intervals, the eighth-note slants going from the outermost space to a note on a ledger line get quite ugly:
Both Ross and UE recommend this, which can only be achieved with a slant value of no higher than 7/8:
I know that there’s a pretty complex interaction under the hood between beam slants, stem shortening, and snapping beams to “legal” positions, but the settings are kind of a weird combination of being very granular while also being too broad. Maybe a solution could be being able to set values for 1, 2, and 3+ beams separately in Beams > Slants, but making the settings for 2 and 3+ beams mirror the eighth note settings by default? Or maybe something like a word processor’s dictionary, where if you make a manual override to a slant in a beamed group you can right-click and save that override as default.
Apologies for the dissertation, I know this is all incredibly nitpicky and there are certainly many more important bugs to squish or features to add, this is just something I care a lot about in my scores and parts.
4 Likes
I spent a lot of time on this a few years back. I haven’t re-evaluated my settings in a while, but here’s what I get by default with my settings:
The eighth notes are very close to the Ross prescriptions, only missing two beam starts and each by only 1/4 space.
You don’t actually want this, because it will place many of the secondary beams in “illegal” positions. A single beam can Sit on a staff line but once a secondary beam is introduced, a primary beam can no longer be placed in a Sit position. If I keep the Sit-Hang beaming of your last example and introduce a secondary beam, I get this:
Obviously the secondary beam can’t be left floating in a space here, so the 2-beam solution can’t match the 1-beam solution.
EDIT: typos
3 Likes
(Just to clarify, that Sit rule I mentioned above is for groupings with secondary beams that are beamed stems-up. It’s flipped for stems-down and the primary beam can’t use Hang.)
I don’t think Ross ever explicitly says this, but if you look through all of his examples on pages 104-110, he never breaks the following rule:
An ascending beam contained within the staff cannot begin with Sit and cannot end with Hang. A descending beam contained within the staff cannot begin with Hang and cannot end with Sit.
So in the case of the 16th beams in the last example, as they are descending beams contained within the staff, they cannot start with Hang and cannot end in Sit. As I just showed in the post above, they also cannot start with Sit as that would place the secondary beam in a space. So the only “legal” start position for the primary beam here is Straddle.
The only real decision is whether to end in Straddle or Hang unless …
… you take the Henle route.
6 Likes
Yeah, by “make the slants for 2 and 3 beams the same as eighths by default” I meant that the current behavior is set by default, where you set a slant angle for a given interval regardless of the number of beams and then Dorico snaps the beams to legal positions to avoid the endpoints creating wedges or a beam in the staff hanging out in midair.
1 Like
