Multiple accidentals problem

General notation questions, including advanced notation, formatting, etc., go here.

Moderators: Peter Thomsen, miker

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Sat May 23, 2020 8:11 pm

Please see attached, 1 page of a long piece, & there are hidden staves. The visible staves have been transposed for French horn in F. I downloaded this file as mus or xml -- can't remember, then transposed the horn parts (from Eb).

There are many cases of triple & quadruple accidentals -- I don't know why they appeared -- which I would like to batch simplify.

I know about Utilities > Respell Notes.

See m31 -- that visible Fb was originally written as bbbbA, but it sounded correctly as Ab, and it still does, but Respell apparently believed the 4 flats, even though Playback (correctly) does not. I left the m31 Hn2 bbbbF alone, for comparison -- it sounds (correctly) as Fb, but if Respell gets it, it changes to a written D, again, leaving playback correct. The other cases of quadruple flats appear to play correctly as a single flat; the triple flats play (correctly) as naturals.

Is there any hope for this?

Thanks,
David
Attachments
Krommer op71 pg1.musx
(166.41 KiB) Downloaded 89 times


User avatar
miker
Posts: 5993
Joined: Fri Mar 13, 2009 4:28 pm
Finale Version: Finale 27.4
Operating System: Mac

Post by miker » Sat May 23, 2020 9:25 pm

Take a look at JW Change Pitch. http://www.finaletips.nu/index.php/download

Make sure you download the correct version for your platform and Finale version.

Jari makes these incredible plugins at no charge. But he accepts (and deserves) contributions.
Finale 27 | SmartScorePro 64
Mac OS 13.2.1 Ventura
Copyist for Barbershop Harmony Society

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sat May 23, 2020 9:44 pm

It is indeed a real mess. You've got the wrong transposition set for F horn, which is part of the problem. It's not a transposition that exists in the diatonic system--you want "Up P5, add 1 sharp," which is built into Finale.

How was it created? That might be a clue to fixing it. If you set up that wrong custom transposition and then edit it or respelled it, you might be better off starting over with the correct transposition. You could use the Repitch tool so you'd only have to enter pitches.

User avatar
zuill
Posts: 4418
Joined: Sat Dec 10, 2016 9:35 pm
Finale Version: Finale 2011-v26.3.1
Operating System: Windows

Post by zuill » Sat May 23, 2020 11:27 pm

Having Independent Key Signatures for the 2 Horn Staves also adds a further level of trouble. Not needed, I believe.

Zuill
Windows 10, Finale 2011-v26.3.1
"When all is said and done, more is said than done."

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sat May 23, 2020 11:34 pm

I didn't notice that. Right--the F transposition will take care of the right key signature.

User avatar
zuill
Posts: 4418
Joined: Sat Dec 10, 2016 9:35 pm
Finale Version: Finale 2011-v26.3.1
Operating System: Windows

Post by zuill » Sat May 23, 2020 11:43 pm

The original is written for Eb horn, and no key signature. The opening written notes are C an E. So, for F Horn, Bb and D. However, somehow the notes here are all multiple flats. So, I did retranscribe and got them looking correct.

How did this get to be such a mess?

Zuill
Windows 10, Finale 2011-v26.3.1
"When all is said and done, more is said than done."

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sun May 24, 2020 4:45 pm

I think the easiest way to do the Eb to F conversion is to set up the staff for Eb transposition, enter the notes, then change to F transposition.

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Sun May 24, 2020 7:11 pm

Strange -- I responded yesterday to your help, but I don't see it here.

The original download (pdf) had the horn parts in C, and marked to be transposed to Eb. We horn players are used to this, and transposing to any other keys, as well. I wanted parts to be in F though so users won't need to transpose. Only yesterday I noticed that Finale does have Horn in Eb trans. That simplifies things. I passed the pdf through SmartScore and when Finale played it, the Bb clarinets & Eb horns were off key, as expected. I (thought that I) had to give those staves key independence so I could manipulate them to their proper keys in both pitch and appearance, and that's where the mess originated. I remember having a nightmare getting pitch right or appearance right, but I was stymied how to get both. Ideally, I should not have had to make the transposing instruments key-independent. Not seeing the existence of Horn in Eb trans, I went to Score Manager > Transpose > Other once the pitch was right, to tweak the appearance to coordinate -- I think that's when the flats proliferated. Somewhere, maybe in a different thread -- I can't find it right now -- motet introduced Respell Notes to me and that cured the flats. The ultimate purpose of all this, BTW, is to send pdf sheets to players of any instruments, along with music-minus-one mp3 files, so that they can play octets & other pieces while in isolation. This is the first time it blew up on me. I much appreciate all of your help!

