The title says it all. Below a comparison with Linux Libertine and Junicode.

Whenever I atempt to git clone the repository, I get the following errors at the end:

     error: transfer closed with 7215141 bytes remaining to read (curl_result = 18, http_code = 200, sha1 = aea33b5e763f74179a91371ded366546dd292f22)
     Getting pack list for http://github.com/georgd/EB-Garamond.git
     Getting alternates list for http://github.com/georgd/EB-Garamond.git
     error: Unable to find 7aa1a4b5e6a7b5b85f579edc494a4c2eb7aca863 under http://github.com/georgd/EB-Garamond.git
     Cannot obtain needed blob 7aa1a4b5e6a7b5b85f579edc494a4c2eb7aca863
     while processing commit d55bd762e7321d349da4421f61aaec40343b9b5d.
     fatal: Fetch failed.

I think this will improve the readability of the font for Slovaks and Czechs if EB Garamond which is serif font has serif-style caron on d, l and t and L with correct positioning and kerning.

In present version those letters look like they have acute accent instead of caron. Caron in fact has two forms: one is visible on letters like ž, č, ň, and the second is on the letters ď, ľ, ť and Ľ.

These letters should have the same kerning as the letters without caron.

My observation (because those letters are used in Slovak language) is that caron placed on letter t should be slightly lower than it is on d and l, as is illustrated in PDF file below on the OFL Sorts Mill Goudy font. Wrong position of this is visible on ť in Adobe Garamond Pro both regular and italic. Next observation is that when connecting t and k to ťk, and ľ and k to ľk, caron may touch the next letter, it is ok if it does (actually it is better if it does - see Adobe Garamond Pro and Linux Libertine font).

PDF file contains words used in Slovak language where this is illustrated.
http://ge.tt/95wSzEk/v/0

Not able to use the font 'EB Garamond Italic' from the 0.014c package within Xe(La)TeX.

Included using \usepackage{fontspec} \newfontfamily{\latin}{EB Garamond Italic}

When the equivalent .otf file is used from the 0.014b package there is not an issue.

Could you offer character variations for m-dash and n-dash so that they have curving (upwards) endings and (downwards) beginnings such as the hyphen has at the moment?

An example is in the following screenshot of another type.

In 0.015a, the "Th" ligature does not appear in MS Word unless "historical and discretionary" ligatures are selected, whereas I believe it should appear if "standard and contextual" is selected.

The pair f è in the roman font seems to need correction. Without it, the space between them
almost breaks up the word; see the top word in the image. Switching off kerning altogether locally
by
{\addfontfeature{Kerning=Off}réfère}
yields the bottom word and looks much better to me.
See also the example `auffällig' on p.31 of Jost Hochuli's nice little book
Das Detail in der Typografie, 2nd ed., 2011. He, too, prefers
to let the letters touch in this case.

As a global solution, I use

\usepackage{filecontents}

\begin{filecontents_}{kernegrave.fea}
languagesystem DFLT dflt;
languagesystem latn dflt;
feature kern {
pos f egrave -1;
} kern;
\end{filecontents_}

\usepackage{fontspec}
\defaultfontfeatures{FeatureFile=kernegrave.fea}

As far as I can see, if a small letter s follows a small vowel with a diacritic, its shape does not change in the same way as following a regular small vowel does in the italic version:

I was wondering if that is a deliberate decision, or just bells and whistles not addressed yet?

The capital Q in 8pt regular seems to be having trouble. Regardless of whether CALT is switched on or off, I always get the long-tailed version. The same thing seems to happen in the italic version as well.

On the other hand, it seems to work fine in the 12pt version.

This is a really great piece of work and I'd like to see it continue to be developed, but unfortunately, I know absolutely nothing about fonts or programming.

When you tag a github commit, github automatically creates a source tarball that you can download:
https://github.com/georgd/EB-Garamond/tags

Could the relevant github commits be tagged ot generate those tarballs? I'm looking into packaging these fonts in Debian and such tags would make my life easier (both for tracking releases, as reference points for packaging, and generating the source code tarball for releases)

You wrote: "As for the shape of de, el and lje, I was informed by the designers from Cyreal, that the triangular form put a strong accent on looking ancient (at least for russian eyes) so I changed default and alternate forms. As I said for sha, I want the less marked forms as default, therefore I agreed to have the modern form as default."

