tablature totally broken in V26

[jzucker]

I wrote this simple exercise, generated some lines in treble clef, copied the treble line down to the tablature line and this is what it produced. It looks like everywhere this is a chromatic passage, the tablature is reversed from low to high.

[jzucker]

I have asked for a refund. This is totally unacceptable behavior. Do they not do *ANY* unit testing???

[zuill]

If you uncheck the option to respect instrument ranges in Preferences, you should get proper behavior. Attached is what I get.

Zuill

[jzucker]

well, it's too late. I exported the musicxml file, uninstalled v26, installed v25 and it works as it always did. Instrument range has ZERO to do with this issue because all notes are perfectly within the instrument range. If the range *WAS* off, I would expect octave transposition, not chromatic interpolation.

[zuill]

I'm sorry that v26 didn't work out for you. I have been cautious with v26, as I know the new articulation paradigm can cause issues if there is a need to open a v26 file in v25 or 2014.5. It messes things up. However, for new work, there is actually some benefit to the new articulation paradigm.

Zuill

[motet]

I'm confused--is the instrument-range thing the culprit? Why would that reverse the fret numbers?

[jzucker]

I'm confused--is the instrument-range thing the culprit? Why would that reverse the fret numbers?

agreed but more concerning is that this isn't 26.0, it's 26.2. That means 3 maintenance releases of this version and an obvious bug of this magnitude is still present? Don't they do unit testing??? This flies in face of the entire selling point for the supposed refactoring/rewrite of the software they did a few years ago. Despite all the promises, I have seen ZERO evidence of increased quality, and little increase in usability. I went back to version 25 and the problem goes away.

It also opened the door for me to start experimenting with alternatives. I am going to spend some time learning musescore. That may be the way forward for what I do. I realize it may not be the answer for large scores...

[zuill]

I want to confirm how to get exactly the result that the original file displayed:

1. Choose to respect instrument ranges in preferences.
2. When copying to the TAB staff, choose 4 as the lowest fret.

If I turn off the instrument ranges option and choose 0 as the lowest fret, then all is as it should be. I believe v25 would do the same thing, so it's not a v26 issue. v26 is just fine.

Zuill

P.S.: I just confirmed this by opening the v26 file in v25 and got exactly the same erroneous result, so v26 is not the culprit. I even created a new document with Guitar and Guitar TAB in v25, copied the top staff there, and then copied to the TAB staff. With the 2 settings mentioned, the same bad result occurred. So, v25 and v26 behave exactly the same in this regard. v26 is not the issue. Often, we forget settings we have made in a prior version, then wonder why a new version doesn't behave the same way. I am guilty of this. When I installed everything again in my new Windows 10 machine, I had to spend time recalling my prior settings.

[jzucker]

sorry bro, a bug is a bug. And if preferences change on a new version, it should copy the settings from the old version and aside from that, reversing the notes in a chromatic passage has *NOTHING* to do with range of the instrument. It's a frickin' bug. If we continue to allow the dog to piddle on the floor, it never learns.

As an aside, I have respect instrument ranges when copying or changing instruments checked in my V25 preferences so your theory doesn't hold true.

[motet]

The bug exists in Finale 2014.5 as well. The culprit indeed seems to be "respect instrument ranges when copying" and gives an erroneous result whether lowest fret is 0 or 4, so I'm surprised it works for jzucker in Finale 25. My guess is this bug was introduced when this feature was added (Finale 2012? I didn't get that one). I wonder if anyone has ever noticed or reported this bug before.

I've posted elsewhere about other bizarre things the feature does (see https://forum.makemusic.com/default.aspx?f=6&m=465376). Since it's aimed at the novice musician it seems like it should be turned off by default.

[zuill]

Why am I not getting the bug?

Zuill

[jzucker]

don't know but I'm guessing that the checkbox is only tangentially related to the issue since in V25 it works fine for me. There's probably an uninitialized variable somewhere that is randomly initialized by memory and may be different on different versions of windows or different computers. My preferences have never changed on that setting. I never even knew about it and I've published literally 600 pages of music with notation and tab and never had the problem until the V26 upgrade.

[motet]

Your bio doesn't say whether you're using a Mac or Windows. Perhaps therein lies the difference.

[jzucker]

Your bio doesn't say whether you're using a Mac or Windows. Perhaps therein lies the difference.

using windows 10 64 bit

[motet]

As am I and, I believe, Zuill. It's a mystery, then.

Anyhow, a totally wacky bug. "Respect instrument ranges" may cause the octave to change (often stupidly, see the link above), but I don't know why the fret numbers would run backwards. Been around since 2014.5, as I say.

