More issues with F27.1: shifted beams

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

Moderators: Peter Thomsen, miker

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 5:18 am

Hello,
In F27.1 for Windows, I noticed that beams are horizontally shifted (for example in 16th notes), resulting in a slight offset with respect to stems. This is visible in Finale on screen and in the exported "pdf". I did not have this issue in an earlier version of Finale.
Was that so in F26.3 Windows also?
(There is also a visible shift between the note and the stem but I believe that is not new.)
It is getting tiring to see how many new issues Finale 27.1 has introduced. :twisted:
Attachments
f27.1issuebeamshift.jpg


User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Fri Jan 21, 2022 7:14 am

The shift between the note and the stem may be causing the misalignment of the stem and end of beam. If so, this can be corrected in Options/Stems/Stem connections.

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 7:18 am

Hi motet. I know all these misalignments can be corrected. The point is they shouldn't be there to begin with. In fact, the beam misalignment was not there before. My question is: is the beam misalignment also in F26.3 in Windows?

User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Fri Jan 21, 2022 7:20 am

Here's Finale 2014.5. Looks pretty close to yours. Are you sure it's new?
Attachments
1152.png
1152.png (94.07 KiB) Viewed 9125 times

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 7:47 am

I don't see it in F2007. Do you see it in exported PDF too?
I thought I could but I was actually not able to correct this in Document Options/Stems/Stem Connections or in Staff/Edit Staff Attributes/Stems/Stem Settings. The beam shifts horizontally with the stem.

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 2:46 pm

Any help?

Questions are:
-Can the issue with the beams showing horizontal offset past the stem be reproduced in F26.3 in Windows?
I see the issue in F27.1 on screen (attached picture) and in the exported PDF file. I do not have this issue in F2007.
-Motet suggested a way to correct the issue. However, anything I tried (via Document Options or Edit Staff Attributes) moves the beam with the stem. Is there a workaround?

Thanks.

User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Fri Jan 21, 2022 5:08 pm

I have a feeling this is just a screen aliasing issue. Here I saved it as PDF, extracted that as a 300 dpi image, and looked at it in a paint program, and the issue isn't there (Finale view included below as well). If I were you I wouldn't worry about it.
Attachments
1154.png
1154.png (6.83 KiB) Viewed 9081 times
1153.png
1153.png (66.62 KiB) Viewed 9081 times

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 5:47 pm

I take it this is F2014.5 you're showing. In my case (F27.1 Windows), I definitely see the issue in PDF too. It is not a display issue.
Though I appreciate the information, you don't really answer the questions motet: I was asking for results in F26.3 and a workaround.

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Fri Jan 21, 2022 5:57 pm

I decided to compare the behavior of Finale 27.1 between Mac and PC (via virtual machine). The situation with the Win version is really deplorable. I tried to correct the relationship between stems and note heads, but the strangest thing is that at high magnification (17,000% in Illustrator) the touches are still inconsistent — some notes are good, others are not.

One can go mad. However, this is the "smaller devil", because this inconsistency in printing is barely noticeable, because there is a touch of ellipses with straight lines, but by the connection beams–steams the situation is touchingly dramatic. Everything is obvious there and unfortunately there is no alternative way for improvement.

I guess Win users need to alert MM about this serious and urgent problem. But let's not be fooled that in Finale (Win) 26.3 things are much different — they are the same (I just did the same test, again via virtual machine). As a university lecturer, I would like to share that most of my students use PCs and they also informed me about this problem (stem connections) two years ago. I admit that we neglected thos inconvenience, because on the one hand the program was used for educational purposes, not publishing, and on the other hand I work on a Mac and there, with such arbitrariness in the Finale, I rarely have to fight (although there, as all Mac users know, there are other headaches).

The strangest thing, however, is that in none of the 3 updates of Win 26.3 these certain annoyances are corrected. I did a few additional tests. I found that even if some small adjustments are made to the stem connections, these settings don't help much for the upstems. The values ​​applied (in Finale for Win) were as follows: down stem: +0.5625 EVPUs; upstem: +0.29687 EVPUs.
This leads to some improvement in the picture, but not at all. The facts show that down stems are starting to behave relatively
close to the expectations, but in the other scenario (down stems) there is complete chaos.

Over the years, I've noticed that for one and the same version, the Mac program has specific issues that soon after MM partially, if not completely, solve, but then there is an alternation and the problem area is transferred to the Win application.
Prank...
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 6:27 pm

Hi wessmusic. Thanks for the extensive testing.

In the paragraph before last where you do some offset tests, do you do that in Staff/Edit Staff Attributes/Stems/Stem Settings?
I tried changing values there (in Horizontal Stem Offsets) but noticed that when I move the stem horizontally, the beam moves with it, so the problem remains intact. You seem to observe some improvement. I suggest you try larger numbers (like 10 evpu) and see what happens. I have not been able to shift beams left/right independently from other objects.