I think the reverse is true, triangular shape are the most striking difference in Cyrillic between Renaissance-like fonts and Didone ones. As for me, round shapes are marked in a Renaissance font, triangular forms should be the default.

Numbers are not correctly typeset in italic when switched to Numbers=Lining mode.

\documentclass[12pt,a4paper,twoside]{book}
\usepackage{fontspec}
\setmainfont[Numbers=Lining]{EB Garamond}
\begin{document}
Sekcia 12 a 13. \textit{Sekcia 12 a 13.} 
\end{document}

With the new file paths/names the download link needs to be updated.

http://www.georgduffner.at/ebgaramond/download.html

Thanks for keeping up the good work! EB Garamond is my main choice of font nowadays. ^.~

I would suggest adding the characters in the range 1DC4-1DC9 (http://en.wikipedia.org/wiki/Combining_Diacritical_Marks_Supplement). These are needed to represent contour tones in the IPA alphabet (cf. http://upload.wikimedia.org/wikipedia/commons/1/15/IPA_chart_2005.png).

Hi
U+0472/U+0473 CYRILLIC LETTER FITA look like U+04E8/U+04E9 CYRILLIC LETTER BARRED O. If you tell me how to insert a pdf here I'll show you the correct glyph.
Regards

This can be confusing but the Unicode name "LATIN CAPITAL LETTER O WITH MIDDLE TILDE" is a misnomer.

The Unicode 1.0 name was "LATIN CAPITAL LETTER BARRED O" and the character notes are "= barred o, o bar".

U+019F should have a straight bar like its lowercase U+0275.

Hello,
please bear in mind that the German "ß" exists only and exclusively as a lower case character. Therefore, in the SC fonts there should appear "SS" instead of a pseudo-capitalized "ß". I do know, of course, that there exist some persons campaigning for a capital "ß" but this is in no way an official tendency and in all probability never will be. To the contrary, one should better consider this leftover from the old German Fraktur (Garamond is Latin) as obsolete, as practised in Switzerland – but this may be a separate discussion. For the time being and in the official future, a capital "ß" simply does not exist and not implementing it would save the labour of correcting this letter manually whereever the SC font is used. Besides that: great, great work. Thank you for it.
Lisa

I just saw this on TeX.StackExchange "EB Garamond" font combines ʿ character and the preceding lette.

This is a bug, modifier letters are not combining marks; they are full spacing letters and should be treated as such, particularly I think the ring is a bit smaller than what linguistics like (remember it is really a letter not an accent), compare it with the one in Amiri which some still consider it smaller than it should.

It would be nice to have "TT" and "tt" ligatures, joining both letters, as the "Th" ligature does.

Would it be a case for discretionary, historical or standard ligature?

Consider modifying the @ glyph to include a “base a” that is less slanted, to better match the character of the roman style, especially in email addresses, where @ occurs intermingled with upright characters. Keeping the current form as an alternative glyph – which may be appropriate for traditional uses such as standalone commercial at – could be considered. If the current glyph is kept, it should be shifted down slightly so that the “base a” sits on the baseline and does not extend beyond the mean line.

I am just checking out that fantastic font and would like to thank you in the first place.

Today I had a look at the Specimen document and would like to post some things, I stumbled onto. Sorry, it will be quite an unsorted list, but maybe some stuff is very easy to fix:

• you are loading +tlig. What is that. Not mentioned in the document. Maybe put a comment in the source
• 2.1 Locals: for english and german you could maybe link to the package selnolig. Your feature works but does seperate ligatures in too many cases
• 2.1 when separating the ligature fl (e.g. auflachen), the two letters touch each other
• 2.2 Character Composition and Decomposition: Some examples would be nice, especially as you load ccmp by default.
• 2.8 Contextual Alternates: does "should be on by default" mean, that you recommend it? I didn't get it when reading it first time. Sounds a bit like: "is switched on, if feature already works in this version". As you notice, I'm no native, so maybe just a "problem" for me
• 2.8 IRAQ - the R and A are touching each other
• 2.10.2 General Variants:
  • cv05: I can't see any differences. Please describe further!
  • cv06: Can't see any difference either. Maybe because you set cv06 in your preamble

OK, that's all I have noticed. Sorry for that messy list. But maybe it is some help. In my opinion, the Specimen.pdf is a great show case, so everything should be clear.

