Automatic Staff/System Spacing

Discuss the rules of notation, standard notation practices, efficient notation practices and graphic design.
Knut
Posts: 864
Joined: 05 Oct 2015, 18:07
Location: Oslo, Norway

Re: Automatic Staff/System Spacing

Post by Knut » 27 Mar 2017, 23:46

jan wrote:
27 Mar 2017, 14:26
Knut, here are two more before/after PDFs which show the optimization better than the video.

One new feature is the automatic violin clef change that allows more condensation on empty staves: Haydn font instead of Maestro (see example score 2 bottom staves).

The dynamic expressions now don't use rectangular boundary boxes anymore, but cut-out boundary boxes as in SMuFL. This allows for a better condensation for example in m.5 piano where the fermata box would otherwise collide with the mp box.

Adding additional clef changes would be another feature that would improve for example m.3 in bassoon 1 (the high note in that measure was only added for demo purposes).
Jan, I want to give you detailed feedback on this, but showing you is much easier than writing bullet points. If you attach the Finale file, I could implement all my preferred adjustments, which hopefully would give you a good Idea of the changes I feel is needed.

Did you just switch the treble clefs from one font (Maestro) to another (Haydn)? I wouldn't recommend this, unless you can switch between different clef styles in the same font.

Also, while I would agree with all of John's points under normal circumstances, a condensed layout (i.e., closest possible vertical spacing) require special liberties, and most, if not all, elements should be allowed to be placed at least partially within the staff. Send me the file and I'll show you what I mean.

Knut
Last edited by Knut on 28 Mar 2017, 07:24, edited 1 time in total.

Christof Schardt
Posts: 23
Joined: 17 Jun 2016, 10:00

Re: Automatic Staff/System Spacing

Post by Christof Schardt » 28 Mar 2017, 05:30

Absolutely astonishing work, Jan.

Agreeing to most of Johns points I would add three remarks regarding the orchestra score:

1. violin 1, bar 4,5: harpins significantly closer to the lower staff, which irritates. Obviously caused by the low mf in the bare before. But later we see other dynamics placed closer to the belonging staff.

2. strings in bar 13: flattening the slurs could improve the optical separation.

3. Bassoon: first note in bar 5: apparently an not input accident, isn't it?

And regarding the other score:

1. Flipping the 6-tuplets to the other side would remove the hairpin-collision without inferring new conflicts.
Opposed to that the triplet-figure 3 flipped to the top in the same bar leads to a new collision, which would require a reshaping of the slur.

2. The inclined hairpin is somewhat irritating, because it follows the melodic line of the lower staff, while its own "melody" is just a long note.
It could make sense to place it higher, even on top of the staff, to make it stay horizontal.
Software Developer, Musician and Engraving Aficionado
PriMus 1.1

User avatar
jan
Posts: 68
Joined: 22 Jun 2016, 08:29

Re: Automatic Staff/System Spacing

Post by jan » 28 Mar 2017, 07:27

Thanks for the input, John, Christof and Knut !
I am busy the next days, but will answer next week.
Here is a bit on Christof's and Knut's remarks.
Christof Schardt wrote: And regarding the other score:
1. Flipping the 6-tuplets to the other side would remove the hairpin-collision without inferring new conflicts.
Opposed to that the triplet-figure 3 flipped to the top in the same bar leads to a new collision, which would require a reshaping of the slur.
Hi Christof, if you read the first entries of this thread, the second (short) score excerpt was intended to be a "replicate" ofthe original condensed score that OCTO posted.
As I stated above tuplet number collision is currently not fully implemented in the Perfect Layout plugin yet, that's why it doesn't notice the tuplet-6/hairpin collision. I am working hard on that issue, but it's rather difficult as Finale doesn't return any metrics information at all regarding tuplets. Currently tuplets are only recognized as part of the staff skyline, but not in collision detection.
Christof Schardt wrote: 2. The inclined hairpin is somewhat irritating, because it follows the melodic line of the lower staff, while its own "melody" is just a long note.
It could make sense to place it higher, even on top of the staff, to make it stay horizontal.
The inclined hairpin is also a result of the replication of the original score. But it looks different as the hairpins below were not corrected as their was enough space. A bit difficult to put this into an algorithm.
Knut wrote: Did you just switch the treble clefs from one font (Maestro) to another (Haydn)? I wouldn't recommend this, unless you can switch between different clef styles in the same font.
You mean "unless I can" or "unless I may"? Finale allows to use clefs from different fonts, so yes I can. It's even possible to use different clefs in score and parts (though a bit difficult to achieve via staff styles). But does it look good?
There was a thread on the Music Engraving Tips Facebook group last week about clefs in condensed scores. And we only found 6 clefs that were designed for condensed scores at all. So there is not really a choice. Haydn and the first clef from Matthew Hindson's Clefs font are probably the only clefs that are accetable in this case:
short clefs for condensed scores.jpg
short clefs for condensed scores.jpg (175.04 KiB) Viewed 1493 times

Knut
Posts: 864
Joined: 05 Oct 2015, 18:07
Location: Oslo, Norway

Re: Automatic Staff/System Spacing

Post by Knut » 28 Mar 2017, 09:01

jan wrote:
28 Mar 2017, 07:27
Knut wrote: Did you just switch the treble clefs from one font (Maestro) to another (Haydn)? I wouldn't recommend this, unless you can switch between different clef styles in the same font.
You mean "unless I can" or "unless I may"? Finale allows to use clefs from different fonts, so yes I can. It's even possible to use different clefs in score and parts (though a bit difficult to achieve via staff styles). But does it look good?
There was a thread on the Music Engraving Tips Facebook group last week about clefs in condensed scores. And we only found 6 clefs that were designed for condensed scores at all. So there is not really a choice. Haydn and the first clef from Matthew Hindson's Clefs font are probably the only clefs that are accetable in this case:
short clefs for condensed scores.jpg
What I mean is that I would not recommend that a plug-in substitutes the clef with an entirely different design from another font. This should only be done if the assigned default music font contains both styles of treble clef (both an upright and slanted/rotated) as stylistic alternates for this specific purpose. Otherwise, the chances of it looking good are fairly slim.