I and others have reported issues here that have not been solved by Makemusic for years. Yet, Makemusic reads these forums where users help them improve Finale for free. This doesn't bode well for my hopes that anything reported for F27.1 is resolved in an upcoming (?) update. Perhaps we are helping competition, if not Makemusic!

(By the way motet, I can see from the PDF image you posted that the beams and the stems are different blacks. How great is that!)

User avatar
Peter Thomsen
Posts: 6601
Joined: Fri Jul 25, 2003 6:47 pm
Finale Version: Finale v27.4
Operating System: Mac

Post by Peter Thomsen » Fri Jan 21, 2022 6:33 pm

sPretzel wrote:
Fri Jan 21, 2022 6:27 pm
… Yet, Makemusic reads these forums …
MakeMusic is monitoring the Finale forum on MakeMusic’s own web site.

This forum here - On The Other Hand - is not a MakeMusic forum.
As Far As I Know, the folks at MakeMusic do not (if ever) read this forum here.
Mac OS X 12.6.9 (Monterey), Finale user since 1996

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Fri Jan 21, 2022 6:43 pm

Hi Peter. I'm pretty sure they do. Anybody who cared about their own product would.

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Sat Jan 22, 2022 3:08 am

I opened the metadata file for the settings of Finale Maestro and it describes that the thickness of the stem is 0.091 sp.
This value corresponds to approximately 2.1875 EVPUs.
By the way, the settings you ask me about are in Document Options - Stems - Stems Connections.

Image

Examining in details the glyph of the note, I noticed that it is located slightly lower in relation to the baseline.

Image

The screenshot shows that the anchor points are also not symmetrical on the axis.
I have shown their values ​​in EVPUs. They correspond to the optimal settings of the stem connection.

The picture becomes even clearer when analyzing the image at high magnification in Illustrator.

Image

In the Mac version everything looks fine:

(clear stem connections)
Image

...and proper stem-beams joints:
Image


But for greatest pity Fin.Win generates insurmountable issue:

Image

What strikes me is that unlike Sibelius and Dorico, Finale for Win and Mac looks like two different softwares with almost the same features (Win. version suffers few facilities), but performed qualitatively differently.
Probably that's why the previous forum was divided into two sections - Finale for Win and Finale for Mac.
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Sat Jan 22, 2022 6:14 am

Excellent analysis wessmusic. Thank you.
You described in one of your earlier posts that you see some improvement by changing values in stem connections (presumably Stem Adjustments/Horizontal Upstem and Downstem values).
The strangest thing, however, is that in none of the 3 updates of Win 26.3 these certain annoyances are corrected. I did a few additional tests. I found that even if some small adjustments are made to the stem connections, these settings don't help much for the upstems. The values ​​applied (in Finale for Win) were as follows: down stem: +0.5625 EVPUs; upstem: +0.29687 EVPUs.
This leads to some improvement in the picture, but not at all. The facts show that down stems are starting to behave relatively
close to the expectations, but in the other scenario (down stems) there is complete chaos.
Referring to the last image in your last post, what I see on my end is that changing values in Stem Adjustments/Horizontal can correct for the stem-notehead connection offset, but not the stem-beam connection. The stem-beam horizontal connection is a unit and I haven't found a place in Finale where I can move the beam without moving the stem too.

To be more accurate, in Special Tools (like Beam Extension Tool, only applies to primary beam, or Beam Stem Tool Adjust), the beam can be moved horizontally with respect to the stem, but that is a manual move on a case by case basis. This isn't practical for an entire document. But, does this mean there could be a setting somewhere in Finale that does the same job? A script maybe (I don't see any relevant command in FinaleScript though)?
Last edited by sPretzel on Sat Jan 22, 2022 7:09 am, edited 3 times in total.

User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sat Jan 22, 2022 6:44 am

If the ends of the beams are determined by the position of the notehead rather than the stem (do you know, Wess?), then fixing the stem connection might indeed fix the beam alignment, as I suggested at the top of the thread.

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Sat Jan 22, 2022 6:47 am

The beam-stem joint position appears to depend on the horizontal position of the stem, not on the position of the notehead.
Last edited by sPretzel on Sat Jan 22, 2022 6:53 am, edited 1 time in total.

User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sat Jan 22, 2022 6:52 am

Yeah, I just experimented and was coming back here to report, and you're right.

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Sat Jan 22, 2022 2:06 pm

motet wrote:
Sat Jan 22, 2022 6:44 am
If the ends of the beams are determined by the position of the notehead rather than the stem (do you know, Wess?), then fixing the stem connection might indeed fix the beam alignment, as I suggested at the top of the thread.
Hi motet,
this idea also crossed my mind, but unfortunately to no avail. I went even further, adding new anchor points to the black note (a quarter note), changing the horizontal position of the origin, shifting the note in different directions - all in FonatLab. Then I recompiled the font, but the problem (I am talking only about Win version) turned out to be persistent.

