Font Forum

I have been working on this font for some months. Although I have worked with many CAD systems throughout my career I have no previous experience with typography.

I'm sure there are many mistakes and anomalies which I have missed and I'm not sure I have got the Panose numbers correct. Any constructive criticism will be gratefully accepted.

This font is under the SIL license so you may copy it and use it as you will but I would like to put a couple of caveats on that. If you are going to make changes to the font I would be grateful if you could rename it to something other than 'Kelvinch'. If you are going to make additions to the font I would be grateful if you could send me a copy, other than that do what you like with it.

This font started out as a Font Forge project but then migrated to Font Creator. Basically I took the Gentium font and altered it (it is now considerably different to Gentium). It was better than starting with a blank canvas.

Many thanks to Bhikkhu Pesala for his example font in Font Creator. The estimated symbol and the musical sharp, flat and natural symbols plus a few others have served as the prototypes for the ones in Kelvinch.

Are there any glaring omissions, characters which ought to be included?

Any comments or advice will be gratfully accepted.

Regards

Paul.

PJMiller wrote:This font is under the SIL license so you may copy it and use it as you will but I would like to put a couple of caveats on that. If you are going to make changes to the font I would be grateful if you could rename it to something other than 'Kelvinch'.

The SIL Open Font License (OFL) already requires that any derivatives use a different name:

OFL web version wrote:The OFL allows the licensed fonts to be used, studied, modified and
redistributed freely as long as they are not sold by themselves. The
fonts, including any derivative works, can be bundled, embedded,
redistributed and/or sold with any software provided that any reserved
names are not used by derivative works. The fonts and derivatives,
however, cannot be released under any other type of license. The
requirement for fonts to remain under this license does not apply
to any document created using the fonts or their derivatives.

Hi Paul

Thank you for posting the four project files.

I downloaded all four and then tried to open Kelvinch.fcp in FontCreator.

However, I cannot open the file as I am using FontCreator 8.0 and the project file was generated using a newer version of FontCreator.

Could you possibly post font files as well please so that I can try them in use in various Serif products, such as PagePlus and CraftArtist please?

Also, I could study the fonts in FontCreator 8.0.

Best regards,

William Overington

31 August 2015

Sorry, I was making some big assumptions there.

Here is an image for anyone who wants to see without downloading it.
Enjoy.

Thank you for posting the fonts.

The first thing I noticed, from the image in your post, is the lowercase e. Ah, a Venetian font.

I like Venetian fonts.

Look at those magnificent digits. They are a beautiful blend of modern digits with the non-lining digits of old style metal type fonts. I have never noticed that effect in a font before.

I have downloaded all four fonts and installed them.

I have also looked at the Kelvinch-Roman.otf font in FontCreator 8.0, primarily to have a look at what is the glyph content of the font.

I visually checked for Esperanto coverage and I am pleased to note that it is present.

http://www.users.globalnet.co.uk/~ngo/t ... eranto.pdf

I checked for Welsh coverage by copying from the following pdf and pasting into WordPad and then formatting to Kelvinch.

http://www.users.globalnet.co.uk/~ngo/t ... acters.pdf

All but six characters are covered by the font.

On my computer, four of the missing characters showed each with a .notdef glyph and two each showed with a sanserif glyph that the computer had found from somewhere.

I am pleased to note the AE macron and the ae macron, which, I think, are sometimes used when transcribing texts from Old English originals.

Quotation marks are there. I tested by copying from the following pdf and pasting into WordPad and then formatting to Kelvinch.

http://www.users.globalnet.co.uk/~ngo/t ... _marks.pdf

The font does not seem to have Hot Beverage. Hot Beverage can be useful when preparing program sheets for meetings, as the symbol can be used to indicate a refreshment break.

http://www.users.globalnet.co.uk/~ngo/t ... verage.pdf

Latvian and Maltese are both good. Excellent.

http://www.users.globalnet.co.uk/~ngo/t ... atvian.pdf

http://www.users.globalnet.co.uk/~ngo/t ... altese.pdf

I did notice that there are no f ligatures, namely ff fi fl ffi ffl which is a surprise. They would be a useful addition.

If you were to add an st ligature and a ct ligature that would be great.

As this is a font with a renaissance look, a Venetian too, if you were to add a good selection of long s ligatures then that could be very useful to some people. It is a niche interest, yet a good design of font with such ligatures would be great.

The font is excellent as it is.