User avatar
jan
Posts: 68
Joined: 22 Jun 2016, 08:29

Re: Automatic Staff/System Spacing

Post by jan » 28 Mar 2017, 09:04

Knut wrote:This should only be done if the assigned default music font contains both styles of treble clef (both an upright and slanted/rotated) as stylistic alternates for this specific purpose. Otherwise, the chances of it looking good are fairly slim.
I agree with that. It's indeed only an optional setting to automatically change the treble clef.
Do you know any font that has both styles of treble clefs?

Knut
Posts: 864
Joined: 05 Oct 2015, 18:07
Location: Oslo, Norway

Re: Automatic Staff/System Spacing

Post by Knut » 28 Mar 2017, 11:58

Jan,

Here's the corrected file. It's by no means perfect, to a large degree because of Finale's default settings and flaws in the design of various Maestro symbols, but I hope it gives you a good idea.

I've applied Patterson Beams (with a Ross-esque setting), first and foremost to limit the stem lengths. Stem lengths are shortened further wherever appropriate, either manually or using JW Change. I'm not sure how best to deal with stem lengths for your plug-in, considering there are numerous workflows available and most serious users will probably use Patterson Beams prior to your plug-in, but you should probably offer a limited shortening of stems relative to the values set prior to plug-in execution.

I have generally moved dynamics closer to the staff, and flipped articulations wherever it will save space. I've even changed some of the staccato-accent marks to separate staccato and accent articulations to improve/conventionalize their placement.
test for system minimization-fixed.pdf
(118.64 KiB) Downloaded 52 times

User avatar
John Ruggero
Posts: 1188
Joined: 05 Oct 2015, 14:25
Location: Raleigh, NC USA

Re: Automatic Staff/System Spacing

Post by John Ruggero » 29 Mar 2017, 11:30

That looks great, Knut. Very readable.

What is such a "condensed" score used for? Movie scores? I thought a condensed score was one in which all empty staves were omitted.
Mac mini (OS 10.8.5) with dual monitors, Kurzweil Mark 5 with M-Audio Midisport 2 x 2,
Finale 2014d with GPO 4, JW Plug-ins, SmartScore X Pro, Adobe InDesign CS4,
Inkscape .48.5 and .91, FontForge 20150526
http://www.cantilenapress.com

Knut
Posts: 864
Joined: 05 Oct 2015, 18:07
Location: Oslo, Norway

Re: Automatic Staff/System Spacing

Post by Knut » 29 Mar 2017, 20:52

John Ruggero wrote:
29 Mar 2017, 11:30
That looks great, Knut. Very readable.

What is such a "condensed" score used for? Movie scores? I thought a condensed score was one in which all empty staves were omitted.
Thanks, John!

You are right, of course, that all empty staves should be omitted for it to be a truly condensed score. I merely used the term in reference to the purpose of the plug-in, which is to space the staves as closely as possible. This kind of spacing is of course very common in large tutti sections or more generally, whenever one is forced to fill pages to their brink with music.
.
Perhaps the test score should have been truly condensed, but I trust that Jan has a reason for doing it this way.

User avatar
John Ruggero
Posts: 1188
Joined: 05 Oct 2015, 14:25
Location: Raleigh, NC USA

Re: Automatic Staff/System Spacing

Post by John Ruggero » 30 Mar 2017, 02:08

Now I understand that this is just an exercise, Knut. In actuality, it would seem best to allow the staff and page size to be determined by the requirements, rather than packing as much music on a page as possible. So the first task would be to determine what is the acceptable range for the positioning of all elements on the basis of readability and good style and then go from there, which I assume is what jan is doing.
Mac mini (OS 10.8.5) with dual monitors, Kurzweil Mark 5 with M-Audio Midisport 2 x 2,
Finale 2014d with GPO 4, JW Plug-ins, SmartScore X Pro, Adobe InDesign CS4,
Inkscape .48.5 and .91, FontForge 20150526
http://www.cantilenapress.com

User avatar
jan
Posts: 68
Joined: 22 Jun 2016, 08:29

Re: Automatic Staff/System Spacing

Post by jan » 03 Apr 2017, 07:27

Regarding clefs for condensed scores: another solution is to slightly rotate the clefs.
So you don't mess up the clef style by mixing different fonts.
In this video the clefs from the two top staves were rotated by 0.06° with a JW Lua script in Finale.
It's a bit tricky to do in Finale's Shape Designer which only has a rotate for arrowheads, but very easy with a script as it supports the rotate command for all (!) elements - seems to be a hidden Finale feature.
https://www.youtube.com/watch?v=J6qnFet9Yqw

Post Reply