Note spacing - voice collision in Dorico et al.?

[John Ruggero]

I was going to suggest the same idea, before your first example yesterday. But I didn’t feel confident that anyone would agree if I didn’t have the time to illustrate it properly. Good to know this was indeed an engraver’s trick for optical alignment.

Thanks so much, MJCube. It is good to know that we are in accord on this one, optical illusions being in the eye of the beholder. I wasn't aware that this particular case of opposite stemming was a standard engraver's trick of the past until Knut mentioned it, but I may now do a little exploring to document this other similar effects.

[John Ruggero]

This is very strange, as it gives me the exact opposite impression. The compensated beats now look clearly misaligned to me.

Oops. I thought you meant that the perfectly aligned ones at A would look better if the staves were closer together. To me they look worse. As far the compensated ones at C, they may not be perfect, as I said, but look better to my eye than A and no worse with the staves closer together. They don't look badly aligned at all to me, actually, so this may be one of those questions of taste or perception.

Do you know of any publisher who used this technique more or less generally in their publications?

No. It is just something that I would do by instinct and probably derived from hand copying, where everything is done by eye. But I might now do some investigation.

I'm guessing there exists literature on the topic of music engraving in both german and french libraries, attics or cellars, but nothing generally available.

That's what I guessed as well. This stuff might have been closely guarded in-house secrets. And there would have been no general market for books on music engraving. But somewhere there exists notes by an apprentice buried in a library or in an attic...

[Knut]

Oops. I thought you meant that the perfectly aligned ones at A would look better if the staves were closer together. To me they look worse. As far the compensated ones at C, they may not be perfect, as I said, but look better to my eye than A and no worse with the staves closer together. They don't look badly aligned at all to me, actually, so this may be one of those questions of taste or perception.

Yes, that's what I meant. To me, the unedited beats in the last example look aligned (more so than in the first example), while the compensated beats look misaligned.

It is just something that I would do by instinct and probably derived from hand copying, where everything is done by eye. But I might now do some investigation.

I'm guessing that my observations of this phenomenon in plate engraving might be due to the engraver working more freely when sketching the music on the plate (which, to my knowledge, was usually done before punching or the symbols in), thereby using similar principles to a hand copyist.

[John Ruggero]

To me, the unedited beats in the last example look aligned (more so than in the first example)

Intriguing. For me, the effect is greater as the staves approach each other, possibly because angle between the stems becomes greater.

What is also interesting, is that I see the illusion only in case A of the following possibilities. B, C, and D look perfectly aligned. For me, the effect lessens as the staves move away from each other and the measure widens. Other melodic shapes seem to display the same effect.

Spacing X.jpg (36.4 KiB) Viewed 10839 times

Spacing Y.jpg (35.52 KiB) Viewed 10839 times

Spacing Z.jpg (58.55 KiB) Viewed 10839 times

[Knut]

To me, the unedited beats in the last example look aligned (more so than in the first example)

Intriguing. For me, the effect is greater as the staves approach each other, possibly because angle between the stems becomes greater.

What is also interesting, is that I see the illusion only in case A of the following possibilities. B, C, and D look perfectly aligned. For me, the effect lessens as the staves move away from each other and the measure widens. Other melodic shapes seem to display the same effect.

That's a much clearer illustration, John, in which I can totally relate to what you're talking about. I also think you are right on the money with your explanation of why the effect increases as the distance between the notes decreases.

However, to me, it seems to greatly depend on where the focus lies when reading. If you focus primarily on the stems (which I'm guessing might be a common point of focus for anyone sight-reading a grand staff), the effect becomes very apparent. With a broader or shifting focus that encompasses the noteheads as well, the voices seem perfectly aligned to me. I also suspect there to be a greater perception of misalignment for staves with facing stems (A) than with facing noteheads (D), but I'm not entirely sure.

In any case, I do not think I would compensate for this effect, since it would just as easily lead me to perceive the compensation as misalignment.

[John Ruggero]

Knut, I'm glad I was clearer this time and you could see the illusion better. As you said, the stems attract attention, and this leads the eye along a strange path and away from the note heads. I agree that if one corrects for it, a sort of entasis, it should be very slight. I have always done such occasional slight adjusting intuitively as I work and without analysis, as if I were hand copying. If a measure doesn't look right, I move things until it does. Now I am going to try to understand the why of it. But it is probably only in the context of fudge-factoring that it has practical general application.

[Knut]

But it is probably only in the context of fudge-factoring that it has practical general application.

I agree completely.

[Knut]

BTW, my question about the spacing of seconds have thus far been ignored by the staff on the Dorico forum. This leaves me a bit worried whether they actually intend for Dorico to handle these situations similarly to Lilypond or not, that is: correctly or incorrectly, at least in my personal opinion.

Update: This behavior seems to be due to a bug, and according to Daniel, Dorico is intended to work the same way as Lilypond when it comes to spacing notes following seconds in different voices.

[OCTO]

BTW, my question about the spacing of seconds have thus far been ignored by the staff on the Dorico forum. This leaves me a bit worried whether they actually intend for Dorico to handle these situations similarly to Lilypond or not, that is: correctly or incorrectly, at least in my personal opinion.

Update: This behavior seems to be due to a bug, and according to Daniel, Dorico is intended to work the same way as Lilypond when it comes to spacing notes following seconds in different voices.

One of the reasons I am looking forward to seeing more from Dorico is the spacing issues found in Finale anf Sibelius. Finale is definitely better than Sibelius at this point, but yet unable to deliver more sophisticated and truly important spacing algorithms. For that reason the issue with Dorico seems to be a bit disappointing.

[Knut]

BTW, my question about the spacing of seconds have thus far been ignored by the staff on the Dorico forum. This leaves me a bit worried whether they actually intend for Dorico to handle these situations similarly to Lilypond or not, that is: correctly or incorrectly, at least in my personal opinion.

Update: This behavior seems to be due to a bug, and according to Daniel, Dorico is intended to work the same way as Lilypond when it comes to spacing notes following seconds in different voices.

One of the reasons I am looking forward to seeing more from Dorico is the spacing issues found in Finale anf Sibelius. Finale is definitely better than Sibelius at this point, but yet unable to deliver more sophisticated and truly important spacing algorithms. For that reason the issue with Dorico seems to be a bit disappointing.

I think most professional engravers and composers share your hopes and expectations for Dorico. It's spacing algorithms are already better than any UI based application, and aims to be at least on par with Lilypond.

As I said, this is only a bug, so it will hopefully be straightened out in the reasonably near future. I'm much more concerned with how manual spacing of notes will be implemented, but optimistic nonetheless.