First of all, thank for this great effort on making an open variant of Garamond!

I noticed that some letters ("л", "д", "с", "ч", "и") in Russian small caps in the EB Garamond look uneven (jumpy):

Compare the same title rendered in Adobe Garamond Premier Pro:

XeLaTeX from TeXLive 2013 was used to build both PDFs, snapshots of which are shown above.

In luatex, the following document produces large horizontal spaces rather than right and left half rings in the italic; the upright is fine.

\documentclass[12pt]{book}
\usepackage{fontspec}
\setmainfont{EB Garamond}
\begin{document}
Kitāb al-Ḥadāʾiq. Ittiṣāl al-ʿAql bi-l-Insān.

\textit{Kitāb al-Ḥadāʾiq. Ittiṣāl al-ʿAql bi-l-Insān.}
\end{document}
% This is LuaTeX, Version beta-0.76.0-2013070317 (rev 4627)
% i386-linux

I don’t know whether this is a bug in the font or in luatex — xetex does give me half rings in the italic (though with a space before the A that may be a kerning problem), but luatex has no trouble with half rings in my other fonts.

Many thanks for making these fonts support so many languages and systems of transliteration!

From R. de Sá-Nogueira Saraiva:
“fo, fa, ffe, etc., have so large a space that the break is noticeable when reading”

I'm trying to find a way to subset the font to make it smaller (I don't need any Latin supplements, Greek, Cyrillic, dingbats, etc.), but I want all of the ligatures (liga and hlig in particular).

Is there any way to do this? I've tried FontSquirrel but their subsetting messes up the ligature lookups (so the ligatures don't work).

Thanks!

It looks like Cyrillic letters included in EB Garamond Regular 08 haven't been adjusted to that optical size at all. Compared to Latin letters from the same font they look much lighter.

Font EB Garamond is not shown in Windows fonts because description is not showing when I run font viewer
. EB Garamond 8 description is OK so the font is visible in Windows in MS Word or other apps. MikTeX recognizes the font correctly (probably because it loads font files directly).

EB Garamond 12 is missing the info section:

EB Garamond 08 does have the info section

I was able to build previous commits, but now it is failing. Both ttf and otf fonts fail

$ git clone https://github.com/georgd/EB-Garamond.git
Cloning into 'EB-Garamond'...
remote: Counting objects: 102304, done.
remote: Compressing objects: 100% (33307/33307), done.
remote: Total 102304 (delta 78610), reused 92641 (delta 68956)
Receiving objects: 100% (102304/102304), 110.47 MiB | 830 KiB/s, done.
Resolving deltas: 100% (78610/78610), done.
$ cd EB-Garamond/
$ make
Generating build/EBGaramond08-Regular.otf
Failed to find NameList: AGL For New Fonts
Save Failed
Traceback (most recent call last):
File "tools/makefont.py", line 77, in
font.generate(args.output) #do somthing about 'old-kern'## should we really?
EnvironmentError: Font generation failed
make: *** [build/EBGaramond08-Regular.otf] Error 1

When using EB Garamond, I am finding that the text is a bit crammed up together and this really does tend to cause bad like-breaks on every page, using (Xe)LaTeX. There is probably not enough space between letters (and words), and also not enough shrink/stretch allowed by default.

Note: my documents generally come out without overflows with more "mature" typefaces (like Adobe Garamond or Junicode). This really is an issue as it makes the text almost impossible to debug properly.

Love your work, otherwise :)

Please offer example of Contextuals=WordFinal in specimen. The 'e' is elongated in italic and would be nice to see documentation about it.

Also if other Contextuals are in there, I could not find any, please document them.

Perhaps more elongated word finals could be added such as the 'r' and 'n'.

I used the latest release of the font 0.015d and printed a page with sample text and tried to read it to see if it would be pleasant, but caron on ľ, and ď is slightly lower - caron on ľ and ď is not on ascender line (see comment of Březina below), this is also visible when I zoom the font a lot. I found this information in discussion on http://typophile.com/node/78226

Please align it correctly as it is shown in the figure below (please use the 1st case as explained below).

