Simplified "Stoic" Sagittal for 13-limit tonality diamond use

User avatar
cmloegmcluin
Site Admin
Posts: 1667
Joined: Tue Feb 11, 2020 3:10 pm
Location: San Francisco, California, USA
Real Name: Douglas Blumeyer (he/him/his)
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by cmloegmcluin »

Wow, the Stoics have really come into their own! Cool to see that y'all have already settled on code points, Sagispeak, and Sagitype for them. And already updated the Smilies and Wincompose sequences.

Eventually I'm going to need to work them into the Sagittal code base I've been working on (eventually to power the online notation calculator, among other things), the written educational materials (not published yet, still a WIP), the educational videos (same), the Xenharmonikôn article (well maybe Dave will do that, or not), and the Sagittal-SMuFL-Map. And we'll actually need to get them submitted to SMuFL, of course.

So there's no multi-shaft versions of these, and they don't need to be added to the JI precision levels diagram.

ndentonprotsack
Posts: 24
Joined: Wed Mar 31, 2021 2:44 am
Real Name: Nicholas Denton Protsack

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by ndentonprotsack »

Hello Both,

Great to hear from you, and, yes, I am very excited to see Stoic getting off the ground in the ways that have been discussed! I will throw in my two cents on the questions about sagispeak and the character map once I have familiarized myself with them a little better.

Regarding the last exchange we had, Dave, about the readability of the symbols, I have taken another stab at them myself. Since I have Dorico and can cross reference them on the staff, I felt that this might help us streamline the process a bit. I edited the double and triple dent symbols to maximize both visibility on the staff, while also preserving their ability to be differentiated from one another. Here are the results (I've attached the font, too):
Image
Stoic_Sagittal.ttf
(12.5 KiB) Downloaded 6 times
At this point, I am more than happy with these symbols. I have started using them in a composition I am working on, and I find they are quite easy to differentiate from one another, and are generally quite readable. However, if you have any final edits to make, please be my guest!

Best,
Nick
PS Agreed about avoiding distributing the "Stoic Sagittal Font", Dave. We can wait on that; however, I may use it for my personal work in the meanwhile as I am working on a few pieces that will incorporate the Stoic symbols already.
Attachments
Stoic Sagittal demo 4.png
(67.48 KiB) Not downloaded yet
Last edited by ndentonprotsack on Sat Sep 04, 2021 8:03 am, edited 1 time in total.

ndentonprotsack
Posts: 24
Joined: Wed Mar 31, 2021 2:44 am
Real Name: Nicholas Denton Protsack

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by ndentonprotsack »

Whoops, I meant to send the FontForge file, not the font itself... my bad.

Here it is:
Stoic Sagittal-7.sfd
(46.86 KiB) Downloaded 7 times

User avatar
Dave Keenan
Site Admin
Posts: 2026
Joined: Tue Sep 01, 2015 2:59 pm
Location: Brisbane, Queensland, Australia
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by Dave Keenan »

ndentonprotsack wrote: Sat Sep 04, 2021 8:01 am Regarding the last exchange we had, Dave, about the readability of the symbols, I have taken another stab at them myself. Since I have Dorico and can cross reference them on the staff, I felt that this might help us streamline the process a bit. I edited the double and triple dent symbols to maximize both visibility on the staff, while also preserving their ability to be differentiated from one another. ... if you have any final edits to make, please be my guest!
Nick, You have done an admirable job of making the double and triple scrolls distinct from each other, and of preventing the triples from getting lost against the staff lines, as my previous triples were wont to do.

Another problem I had with version 6 was that, although the doubles did not get lost against the staff lines, some features of them did. I find that this problem remains in version 7. Your work did not make this worse, but didn't fix it either.

So I have produced a version 8 (attached below). This time, I did what I should have done long ago and made a piece of staff-line as a glyph in the font, so that I could temporarily combine copies of it (by reference) with a Stoic glyph, to test the Stoic's appearance on the staff, in both positions.

I have tweaked the doubles to open up some half-pixel gaps between them and the staff lines, at strategic locations. I hope you find that, in doing so, I have not significantly reduced any of the gains you have made.

I felt that your triples had the appearance of flaring towards their ends, where other scrolls taper (before being cut off horizontally). I realise this contributed to the required distinctness from the doubles and stafflines, so I have tamed this only to the point where (I hope) it now appears to maintain constant width — neither flaring nor tapering.

