[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Accidentals' font

From: Paolo Prete
Subject: Re: Accidentals' font
Date: Sat, 4 Jul 2020 16:20:51 +0200

On Sat, Jul 4, 2020 at 1:46 PM Han-Wen Nienhuys <> wrote:

> On Fri, Jul 3, 2020 at 11:42 PM Paolo Prete <> wrote:
> > *All* my opinions are personal opinions. And I don't expect they will be
> > followed or implemented.
> If you really expect that, why bother sending this email?
Because it is important, in my opinion, in any case, to leave feedback.
This is because I consider Lilypond so powerful that I consider absurd that
its power cannot be highlighted by such defects (which are easy to fix).
Personally, I have no problem using workarounds, which is why it is not
very important *for me* that these problems are fixed. But I think it is
appropriate to leave feedback. Then everyone will judge according to their
own criteria.

> Let's all just assume that they're not just your personal opinions,
> but that you have found genuine defects in the font. We're not going
> to adjust glyphs because they don't conform to your opinion. We can
> adjust glyphs if we find there are bugs in them.
TBH I can't preview what you are going to do in * any * case, even if you
find my explanations convincing. I just leave feedback.

> > Meanwhile, for anyone interested I focus on some aspects that IMHO make
> > that font really desirable for Lilypond, at least as an option. Be free
> to
> > agree or not agree: here is my list
> >
> > 1) Look at the treble clef. IMHO Lilypond doesn't need to be so high, and
> > in fact the author of the font has reduced its height. Using a treble
> clef
> > higher than necessary leads to spacing problems between staves.
> I actually doubt this. The clef is at the start of the line, and there
> are usually no things sticking out of the bottom of the at the start
> of the previous line.

Look at the attached (very simple) example, which illustrates the case in
which a part must be extracted from the score (and therefore there is the
need to have * many * staves on the same page, and avoid turning the pages
too often)
If the highest point of the clef is too far from the first staff line, we
have the impression that the glyphs of the upper staff near the below clef
invade the lower staff.
If, however, the clef remains as much as possible within the staff, this
problem is reduced. It is not a simple problem to explain, but I can add
details, and you have to imagine this situation when there are many glyphs.

> > 2) In general, for keys, there is no need to use bold glyph. The default
> > bold glyphs, which are Bravura's philosophy (AFAIK), were fine when
> Can you explain what you mean with "bold" glyphs. The accidentals only
> exist in one flavor.

I do not understand exactly your question. Anyway, the vertical line of the
flat glyph, for example, is bold (let's say: too "fat", and it was
corrected by Gonville). The alto clef is too bold as well and has also been
lightened in Gonville. Note that it is bad practice to use fat symbols when
the symbol itself has many lines and curves (for example: alto clef),
because if a symbol is complex in itself, it is distracting in itself,
therefore it needs to be rendered as light as possible.

> > 3) Look at the time signature. In the Feta font the two numbers overlap
> and
> > can be annoying at the sight, IMHO. This problem was solved by the
> Gonville
> > font.
> Can you file a bug with an image? Over here, the digits do not overlap
> with each other, but do overlap slightly with the staff lines. (see
> attachment).
It is exactly the overlap with the central line which (IMHO) is very
annoying to see and creates a unique glyph with the numerator and
denominator. Note that even in the plate engraving of 1902 they avoided

> > 4) Look at the sharp glyph: it is reduced in width, saving space and
> > remaining perfectly legible at the same time.
> The gonville font is well done, but we usually base formatting
> decisions on plate engravings. Could you do some measurements to
> demonstrate what ranges of dimensions (relative to staff space) are
> common?

In order to understand the defect of the Feta sharp (or, better: what in my
humble opinion is a defect), look at the example I have attached.
There is a very irregular spacing between accidentals and note heads.
It is also too wide and this is annoying when there are too many notes
between (see the attached example as well).
Now see how the spacing becomes regular with Gonville (I can make more
accurate examples, but this gives you an idea). You see, now, vertically
ordered columns, and not a mess of glyphs.
This is because the sharp symbol is itself larger than the flat one. But
they have the same importance. To solve this problem, a compromise must be
found. The Gonville font uses a sharp symbol that is *slightly* wider than
the flat symbol. And this width is carefully calibrated.

> > 5) Look at the flat glyph: the horizontal line has been reduced in
> > thickness, which once again pairs well with lighter staves and stems.
> The flat glyph has no horizontal line. You can see the glyph it was
> based on over here:
Forgive me, it's a typo: I meant the vertical line (as I also wrote in the
previous post)

> > 6) Look at the glyph of forcing: it forms a single word, and not the
> union
> > of three different characters (which is very distracting, in my opinion)
> what is 'forcing' ?
Forgive me: a typo again: I meant the "sfz" symbol.

> > 7) Look at the trill glyph: it's simpler, cleaner and less "baroque",
> Maybe Jan can comment. The comment says it's based on a Saint Saens
> Cello concerto, which I can't recall ever having seen.
Ok, but there is no consistency with the general style. And consequently
the "baroque" look (which does not necessarily mean what belongs to the
Baroque era: in Italy we use this term to denote what is strange or
bizarre). There is much, much more to say, but if I were you I would ask
all of these things to the author of the font. Describing such meticulous
and long work in a few lines is a really difficult task.

> > 8) Look at the bequadro glyph: the Feta one is too large compared to the
> > NoteHead and this has also been corrected in the gonville font.
> Scans? "Too large": do you mean too wide, or too high?
Too high.

> Thanks. I could go on asking questions, but I am really looking for
> data rather than opinions.
Again, I hope I am not unpleasant and polemic, but I disagree. The criteria
of an accurate engraving are not obtained only with the data.
They are also obtained with opinions, comparing everyone's opinion. The
more lively the debate, the better the results.
A too politically correct and too weak debate on issues so controversial
and difficult to interpret means reaching no conclusion.

Attachment: accidentals-gonville.pdf
Description: Adobe PDF document

Attachment: accidentals-feta.pdf
Description: Adobe PDF document

Attachment: clef-test-feta.pdf
Description: Adobe PDF document

Attachment: clef-test-gonville.pdf
Description: Adobe PDF document

reply via email to

[Prev in Thread] Current Thread [Next in Thread]