This is David Brezina comment:
"Preferably (1). You want to maintain clean ascender line (as one would do it similarly with the x-height). The second case (2) could happen in handwriting, though. You might also need to do it in small cap /lcaron since there may not be enough space in bold. I also do it with Lcaron so all "vertical carons" in the font are at the same vertical position."

Other letters are fine to read (ť and Ľ).
See also how the vertical position of the caron is aligned on other correct fonts:

In the word "vis-à-vis", the "is" ligature is triggered the second time, but not the first:

I believe the dash shouldn't prevent this end-of-word ligature.

Oh my gosh I am so pleased with this typeface, and especially with the SS06 set you've cut for IPA usage! If I may be bold, can I suggest a few more stylistic alternates that would enhance SS06?

In IPA, the glyphs a (U+0061 LATIN SMALL LETTER A) and ɑ (U+0251 LATIN SMALL LETTER ALPHA) represent two distinct vowel sounds. It would be great if italic a retained its two-story configuration in SS06. Oblique double-story 'a' is an admittedly weird glyph, but there are a few precedents: IPAPhon and Cambria.

In IPA, the voiced velar stop g is always drawn in a one-story style. U+0261 LATIN SMALL LETTER SCRIPT G is the correct encoding for this symbol, but most linguists use the regular codepoint U+0067. It would be great if this could pick up a schoolbook style in SS06.

Similarly, the IPA symbol for the voiceless bilabial fricative ɸ is properly encoded at U+0278 LATIN SMALL LETTER PHI, but many users use the regular Greek codepoint U+03C6. It would be awesome if SS06 restyled this to match the latin-style phi symbol you've already drawn for U+0444 CYRILLIC SMALL LETTER FE.

Likewise, for when these glyphs are drawn: U+0263 LATIN SMALL LETTER GAMMA and U+03B3 GREEK SMALL LETTER GAMMA; U+025B LATIN SMALL LETTER EPSILON and U+03B5 GREEK SMALL LETTER EPSILON; U+028A LATIN SMALL LETTER UPSILON and U+03C5 GREEK SMALL LETTER UPSILON

Thank you so much for making this amazing typeface, and for contributing it to the open source community!

Font names show issues in Linux (Mint) and as result small caps (08 and 12) versions are not showing.
Can this be fixed by separating all font files or similar.
(using EBGaramond-0.015d.zip)

Thanks for great font

To me (and I am not an expert at all) it seems the kerning can be improved for f followed by a space followed by a singe left quote.

See this with XeLaTeX

Of ‘Never — nevermore’.

and the result in http://img442.imageshack.us/img442/1149/kerning.png where--to me--it seems they are too close.

Hello,

I just played around with EB Garamond as a web font when I noticed something odd with the woff/ttf files, they contain fewer glyphs than the .otf one.

Here's a list of some of the ones I think are missing,

B C D E H I J K L M P R S T U V W Y Z

Note: I didn't try the .eot files.

Also, thanks for an awesome font!

Is this font hinting issue?

see screenshots:
http://s9.postimage.org/6sussjgxr/ebgaramond.png
http://s12.postimage.org/rgdqlmihp/ebgaramond120.png

document was written in Office 2010 and exported to PDF using PDFCreator
http://www.filehosting.org/file/details/376137/doc1.pdf

As of now, the glyph at U+03C6 is the stroked version of phi ϕ in Bold, Regular and Italic. Regular and bold have the same at U+03D5, while Italic has nothing there. Most fonts have the loopy glyph φ at U+03C6 and the stroked glyph at U+03D5 and it would make EB Garamond more usable if it was the same.

And if you'd rather have the stroked one at U+03C6, we still miss the loopy one.

Thanks for all your much appreciated hard work on this font.

The two .otf fontfiles EB Garamond12_Regular & EB Garamond08_Regular are seen as the same font, and so it is not possible to install both fonts.

From what I understand, the font that was previously known as EB Garamond Caption is now optical size 08, while the previously regular font is optical size 12. I'm a bit surprised, because as a result, normal text gets rendered with optical size 08 while titles get rendered with 12 (using fontspec automatic optical size selection), so normal text looks very bold, and titles look very thin. Is that the normal behavior?

