Page 1 of 5

"Ten Music Notation Programs", including Dorico

Posted: 24 Oct 2016, 18:46
by jrethorst
I've revised the Seven Music Notation Programs announced here earlier, adding Dorico, Encore (still more popular in Europe and especially South America; turns out it's a program of choice in Brazil) and Graphire Music Press, which is once again available for sale and which had some discussion in this forum a little while ago. The comparison of ten programs' engraving is at:

https://dl.dropboxusercontent.com/u/643 ... ograms.pdf

Unsolicited comments on earlier editions of this document include:

"Absolutely amazing! . . . Interesting! . . . This is certainly a worth while endeavor . . . a really great idea and I’m glad that someone is taking the time to do it the right way . . . I love seeing how each program handles music . . . I don't think I've ever seen a lineup of professionally engraved scores like this before. It's a great idea. I must say, I find it very encouraging that there is still a large element of personality to be found in every one of the engravings, despite the nature of computers. Long may that continue! . . . Very interesting indeed to see the subtle differences in the notation and how the eye perceives it for clarity."

Re: "Ten Music Notation Programs", including Dorico

Posted: 24 Oct 2016, 20:24
by jrethorst
When I posted the earler Seven Programs in the Delcamp classical guitar forum [http://www.classicalguitardelcamp.com/ ... ?t=104735], I got several responses including this detailed one, which I take the liberty of reproducing here as it may be of interest to the many sharp eyes of this forum:
My observations:

1. Finale: Looks like music is supposed to look. Any errors are the fault of the person who engraved this score (especially the Coda). Has a "modern" look that while initially reviled 30 years ago, now is commonplace and even wins awards. Alfred, Hal Leonard, etc. use or have used it. Some elements, like the ledger lines, are too thick, and one rest in the 2nd ending is on a line instead of a space where it belongs (fault of the engraver). All those issues are adjustable.

2. Lilypond: Not bad. It's an "older" style like older Schirmer scores (nothing wrong with that). The "rolled chord" wavy lines are a little angular, which is less common, but not "wrong" per se. The Fermata in 25 is touching the beam but that could be the notator's error. This is very "traditional" and can't really be faulted. I don't care for the numerals in the endings, but I do like the "thickness" or "boldness" of a lot of the text and symbolic elements - again, very traditional (because once engraved, Schirmer's plates were used over and over and over and over again, gradually wearing down and becoming "blobby"!). I used to customize Finale to mimic this kind of "boldness" of the fonts/symbols, but with the modern stylings it had.

Note, some programs do not convert to PDF well, and that's an issue in this day and age. I'm not sure how these were generated and put into this PDF, but as an example, the Lilypond PDF has ledger lines that constantly vary in thickness and some even have "rounded ends". These kinds of imaging errors are often only present at certain zoom levels. But when the Finale score gets down to closer to 100%, it's ledger lines look think like they're supposed to, and Lilypond's are not only still thick, but "blurry" because of it.

Rest in m. 25 is in correct place.

3. Musescore. A LOT of the 8th rests are on the line, which is wrong. This tells me that it's possible the program defaults to this. If it can't be over-ridden by the user, it's no good. It could engraver error, but with this many in the wrong place, looks more like the way the program handles it. The 8th rests in general are a little "blobby" compared to the previous two. The dotted quarter rests also have dots that are on the line. That's no good. However, at 100% zoom, the "darkness" or "boldness" that results makes for easier (or more traditional :-) reading from a stand. Where Finale is "crisp" and "clean", the next two are rather "bolder" in general - the LP score is also reduced even further in size so not a 1:1 comparison here. Just a general observation.

The only other questionable thing I see is the "segno" sign in the D.C. indication - it's very small and if it can't be adjusted independently of the rest of the text, that could be a dealbreaker.

4. Notion. Notion's output in the PDF looks very bad at most zoom levels - it's almost as if it's only 200 dpi instead of 600+ on a laser printer. Very "blurry". It also uses some rather non-traditional looking symbols.

The 8th rests are particularly bad. They're also all on the 3rd space leading me to believe this is the default. There are some collisions because of this (unacceptable ones). In the old days, even Finale would do this and you'd have to manually adjust, but nowadays they avoid most such collisions on input. Maybe the notator here didn't know how to move them, but if they're unmovable, that's a dealbreaker.