I also felt that the triple barbs (12 dents) were getting lost against the staff lines. So I have increased their slope, and reduced their vertical spacing, from 4 pixels (half a staff-space) to 3.5 pixels.

I also ran FontForge's Verify command and corrected a problem it found where one of the shafts had two control points in the same place. There is a remaining problem whereby all the even-numbered (downward) Stoics have their outlines going anticlockwise when they should be clockwise, because they are flipped references to the odd-numbered (upward) Stoics. But that can only be fixed by breaking the references, so I'll leave that until we are definitely finished tweaking.

I hope you find version 8 to be acceptable, and I look forward to seeing the results of your Dorico test using it.
Agreed about avoiding distributing the "Stoic Sagittal Font", Dave. We can wait on that; however, I may use it for my personal work in the meanwhile as I am working on a few pieces that will incorporate the Stoic symbols already.
Sure. No problem. I look forward to seeing the scores, and hearing the pieces. :)
Attachments
Stoic Sagittal-8.sfd
(47.09 KiB) Downloaded 9 times

ndentonprotsack
Posts: 24
Joined: Wed Mar 31, 2021 2:44 am
Real Name: Nicholas Denton Protsack

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by ndentonprotsack »

Dave,

These changes are great! I like them very much, especially the double dent redesign and the increased slope on the triple barb (12 dent).

I did some extensive testing of them on Dorico, and I noticed two small areas for improved visibility---so I made some extremely slight adjustments. Hardly noticeable, and definitely not any kind of redesign, so I opted to call the new version 8.5.

Here is what I did:
- I made the curve or "belly" of the double dent symbol slightly more pronounced, because I noticed that it when it appears on a space on the staff, a greater curve allows for more white space to appear between the upper double dent and the staff line. Thus, it obscures the feature of the symbol just slightly less, without compromising its appearance in any other position.

- I took what you wrote about the triple dent symbol to heart and took it even further. I now made it taper at the tip ever so slightly, increasing the verticality of the outer edge as a result... This actually allows for more visibility when it appears on a space of the staff, as it juts out from the staff line a little less shallow than before.

Here are the results:
Image
Stoic Sagittal-8.5.sfd
(46.89 KiB) Downloaded 2 times
I am 100% happy with these. They are super readable now; I even can make them out distinctly when the page is 30% the size of a typical 8.5"x11"!! I think these are definitely good to go, if you are in agreement!

Best,
Nick
Attachments
Stoic Sagittal demo 5.png
(70.54 KiB) Not downloaded yet

User avatar
Dave Keenan
Site Admin
Posts: 2026
Joined: Tue Sep 01, 2015 2:59 pm
Location: Brisbane, Queensland, Australia
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by Dave Keenan »

I agree with your changes, except that when, for the double scroll on a space, you increased the sub-pixel-sized semicircular whitespace between the upper scroll and the staff line (a good idea), you decreased the sub-pixel-sized triangular whitespace between the lower scroll, the shaft and the staff line. I have restored this, and possibly increased it a little, in line with the increase you made in the semicircular one, by moving the on-curve point above the triangle, up by 4 fu, and moving the on-curve point to the right of the triangle, right by 5 fu.
Attachments
Stoic Sagittal-9.sfd
(46.99 KiB) Downloaded 2 times

ndentonprotsack
Posts: 24
Joined: Wed Mar 31, 2021 2:44 am
Real Name: Nicholas Denton Protsack

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by ndentonprotsack »

Awesome! Great eye on that final adjustment, Dave. I tested them one more time on Dorico, and I am very happy with them. So cool to progress all the way from sketches on a piece of paper to these beautiful symbols!

Now, I think we are ready to move on to the next order of business, if you are as well! How do you suggest we proceed?

Best,
Nick

User avatar
Dave Keenan
Site Admin
Posts: 2026
Joined: Tue Sep 01, 2015 2:59 pm
Location: Brisbane, Queensland, Australia
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by Dave Keenan »

The next step was to convert the new symbols into a form compatible with the Bravura font. The main difference being that it uses 1000 fu per em (250 fu per staff-space) where we've been using 2048 fu per em (512 fu per staff-space). We designed them using 512 partly because that's the TTF (true-type font) standard to which all the previous Sagittals were designed, and partly because it's divisible by 8, while 250 is not. So our 8 pixels per staff-space were 64 fu per pixel, whereas they would ideally be 31.25 fu per pixel in the OTF (open-type font) format used by Bravura, but in fact they will be rounded to 31 in three-quarters of cases, and rounded to 32 in the other quarter. Horrible, but it can't be helped, since Steinberg chose to use 1000 fu per em.