When I use:

\documentclass{article}

\usepackage{fontspec}
\usepackage{xunicode}
\defaultfontfeatures{Ligatures=TeX}
\setmainfont{EB Garamond}

\usepackage[bottom,symbol*]{footmisc}

\begin{document}

This is a test\footnote{really a test}
with 2\footnote{aka two} footnotes.

\end{document}

I expect to see a dagger for the second footnote symbol, but nothing shows up:

The font seems to have the dagger, but I'm wondering if it's located in the right place… or is it maybe using EBGaramond08 because of the small size?

Hi

I seen latest available ttf font release available on bitbucket is very old

EBGaramond-0.015d.zip
2013-06-28

while development added meanwhile many other fixes and features

Can you provide a more recent build for people, like me, interested to test and inspect font behavior, but unable to build fonts fom sources? Sorry for disturbing and thanks in advance

Using the swsh font feature for the italic font leads to big gaps between capital V/W and a following vowel.

I use fontspec to load EB Garamond to lualatex. When my document contains 'fl' characters, the automatically generated ligatures look fine. But if I copy the text, it will become 'flf'. This is font specific, since if I switch to 'Minion Pro' or the bulit-in Type 1 'Computer modern', the problem would be gone.

Here's my minimum working example:

\documentclass[letterpaper,12pt]{article}

\usepackage{fontspec}
\setmainfont{EB Garamond}

\begin{document}

\Huge{Fl Fi ffi fl fi}

\end{document}

Hi, Georg.

First of all, it is a real pleasure to see EB-Garamond becoming this beautiful and with strong development. In fact, it is getting so beautiful that I'm tempted to switch over completely to it. :)

But to write documents in XeTeX, I would need to have some kind of support for mathematics. Not exactly in a hurry, but I see that @khaledhosny has accumulated a good amount of knowledge with his XITS font. (Actually, it would be superb if he could document the experience that he gained and, perhaps, transform those in some scripts---as much as possible).

Anyway, I can help with the mathematics part, as I am a Computer Scientist. What I can't really help with is with the artistic side of things (as you probably noticed with the disaster that my URW Garamond No.8 turned out to be).

Thanks for all the beautiful work.

(I accidentally posted this in the wrong thread earlier. I apologize for the confusion)

Dear all,

My name is Marten Soderblom Saarela. Although I'm not a font developer, I've been following this Git hub group for a while because I really love the look of EB Garamond and I'm looking forward to the day when I will be able to use it in my academic writing. I'm doing Chinese history, and I often need to type Chinese transcriptions (the pinyin system) with tone marks. For the tone marks, I'm using the macron mark, acute and grave accent, and háček. I know that the acute and grave accents, which are actually not entirely uniform in width from top to bottom, as the tone marks are supposed to be, are not ideal in this context, but I haven't found a way to type the uniform, "proper" tone marks. (If anybody knows about this, please let me know!) Anyway, I just saw some undesired results when typing transcriptions using these marks, such as in the word "Tàizōng," where the grave accent on the à intersects with the T. I've attached a picture. I just wanted to bring this to your attention. Thanks for a great font.

https://dl.dropboxusercontent.com/u/402161/Taizong.png

Hi @georgd, it seems that you opened a Bitbicket repo only to make the zip files downloadable. However it's possible to attach binaries on github using Releases. Here's an example I've done to show you how it can be done:

https://github.com/benjaminabel/EB-Garamond/releases

I think you could use the semantic versionning if you want to publish multiple release this way. For example this release should have been 0.16.1 has it is a prerelease.

I could treat that manually, but maybe you haven't seen this collision yet.

Thanks for all your great work!

Hi there,

My apologies if this is not quite the correct outlet for this inquiry.
I'm having some puzzling difficulties using the italics in EB Garamond in XeTeX -- I just get an error that the italic font shape is undefined, and I've been unable to obtain italics adding a +ital feature or anything. I've tried this on two computers so far, and haven't had this problem with any other font. I also downloaded your specimen.tex and compiled it, and the italics simply show up as plain roman text.

Any suggestions you could provide with regards to troubleshooting this issue would be appreciated!

Thanks,

Radu

Bug report submitted anonymously on bitbucket. Attachment: http://georgduffner.at/ebgaramond/res/A.pdf

Hello, Please see attached PDF for strange behaviour of capital letter A compared to normal behaviour of the same letter with diacritical signs. [email protected]