Downstemmed notes with dots should have the dot in the space BELOW when on a line, even the ledger line notes. This would be another dealbreaker if you can't change it.

5. Overture: Very reminiscent of the early days of Music Notation Software where I suppose programmers thought people would be so delighted to have such programs that it didn't matter if the music didn't look "traditional" and they figured people would accept the "new look". That never happened :-). All the 8th rests are wrong. The ties all have a horizontal grey line with them, that may be an artifact from the conversion to PDF. Would be unacceptable in direct conversion and especially printing. The ties also are a bit too arced and far from the notes - tends to make them look a but more like slurs in this context (becuase ties and slurs are essentially the same type of symbol). Running back through the previous versions to check the ties, again Finale handles them nicely (and was a huge upgrade for them in the past) and the rest all look pretty good actually.

The important part is whether you can globally and individually adjust ties - set default parameters and/or adjust them once entered. These don't look like you can, or else someone would have fixed them!

The text at the top (attributed to) is a little sketchy - the 2nd line is too close to the line above - but that could have been engraver error. Still if the program doesn't allow adjustment of these elements.

6. Score. Wow, not bad. Not unlike Lilypond. You're really nitpicking at this level. The fermatas seem small to me and the augmentation dots, and especially the dots on the repeat bars seem "light" to me. Otherwise it's a very traditional output.

I'm not crazy about the G Clef (and small fermata and dots) but otherwise, this may be the best "average" of all of them so far. The "rolled chord" is curvy, but not so "modern" like Finale, and not so angular like LP.

If you look closely, the Half Notes can be a deciding factor - each program deals with them a different way. In some cases they are merely non-blackened quarters (or quarters are filled in halves if you like) and in other cases a completely different symbol. I wish there were some WHOLE NOTES in this piece because that's where things get really noticeably different. Whole notes have a completely different notehead than the rest traditionally (no, I'm not talking about semi-breves and breves).

Of all of these, Musescore's Half notes probably look the best! Otherwise, the nod goes to Finale - so far.

7. Sibelius looks great, and not unlike Finale, as expected given Sibelius kind of came along with the intent of out Finale-ing Finale! There are some font things I don't like, but again, most programs allow you to customize that stuff. This is the only one with the "8" below the Treble clef (which is not officially necessary as we know guitar is a transposing instrument, but I suppose for the sake of non-guitarists it's nice to have). Finale can certainly do this too (easily) so it's kind of a strange choice as it makes it look like it's something none of the other programs could do.

I don't like the font on it here, and I know you can't change that becuase it's embedded in the symbol. IMHO the Finale "octave down treble" looks better.

FWIW, I don't like any G clef that looks too "pointy" at the top (LP) or "leans" (score). I do like the rolled chord lines better than Finale (which I know you can't change in Finale) and the Segno sign (can change size of in Finale).

Again, I wish they all used the "target" coda symbol because again it's an unfair comparison - it makes it appear that those without it don't have it. The ties are slightly funky - like Finale's was before they fixed them - but again as long as you can go in and edit the start and end points and curvature, it's OK.

There's one thing that absolutely keeps me from buying Sibelius: You can't resize the little floating keypad. That's absolutely ridiculous. I mean, really? Come on.

Also, having used Sibelius this last semester, both myself and students who had used Finale before, as well as those who've never even used a notation program before, found the way that Sibelius deals with selecting notes really counter-intuitive (I'd have to say that about the program as a rule, but Finale certainly has it's stupid things!). No one likes the way it deals with a note that was played in, and selecting that note or the previous note using the arrow keys - it wants to change the selected note when it's highlighted only under certain circumstances (i.e. selecting it rather than having just entered it).

I haven't checked if that's something you can change (because Sib has an entire set of parameters you can change to make it work like Fin ;-) but boy is it annoying.

After all this, you have to consider not only the output, but the input method. While Lilypond and Score are both acceptable outputs, and may in fact be customizeable beyond what we see here.

Score runs on DOS. They might as well include a non-resizable keypad palette!
Lilypond is text input. You essentially have to be a computer programmer to use it (in addition to understanding music notation elements!)

This explains why FInale and Sibelius tend to be the preferred "modern" programs, running on both Mac and Windows, using MIDI input and output (Finale now includes the ability to have audio tracks), working like other DAW programs, multiple (proprietary and non-proprietery) output formats and so on.