[jzucker]

As am I and, I believe, Zuill. It's a mystery, then.

Anyhow, a totally wacky bug. "Respect instrument ranges" may cause the octave to change (often stupidly, see the link above), but I don't know why the fret numbers would run backwards. Been around since 2014.5, as I say.

The interesting thing is that I have not changed preferences since I wrote my first sheets of sound for guitar book which was in 2003 or 2004. My next book was written using 2014.5 and the last two using 25. I have never varied that setting and have been doing the same tablature workflow for 15+ years with no issues. Certainly never got wrong notes. Interestingly enough, when you see the bug, you can click on the staff in the speedy edit tool and it shows you a treble staff instead of tablature and in the treble staff, the notes are correct but as soon as you click off the staff and it re-generates the tablature, it goes back to the incorrect interpolation.

Again, all software has bugs but as an engineer, I can tell you that you write unit tests for these sorts of things. Additionally, I thought it was just a V26 bug and I was flabbergasted that they were on 26.2 (3 releases) and the bug is still here but now you're telling me it goes all the way back to 2014.5???

[motet]

And it's by no means the oldest of the bugs! Or the most disastrous.

When you upgrade, by the way, Finale doesn't try to replicate your previous settings, but rather gives you a clean install every time. The brain-dead "respect ranges" feature appeared Either in Finale 2012 or Finale 2014 and was turned on by default (and still is, apparently).

[Bill Stevens]

>> don't know why the fret numbers would run backwards

They are not really backwards, are they? It's not clear on the sample, but the fret numbers all have a minus before them because the tab is off by an octave, for whatever reason, possibly related to the music range thing. So for instance, in the yellow area E down 5 frets would be B and down 6 frets would be B-flat.

Bill

[jzucker]

>> don't know why the fret numbers would run backwards

They are not really backwards, are they? It's not clear on the sample, but the fret numbers all have a minus before them because the tab is off by an octave, for whatever reason, possibly related to the music range thing. So for instance, in the yellow area E down 5 frets would be B and down 6 frets would be B-flat.

Bill

yes you're right and I was just focused on the chromatics because they were so obviously going in the right direction but the whole score is off.

[zuill]

I've just spent an hour exploring TAB in various versions of Finale. When I create a clean file in v26, no issues. There is some corruption in this file, but I can't say what. I exported to XML, and Finale recognized several errors when trying to open the XML file. It finally opened, and the odd TAB behavior was gone. So, I think the only issue here was file corruption.

One thing that gave me an indication of file corruption was odd behavior when selecting one or more measures. When I copied some of the measures by themselves, no problem. When I copied bars 1-3, misbehavior. When I selected 1-4, proper behavior. This was just the tip of the iceberg. Needless to say, it is a file error issue, since a new file in v26 does not have this problem.

Zuill

[motet]

Bill Stevens solves the "respect ranges" thing--thanks. I didn't notice the minus signs--what a concept!

[jzucker]

I've just spent an hour exploring TAB in various versions of Finale. When I create a clean file in v26, no issues. There is some corruption in this file, but I can't say what. I exported to XML, and Finale recognized several errors when trying to open the XML file. It finally opened, and the odd TAB behavior was gone. So, I think the only issue here was file corruption.

One thing that gave me an indication of file corruption was odd behavior when selecting one or more measures. When I copied some of the measures by themselves, no problem. When I copied bars 1-3, misbehavior. When I selected 1-4, proper behavior. This was just the tip of the iceberg. Needless to say, it is a file error issue, since a new file in v26 does not have this problem.

Zuill

Zuill, this file was generated cleanly using Finale 26. Additionally, i exported it as xml and imported it into musescore and finale V25 with ZERO issues in the tablature.

[motet]

Zuill, with the misbehaving partial copy, was "respect instrument ranges" selected? Finale will "consider" (if that word can be applied to such a mess) the notes in the entire range of what you're copying in deciding what to "respect," and you will indeed get different results, depending. If I uncheck that option, copying is consistent no matter how many measures are selected.

[zuill]

I am talking about having that item checked.

Zuill

[zuill]

Okay. I created the file from scratch and copied to the TAB and got the same erroneous behavior. Exporting to XML caused the same corruption. If I manually transposed the TAB staff to the correct octave, or if I copied with[out] the option regarding respecting instrument range, then the result is correct and the XML is not corrupt.

So, I guess this should be reported. However, since this is not my file or music, I think you, jzucker, should report it and send the file for MakeMusic to examine and study. I would do the reporting if I had your permission.

Zuill