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sun May 24, 2020 8:43 pm

It may be possible in SmartScore to correct the transposition for the horns and clarinets before making the MusicXML file. Maybe Mike Rosen will know.

If not, I think the way to do in in would be to import the horn part into Finale into a normal, non-transposing staff. Since what looks like a C is in fact an Eb, then transpose the notes down a major sixth with Utilities/Transpose. Finally, go to the Score Manager and pick F transposition and it should be correct, with the correct key signature.

For Bb clarinet, same thing, except transpose down a major second.

User avatar
miker
Posts: 5993
Joined: Fri Mar 13, 2009 4:28 pm
Finale Version: Finale 27.4
Operating System: Mac

Post by miker » Sun May 24, 2020 11:04 pm

I’m not 100% sure the problem is all with SmartScore. Some of it may lie in the XML translation. I get situations where parts that look fine in SS are subject to weird transpositions coming into Finale. It’s especially bad if there are multiple key changes in the piece, with staves that may he hidden during the change. I just sort of cuss, and spend some time with the key sigs, and moving up or down with the 6 and 7 keys. Transpositions are way over my head. I sort of understand them, but I don’t really internalize how they work.
Finale 27 | SmartScorePro 64
Mac OS 13.2.1 Ventura
Copyist for Barbershop Harmony Society

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Mon May 25, 2020 8:09 pm

I think I got it now, and largely per motet's recipe. But 5+ hours of aggravation today led to the discovery that the culprit was that the SmartScore setting for Unify Key Sigs was not based on topmost Staffline. With that setting, the notes I see in Finale are really what they look like they are, and transposition puts them where they need to be, for any instrumentation. miker, thanks for mentioning key 6 & 7 in passing -- I didn't know that -- neat time saver.

I think I had the SS Unify setting accidentally on 1st measure, and when Finale got the score I had bizarre errors, so my time went into fixing them whackamole style.

Surely I can't be the first person who got bewildered how to make a mixed-instrument score play back correctly, where each staff of the score shows what each musician sees. But I couldn't find prior cases in the forums. It would be a wonderful addition to Finale's magic bag for a command like: Match playback to appearance. I.e., I name the instrument for each staff, invoke this tool, and the entire score is updated.

User avatar
Michel R E
Posts: 702
Joined: Sat Dec 10, 2016 9:16 pm
Finale Version: Finale 2012, 25, 26
Operating System: Windows

Post by Michel R E » Mon May 25, 2020 8:24 pm

scanning a score is always going to have SOME sort of difficulty.
User of Finale since version 3.0 on Windows.
Now using a mix of Finale 2012, Finale 25, and 26.1
GPO, Garritan Solo Stradivari violin, Gofriller Solo Cello.
XSamples Chamber Ensemble.
Absolute convert to NotePerformer3.

User avatar
miker
Posts: 5993
Joined: Fri Mar 13, 2009 4:28 pm
Finale Version: Finale 27.4
Operating System: Mac

Post by miker » Mon May 25, 2020 9:29 pm

Michel R E wrote:
Mon May 25, 2020 8:24 pm
scanning a score is always going to have SOME sort of difficulty.
Amen to that. But (for me, at least) it’s still the fastest way to get a score into Finale.

But I’m going to take a look at my SmartScore settings. That Unify Key Sig might be my problem, too. Thanks for pointing it out.
Finale 27 | SmartScorePro 64
Mac OS 13.2.1 Ventura
Copyist for Barbershop Harmony Society

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Mon May 25, 2020 10:42 pm

There's still some wonky, and I think it's on Finale's side. See attached tif (what I gave to SmartScore many times) and jpg (screenshot of the xml it produced with Unify Key Sig completely turned off). All the tif key sigs are correctly retained.

1) Staves 5 & 6, the Eb horns -- the m1 notes (and the rest of the notes) changed from c5,e4 to A4,C4. See the xml screenshot -- it is the xml that SS passed to Finale, fed back into SS; the horn notes in SS match the tif.

2) Only the horn staves, per Score Manager, are marked "Hide key signature and show all accidentals".

3) When I select Clarinet in Bb for staves 3 & 4, the single flat becomes a # (correct). If Score manager Trans is used to add 2 flats, it goes back to KS 3 flats, not the necessary 1# - 2 flats = 1 flat.