I made many passes before I got the following correct procedure: I made a copy of "Stoic Sagittal-9.sfd" and renamed it to "Stoic Sagittal-9-otf.sfd". Opened it, shift-clicked to select all the upward Stoics and chose Edit->Unlink Reference, then Element->Overlap->Remove Overlap. Then, annoyingly, I had to edit each upward Stoic, to turn all the on-curve points at the bottoms of the dips, back into HV-curve points (from circles back to diamonds). Then went to Element->Font Info->General and set Ascent: 800, Descent: 200. Then went to Element->Font Info->OS/2 and set OS/2 Version: 4, and clicked OK. Then chose Element->Validation->Validate. There were the expected "Flipped references" for all the downward Stoics. I ignored them for now. I also ignored all errors in the 1, 4 and 8 dent symbols since they already exist in Bravura. And I ignored errors in any downward symbols that would automatically be fixed by fixing the corresponding upward symbol. But in the 3-up, 5-up, 7-up and 11-up there were errors of the type "Missing points at extrema", which required manual adjustment of control points by 1 fu to realign them with their on-curve points to make the warning circles-with-cross-hairs go away.

Then when there were no more errors, I selected the downward Stoics (actually it was easier just to select them all) and chose Edit->Unlink References, then Element->Correct Direction. Stupidly, this made "Missing points at extrema" errors in 3, 5, 7 and 11 down. How this can be, when they are supposed to be just flipped copies of the upward symbols, I don't know. But I manually fixed these too. I then selected them all and chose Metrics->Set RBearing and set the right side-bearing to zero, because that's the rule for accidentals in Bravura. Steinberg figure it's up to the notation software to decide how far apart to space them, and how far from the notehead. Then I checked that all the widths were multiples of 31.25 bankers-rounded to an integer.

Here's the OTF-ized SFD file for you to test, and possibly adjust, before I copy the glyphs into a copy of Bravura and rename it "Bravura-S". You should generate a .otf file for testing. What format have you been using up 'til now?
Attachments
Stoic Sagittal-9-otf.sfd
(55.14 KiB) Downloaded 2 times

User avatar
Dave Keenan
Site Admin
Posts: 2026
Joined: Tue Sep 01, 2015 2:59 pm
Location: Brisbane, Queensland, Australia
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by Dave Keenan »

Nick, don't spend any more time on my previous upload. Sorry. I just tried copying those glyphs to a copy of Bravura and found that they no longer passed validation, due to non-integer coordinates. When I investigated why some coordinates suddenly grew fractional parts, just through being pasted into Bravura, I realised I had forgotten a conversion step. Bravura uses cubic curves whereas we designed the Stoics (like all Sagittals) using the easier-to-manage quadratic curves.

It will be better if I do the conversion over from scratch, and do the quadratic to cubic conversion before the 2048→1000 down-scaling, so there is only one round-to-integers step. If there were two such steps, the errors might accumulate.

I'll post a new OTF version soon.

User avatar
Dave Keenan
Site Admin
Posts: 2026
Joined: Tue Sep 01, 2015 2:59 pm
Location: Brisbane, Queensland, Australia
Contact:

Re: Simplified "Stoic" Sagittal for 13-limit tonality diamond use

Post by Dave Keenan »

I decided to go all the way, and add them to Bravura-S now. So in testing them, you'll have to deal with the less convenient non-consecutive code points. Sorry.

E300 :|(:	E301 :!(:
E40C :|((:	E40D :!((:
E40E :|(((:	E40F :!(((:
E302 :/|:	E303 :\!:
E410 :/|(:	E411 :\!(:
E412 :/|((:	E413 :\!((:
E414 :/|(((:	E415 :\!(((:
E306 ://|:	E307 :\\!:
E416 ://|(:	E417 :\\!(:
E418 ://|((:	E419 :\\!((:
E41A ://|(((:	E41B :\\!(((:
E41C :///|:	E41D :\\\!:
Attachments
BravuraS.otf
(600.46 KiB) Not downloaded yet
BravuraS.sfd
(3.61 MiB) Not downloaded yet

Post Reply