Why SCORE is so good, an opinion

[jrethorst]

Interesting discussion of notation strategies:

http://www.jeffreygrossman.com/engraving.html

I wonder how much of his discussion applies to LilyPond.

[tisimst]

Interesting discussion of notation strategies:

http://www.jeffreygrossman.com/engraving.html

I wonder how much of his discussion applies to LilyPond.

This is a great explanation of what really does make SCORE so nice, to name just a few reasons. No doubt there are more.

How does his discussion relate to LilyPond? Quite similarly, actually. LP's horizontal AND vertical spacing algorithms are excellent and flexible. Each element (called graphical objects, or "grobs" for short) has numerous position properties, all available to the user. Changing some values may allow LP to adjust its calculations (like X-offset, Y-offset, etc.), others are taken into account after LP does its magic (i.e., extra-offset), causing the manual adjustments to have no change whatsoever on any other element on the page except the one the user wanted to change.

1) Note Spacing. Normally, LP uses optical spacing so that consecutive notes "appear" the same distance apart, even if they might not actually be. This aids readability, of course. You can turn this off to make the spacing more uniform. You can also force the spacing to be exactly proportional to its duration (i.e. the horizontal space given to a whole note is the exactly the same as four quarter notes).

2) Space Allotment. If an added accidental doesn't barge into the allotted "space" of other elements, then the horizontal spacing is not disturbed. This also means that various music elements are "kerned" where possible.

There are only two things I can think of that really could use some improvement:

1) Vertical flexibility of spanners within a system. For example, in a system of piano music, it is common to put the dynamics between the two staves. In LP, this is most easily done with a dedicated dynamics line. However, this forces the dynamics to remain at the same vertical height which can cause the staves to be farther apart than necessary when part of the dynamics line can move up and the other down. This can be adjusted manually, but that's the only way if done with a dedicated dynamics line. I would love to see an implementation that allows for this flexibility automatically, just as there already is between staves, systems, etc.

2) Lyric horizontal positioning. LP already does a pretty good job with this, but, like the vertical rigidity of a dedicated dynamics line, each lyric syllable is quite rigid in its horizontal placement. I think that if Lyrics could "flex" horizontally to favor the original note spacing, then this would be a tremendous leap forward. Again, you can manually adjust a syllable's horizontal placement, both with or without affecting other syllables.

So, the takeaway to all this is that I believe that under the hands of a knowledgeable user, LP is more than capable of beautifully handling challenging scores and giving the engraver a better starting place when manual edits are necessary.

P.S. Just for fun, here's LilyPond's default output of the complex tuplet spacing example on Jeffrey Grossman's site:

tuplets-LP-2.19.36.png (465.64 KiB) Viewed 11482 times

[DatOrganistTho]

Interesting discussion of notation strategies:

http://www.jeffreygrossman.com/engraving.html

I wonder how much of his discussion applies to LilyPond.

This is a great explanation of what really does make SCORE so nice, to name just a few reasons. No doubt there are more.

How does his discussion relate to LilyPond? Quite similarly, actually. LP's horizontal AND vertical spacing algorithms are excellent and flexible. Each element (called graphical objects, or "grobs" for short) has numerous position properties, all available to the user. Changing some values may allow LP to adjust its calculations (like X-offset, Y-offset, etc.), others are taken into account after LP does its magic (i.e., extra-offset), causing the manual adjustments to have no change whatsoever on any other element on the page except the one the user wanted to change.

1) Note Spacing. Normally, LP uses optical spacing so that consecutive notes "appear" the same distance apart, even if they might not actually be. This aids readability, of course. You can turn this off to make the spacing more uniform. You can also force the spacing to be exactly proportional to its duration (i.e. the horizontal space given to a whole note is the exactly the same as four quarter notes).

2) Space Allotment. If an added accidental doesn't barge into the allotted "space" of other elements, then the horizontal spacing is not disturbed. This also means that various music elements are "kerned" where possible.