These 4 are the only ones whose printed output I would consider "publisher quality", unless Musescore can be tweaked. But assuming these were done with basically the "default" settings and simple little edits where easy to get to, the further you get from these 4 the more work you'd have to do to get things looking right.

However, there seems to be an entire contingent of people using even Finale and Sibelius who don't care to have the music look right, so there you go.

Re: "Ten Music Notation Programs", including Dorico

Posted: 24 Oct 2016, 20:35
by tisimst
Thanks, John, for sharing those observations.

Re: "Ten Music Notation Programs", including Dorico

Posted: 25 Oct 2016, 02:39
by OCTO
One thing might be very considerable here: why not to create documents with the same fonts? I dislike Finale with Engraver (at this size, perhaps 22 pt is better), and like MuseScore 2 output. I also dislike Opus font...
So my judging is very much based on that.

Re: "Ten Music Notation Programs", including Dorico

Posted: 25 Oct 2016, 11:56
by Schneider
I fully agree OCTO

Re: "Ten Music Notation Programs", including Dorico

Posted: 25 Oct 2016, 12:18
by Fred G. Unn
I know its shortcomings have been discussed elsewhere, but wow, in comparison it is clear Dorico is not close to being able to be used for professional work yet. There are too many essential elements of notation that just aren't possible with it.
OCTO wrote:One thing might be very considerable here: why not to create documents with the same fonts? I dislike Finale with Engraver (at this size, perhaps 22 pt is better), and like MuseScore 2 output. I also dislike Opus font...
So my judging is very much based on that.
I agree, the choice of font can be very influential on my overall opinion of the work. Here's a version of this I did last spring with my defaults, and another version of the same file with the Finale defaults. I think there is also a big difference in what software does by default and what "can" be done with it. For example, I didn't run Patterson Beams on the Finale default version, but did so on my version. A lot of people never bother to run it, although it produces much improved results IMO. Varying the choice of text font also obviously can have a strong influence as well. Taking my version and changing the text font to Adobe Jenson creates a very different feel to the overall look too.

(I know everyone hates the winged repeats, but most of what I do is for immediate live performance/recording, and I think they help the performer. I might use standard repeats for publication. I also did away with the archaic "DC al [sign] e poi la Coda" in favor of the more familiar D.C. al Coda. It is unfamiliar to most performers and even the Notion copyist misinterpreted it. )

Re: "Ten Music Notation Programs", including Dorico

Posted: 25 Oct 2016, 12:26
by tisimst
Thanks for sharing those, Fred!

Re: "Ten Music Notation Programs", including Dorico

Posted: 27 Oct 2016, 04:58
by jrethorst
OCTO wrote:One thing might be very considerable here: why not to create documents with the same fonts? I dislike Finale with Engraver (at this size, perhaps 22 pt is better), and like MuseScore 2 output. I also dislike Opus font...
So my judging is very much based on that.
That would be hard to do. Not many of these programs support SMUFL yet, so making a font map for each program would take a lot of time. Also, not all programs (MuseScore at least) support different fonts. In any case, part of the point of the comparison was to illustrate differences between programs using more or less their defaults, including fonts.

Re: "Ten Music Notation Programs", including Dorico

Posted: 27 Oct 2016, 06:40
by OCTO
jrethorst wrote:
OCTO wrote:One thing might be very considerable here: why not to create documents with the same fonts? I dislike Finale with Engraver (at this size, perhaps 22 pt is better), and like MuseScore 2 output. I also dislike Opus font...
So my judging is very much based on that.
That would be hard to do. Not many of these programs support SMUFL yet, so making a font map for each program would take a lot of time. Also, not all programs (MuseScore at least) support different fonts. In any case, part of the point of the comparison was to illustrate differences between programs using more or less their defaults, including fonts.
Yes, actually that makes a perfect sense.

Re: "Ten Music Notation Programs", including Dorico

Posted: 31 Dec 2016, 00:53
by mducharme
jrethorst wrote:I've revised the Seven Music Notation Programs announced here earlier, adding Dorico, Encore (still more popular in Europe and especially South America; turns out it's a program of choice in Brazil) and Graphire Music Press, which is once again available for sale and which had some discussion in this forum a little while ago. The comparison of ten programs' engraving is at:

https://dl.dropboxusercontent.com/u/643 ... ograms.pdf
Could you add NoteAbility to this comparison? It is free software:

http://debussy.music.ubc.ca/NoteAbility/index.html