This is the kind of thing I am battling for a few days, and miker, I think what you mentioned happens to you. I will try some more like this, with SmartScore set for Unify per staff#1, and try not to come back here, except to report that that solves all the problems.
Attachments
Krommer 71 xml.JPG
Krommer 71 xml.JPG (75.86 KiB) Viewed 5303 times

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Tue May 26, 2020 1:27 am

I found the score on IMSLP and played around with SmartScore. You definitely don't want Unify Key.

The program crashed when I tried to recognize the whole file, but I got it to work on the first seven pages.

With much horsing around I was able to get it horns and clarinets to work! SmartScore neglects to output the <transpose> XML tag, at least the way I was using it. When I inserted those manually with a text editor, all was well. There may be some switch in SmartScore I'm missing, but all I found under "transpose" in the documentation had to do with playback in SmartScore.

I created a Finale file and found that for Bb clarinet you want the tags

<transpose>
<diatonic>-1</diatonic>
<chromatic>-2</chromatic>
</transpose>

and for Eb horn you want

<transpose>
<diatonic>-5</diatonic>
<chromatic>-9</chromatic>
</transpose>

When I inserted them in the Krommer file and imported into Finale, everything was correct. (To get F horn you could then change the transposition in Finale from Eb to F).

This is all very arcane, though, and I don"t expect you to be able to do that. I would get a hold of SmartScore customer support and ask them about this; you have my permission to email them this posting. It's very possible we just neglected to turn on some feature.

User avatar
miker
Posts: 5993
Joined: Fri Mar 13, 2009 4:28 pm
Finale Version: Finale 27.4
Operating System: Mac

Post by miker » Tue May 26, 2020 4:45 am

motet wrote:
Tue May 26, 2020 1:27 am
I would get a hold of SmartScore customer support and ask them about this; you have my permission to email them this posting. It's very possible we just neglected to turn on some feature.
Yes, please! In fact, link them to the entire thread.

They are promising the new version in the next few weeks. Unfortunately, they have been promising that for a very long time.
Finale 27 | SmartScorePro 64
Mac OS 13.2.1 Ventura
Copyist for Barbershop Harmony Society

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Tue May 26, 2020 4:11 pm

Last night I made up the attached test file -- I wrote the same C-scale 9 times, then transposed to various different-key instruments as I know are listed in Score manager, with a couple duplications. Then I made a pdf, imported to SmartScore, then sent SS's xml back to Finale for curing. The point being to check for consistency of curing procedures, to get practice, and to figure out how much time is expended getting a correct-playback score. motet, you are reporting that a long file presents more problems -- I suppose 10 or 20 pages of this tune could check for that also.

Is it the case that if SS is set for No Unify, that it will (theoretically?) create an xml that exactly matches the pdf or tif?

I'm not sure if your suggestion to contact SS customer support is for me or if you are discussing it amongst yourselves. As far as getting the translation job done, would it be best to contact say, JW to make a magic plug-in, or implore the Finale programmers to make v26.3 take care of this with one click? Or at most, one transaction per staff -- select the instrument, done? This would be a great boon to Finale.

I'm happy to contact SS, but I come in with a lot less knowledge than any of you.

Since I passed out at the computer last night while trying to cure this test doc, it will be today's project.

Edit: I don't see my trans test.tif or pdf attached, so now I attach the original musx version.
Attachments
trans test.musx
(104.13 KiB) Downloaded 68 times

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Tue May 26, 2020 4:32 pm

My guess is you won't get any satisfaction from JW or MakeMusic any time soon.

One more thing to try is to set the playback transposition in SS and see if that results in the correct tags being set in the XML file.

I am recovering from a bad back so can't do it today, but I will scan the file for you, edit the XML file to include those tags, and make a .musx file. Since I bought SS a while back and haven't used it, it's a learning opportunity for me. Plus it will save you from passing out at the computer.

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Tue May 26, 2020 5:38 pm

Thanks for all this work and interest, motet & miker.

In the past few minutes, I tried an experiment with my scale test-file:

1) import to SS, Unify per 1st staff -- although I think further experiment will show that this setting doesn't matter

2) in SS, erase all key signatures -- quick & easy

3) Pass SS xml back to Finale

4) With the original pdf or tif visible, select appropriate instrument for each staff -- I notice that the notes in all of the staves jump around. Don't care.

5) Start over at staff#1 with the keyboard 6 & 7 trans shortcuts to match the notes with the pdf

Done.