Such headaches are more common in Sibelius, especially when the font is not properly compiled, but even then these issues are solved successfully by applying some of the above actions. Stem connections in Sibelius and Dorico are connected to the anchor points, which are located horizontally at the left and right side of the oval. If this is the case, everything from then on is trouble-free.

By the way, as you can see from my post above, FinaleMaestro follows these rules, too, but what comes out of this, since the beam-stem joints do not meet logical expectations, and worse, this discrepancy is visible when printed.

Unfortunately, it turns out that I am not able to solve this problem in general, except of course by using the Special tool.
However, this method seems to me too humiliating, resembling even a mental disorder, if I'm even forced to accept, however it is the last resort.

I also wondered if MM is looking from time to time at what this forum is about.
Judging by the issues discussed, as Peter Thomsen replies above, rather, they do not pay attention to our problems stemming from their negligence and in particular, this part of the team, that develops the Win version.
After all, we are human, no one is sinless...
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Sat Jan 22, 2022 4:04 pm

Hi wessmusic. Did you actually print to state that it's "visible in print"? The problem is real but I think how visible it is in print depends on the type of printer and the paper, in particular.

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Sat Jan 22, 2022 5:25 pm

Hi sPretzel,

here is a printed and photographed fragment at 1200 DPI (real, not expanded resolution), which shows the details we discussed.
The photo is just from iPhone and IMO the problem is clearly visible.

Everyone could judge for himself whether what we see brings satisfaction or needs to be persistently reported at MM.

Image
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Sat Jan 22, 2022 5:39 pm

Yes, it's visible in your print. Sad. Laser printer?

MM should listen and fix it quick. Especially since it's an issue that wasn't there before (but it's been introduced for long enough, as motet reported it's present in F2014.5). 🤞 There has been no update to Finale 27.1 in a while, maybe since it came out in November 2021.

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Sat Jan 22, 2022 5:54 pm

sPretzel wrote:
Sat Jan 22, 2022 5:39 pm
Laser printer?
HP 2015 DN (PostScrip, DIN A4 Duplex, Netzwerk) — almost 20 years old :)
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

sPretzel
Posts: 165
Joined: Sun Mar 11, 2007 8:38 pm
Finale Version: Finale 2007
Operating System: Windows

Post by sPretzel » Sat Jan 22, 2022 7:22 pm

motet, you posted two images for Finale 2014.5 in Windows, one which is the PDF and one which is the view on screen in Finale. Bizarrely, the PDF does not show the beam-stem offset issue.
So my question is for wessmusic: you said you see this issue in Finale 26.3 in Windows. Was that statement based on the PDF or based on what you saw on screen? I am wondering if, like motet, the PDF might look okay?

User avatar
wessmusic
Posts: 109
Joined: Fri Mar 22, 2019 10:38 am
Finale Version: Finale 27.4
Operating System: Mac

Post by wessmusic » Sat Jan 22, 2022 11:46 pm

sPretzel wrote:
Sat Jan 22, 2022 7:22 pm
So my question is for wessmusic: you said you see this issue in Finale 26.3 in Windows. Was that statement based on the PDF or based on what you saw on screen?
I'm talking about both options, of course - Finale and PDF.
The sample I showed the mismatch between stems and beams was printed from the PDF file on Mac OS because I intentionally shut down the virtual machine's network (although the printer you asked me for supports both network and USB). Unfortunately the problem is undoubtedly serious with both Win versions — 26.3 and at 27.1.

Here are screenshots taken from Illustrator of how things look "up close" at 6400% zoom. I would advise you to look at the images through the browser at 100 percent being opened in a new tab/window. So it can be seen quite clearly, especially since the pattern is the same in both tests. I want to clarify - this is not the same file. The musical text is entered separately for each try out.

26.3:
Image

27.1
Image

When working with orchestral scores, this problem may not be so pronounced and can be safely ignored, but when it comes to piano music, I think the situation is intolerable.
CUSTOM FONTS for FINALE and SIBELIUS
https://www.dropbox.com/sh/f0udkrvb8xvh2zj/AAD_8mVlRzzr5mjZKI7BR7Kza?dl=0
________
Finale user since 1994

User avatar
motet
Posts: 8226
Joined: Tue Dec 06, 2016 8:33 pm
Finale Version: 2014.5,2011,2005,27
Operating System: Windows

Post by motet » Sun Jan 23, 2022 1:05 am

This sounds like Finale is doing fixed-point (integer) arithmetic with not enough precision.

Post Reply