There are only two things I can think of that really could use some improvement:

1) Vertical flexibility of spanners within a system. For example, in a system of piano music, it is common to put the dynamics between the two staves. In LP, this is most easily done with a dedicated dynamics line. However, this forces the dynamics to remain at the same vertical height which can cause the staves to be farther apart than necessary when part of the dynamics line can move up and the other down. This can be adjusted manually, but that's the only way if done with a dedicated dynamics line. I would love to see an implementation that allows for this flexibility automatically, just as there already is between staves, systems, etc.

2) Lyric horizontal positioning. LP already does a pretty good job with this, but, like the vertical rigidity of a dedicated dynamics line, each lyric syllable is quite rigid in its horizontal placement. I think that if Lyrics could "flex" horizontally to favor the original note spacing, then this would be a tremendous leap forward. Again, you can manually adjust a syllable's horizontal placement, both with or without affecting other syllables.

So, the takeaway to all this is that I believe that under the hands of a knowledgeable user, LP is more than capable of beautifully handling challenging scores and giving the engraver a better starting place when manual edits are necessary.

P.S. Just for fun, here's LilyPond's default output of the complex tuplet spacing example on Jeffrey Grossman's site:

tuplets-LP-2.19.36.png

This is a fabulous response. Thanks for it!

Also, @jrethorst, I've been looking for that website for a while.

[DatOrganistTho]

To add my own comment to your link: They are very similar. What differences there are boils down to the editing process. LilyPond loves it when you enjoy its default output, but doesn't care so much for customizing the final look down to a tee. The tweaks are obtuse and clunky, and you are required to learn a fair amount of coding in order to do any heavy customization yourself (without pre-written tweaks).

That being said, I'll take LilyPond's default (and all the musicians that work with me and my scores) over Finale without a heartbeat or eyelash past.

[jrethorst]

Very nice engraving, Abraham. What font is that?

[tisimst]

Very nice engraving, Abraham. What font is that?

Thanks! It's MTF-Cadence.

[OCTO]

SCORE allows absolutely unparalleled, complete control over every item on the musical page.

Yes yes..

you can manipulate all 19 of SCORE's slur parameters

What to say more?

When I see a slur colliding with an accidental, I can just adjust exactly the part of the slur that needs to be changed

Finale? Sibelius?

The other part of SCORE... is that nothing changes unless you want it to.

What to add more?
No fonts. There is a very little chance that your output will be a messy PDF or collide with printer. Just outlines. And for all outlines you can adjust their size.
IMO, the biggest problem of F and S is that they are sequencers, and everything must be adjusted vertically. People like to play what they engrave (sometimes it is very reasonable), but for me this is A BIG HANDICAP of those programs.

[jrethorst]

As info, I asked the author about WinScore -- whether it was simply a port to Windows, or whether functionality had been changed or added. He said:

WinScore is, in my opinion, not suitable for production work -- it is filled with bugs, missing functionality, and generally a disaster -- though technically it is the "latest version" of the software. The general symbols and mathematics of the program (spacing, etc.) did not change, though, so it really only matters to those working in the program. I still use DOS SCORE for all of my work.

[DatOrganistTho]

SCORE allows absolutely unparalleled, complete control over every item on the musical page.

Yes yes..

you can manipulate all 19 of SCORE's slur parameters

What to say more?

When I see a slur colliding with an accidental, I can just adjust exactly the part of the slur that needs to be changed

Finale? Sibelius?

The other part of SCORE... is that nothing changes unless you want it to.

What to add more?
No fonts. There is a very little chance that your output will be a messy PDF or collide with printer. Just outlines. And for all outlines you can adjust their size.
IMO, the biggest problem of F and S is that they are sequencers, and everything must be adjusted vertically. People like to play what they engrave (sometimes it is very reasonable), but for me this is A BIG HANDICAP of those programs.

Couldn't agree more.

[DatOrganistTho]

As info, I asked the author about WinScore -- whether it was simply a port to Windows, or whether functionality had been changed or added. He said:

WinScore is, in my opinion, not suitable for production work -- it is filled with bugs, missing functionality, and generally a disaster -- though technically it is the "latest version" of the software. The general symbols and mathematics of the program (spacing, etc.) did not change, though, so it really only matters to those working in the program. I still use DOS SCORE for all of my work.

I've been trying so hard to buy this program outright. I can't get the people over at the website to pickup the phone or answer an email. I would love to use DOS SCORE with DOS-BOX.