Except that the three Eb instruments turn up the same error on the concert A#, having received sharp-doublesharp, and sounding a half-tone high. Except for this one error, this process is relatively quick, and most importantly (for me), mechanical -- no thinking about transposition theory.

If this is as bad as it gets, I can at least go forward with some projects for now.

Is there some software other than SS that can convert an image to xml? That might be useful to determine if SS is complicit.

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Tue May 26, 2020 5:45 pm

Please try the experiment of setting the right transposition in SS playback to see if that fixes it before writing them off. I'm sure they must have encountered this situation before.

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Tue May 26, 2020 6:33 pm

I think I misstated the process -- here is an almost perfect mechanical sequence. There are a few errors that I can't trace:

1) in SS, Unify to 1st staff -- I think it doesn't matter what key results.
2) Pass the SS xml to Finale
3) in Finale, use the Key Signature Tool to transpose all staves to C
4) Select instrument for each staff normally -- each staff will adopt new KS and transpose notes
5) Use 6 & 7 keyboard trans to match notes (1st note?) on all staves with pdf.

I don't know if the few errors are theoretical or programming bugs.

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Tue May 26, 2020 6:38 pm

And that doesn't break any accidentals?

If SS would put in the transpose tags, none of that would be necessary. Aren't you curious to see if setting the SS playback transposition accomplishes that? When I put the transpose tags in by hand, it was almost perfect with none of that fidding.

montedoro44
Posts: 144
Joined: Sat Feb 29, 2020 4:21 pm
Finale Version: 27
Operating System: Windows

Post by montedoro44 » Tue May 26, 2020 10:32 pm

motet, I'm not sure I know enough to deal with SS customer service, at least about transpose tags -- though I do recognize them as html tags, which I know how to use for text manipulation. I'm happy to try though.

The core of the problem is that the pdf/tif image of the score designates the instrument for each staff, but neither SS nor Finale can read those words so as to implement them. Short of that, isn't it the case that the user will still have to visit each staff once to select its instrument, either in SS or Finale? -- so that would be the minimum amount of work for the user.

If that is true, then I just stumbled on a near-perfect Finale solution -- ignore my previous hasty & flawed remedies. :(

1) pdf to SS: No Unify; xml to Finale (key sigs & notes are identical to the pdf)

2) Finale: Plug-ins > Scoring & Arranging > Global Staff Attributes > Ignore Key Signatures
[except for already-key-of-C-staves ?! -- see below]

3) Score Manager > Select instruments

I was not conscious of this "Ignore Key Signatures" tool -- it claims to transpose all keys to C. After it does that, visiting each staff to make the instrument choice produces a perfect copy of the pdf, and playback is correct - - -

except it seems to not work for the last staff (oboe) on my Trans Test doc. So I started over at the import xml step (if I remember right, using Undo to precede the plug-in didn't work). At Global staff Attributes I changed only all of the not-key-of-C staves, and then the page & playback came out perfect. Does this sound like a bug in the Ignore Key Signatures tool that Finale ought to fix? This is a very powerful tool, otherwise.

User avatar
motet
Posts: 8225
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Tue May 26, 2020 11:04 pm

I set the transposition in SmartScore and it works perfectly! Don't use "Unify key signatures" of either flavor. Set the transpositions in SmartScore's mixer as follows, -2 for Bb clarinet and -9 for Eb horn (the number is the number of half steps the note should sound from what's written):
0467.png
0467.png (19.76 KiB) Viewed 5189 times
Save as XML and you won't have to do any tweaking in Finale except change the instrument names, playback sound if you care about that, and change Eb horn to F horn if you want (use the Transposition dialog in the Score Manager).

So both Finale and SmartScore are absolved, except perhaps for minimal documentation from the latter. It seems to have rendered the notes almost perfectly. It didn't do so well on dynamics or other text, so you'll have to enter those yourself.
Attachments
0466.png
0466.png (168 KiB) Viewed 5210 times
Last edited by motet on Wed May 27, 2020 4:54 am, edited 1 time in total.

User avatar
miker
Posts: 5993
Joined: Fri Mar 13, 2009 4:28 pm
Finale Version: Finale 27.4
Operating System: Mac

Post by miker » Tue May 26, 2020 11:32 pm

Excellent detective work. Thanks to both of you!
Finale 27 | SmartScorePro 64
Mac OS 13.2.1 Ventura
Copyist for Barbershop Harmony Society

Post Reply