William Overington

31 August 2015

Hi Paul

If you do decide to include a ct ligature glyph, then the glyph could be accessed by an OpenType table.

In addition you might like to consider a direct access through the Private Use Area so that a display could be produced in non-OpenType-aware application programs, like WordPad. If you do decide to do that, then you could map the ct ligature glyph to anywhere you choose in the Private Use Area. However, you might find a mapping to U+EEC5 would be a good idea as the font would then be compatible with the suggestions available from the following webspace.

http://folk.uib.no/hnooh/mufi/

http://folk.uib.no/hnooh/mufi/specs/MUF ... rt-3-0.pdf

Best regards,

William

Dear William,

Thank you for your praise. I also like venetian fonts, as for the digits, I first tried doing the digits old style with the ascenders going up to the 'CapHeight' and the descenders going to 'WinDescent' but it looked ridiculous, then I tried lining digits and that didn't look the way I wanted so then I tried it half way and that achieved the look I was after. It also aids legibility because the size of the zero is half way between lower case o and upper case O so they are easily distinguished.

The extra characters for Welsh will be in the next iteration.

I am currently tidying it up. There are a lot of extra contol points left over from the copious use of the 'Add on-curve extremes' button in the real time validation window. I have also reduced the size of the bowl on the lower case b.

I never cared much for ligatures. A few years ago I used some ligatures in a document I was producing, I had to insert them manually and then could not 'find' the words which contained ligatures.

Example :- I live in Sheffield. When I put the 'ffi' ligature into Sheffield then it's spelling is not 'Sheffield' any more it's 'She<some strange character>eld' and so a search for 'Sheffield' in the word processor doesn't find it.

I am not sure if I got the Panose numbers correct. The Panose website is very abstruse and I'm not certain I got all the measurements correct. Is there an easier way of doing this?

Regards

Paul.

PJMiller wrote:I never cared much for ligatures. A few years ago I used some ligatures in a document I was producing, I had to insert them manually and then could not 'find' the words which contained ligatures.

That's why OpenType Features should be used for ligatures. The text string doesn't change — it will still be "ct" or "ff" but when an OpenType aware application finds those strings it will replace them with ligatures.

The ct ligature should be a Discretionary Ligature (dlig) unless you're creating an archaic font, and the "ff, fi, ffi, fl, ffl" ligatures should be Standard Ligatures (liga)

There's a tutorial on Panose Classification. I don't think it's important to get it 100% right. Most fonts don't include this information. All I do is copy the Panose number from a similar font that does have it, and paste it into the Font Properties dialogue.

The number for Gentium is 2-0-6-6-8-0-0-2-0-4

Dear Bhikkhu,

I will have a look at putting some ligatures in the font and see what I can come up with.

I tried to make the Panose numbers as accurate as possible based on the information on the Panose website and the number I came up with for Kelvinch Roman was 2-5-5-3-4-5-6-2-2-3 but if you compare Kelvinch to Gentium they are considerably different. Kelvinch started out as Gentium but it ain't Gentium anymore.

I do agree that PANOSE is not very important, but it is recommended to at least provide accurate information there.

Here is a direct link to a PANOSE chart (which is also mentioned in the topic that BP pointed out) which might be useful:
PANOSE Chart

For archiving fonts, whether in *.fcp or *.otf formats, it's well worth using 7-Zip LZMA compression. Those four font files are just 272 Kbytes instead of 743 Kbytes, or 402 Kbytes as a Zip archive.

One anomaly I noticed with your font is that the .nul glyph uses the .notdef glyph contours. I think it should be empty.

Does everyone on the forum have 7-zip ?

It's free, Open Source, and only a 1 Mbyte download so there's no good reason for them not to get it. Other programs like WinZip can also open the archives. If you host fonts on your site, you and your visitors will save a lot of bandwidth.

Kabala, for example, is 593 Kbytes as a 7z archive or 1.58 Mbytes as a Zip archive, or 3.31 Mbytes as separate font files.

PJMiller wrote:Does everyone on the forum have 7-zip ?

I also have used it for years. Often I just use ZIP compression when I exchange files, but the performance in the way of compression of the 7z format is great. I use it for archiving a lot of jobs.

Mike

Ok ... here is the latest iteration of Kelvinch Roman, the others (italic, bold and bold-italic) will follow as soon as they are finished, probably sometime next week.

I have included the Welsh characters and a few others, I've erased the .null characters contours. I have decreased the size of the bowl of the lower case b and corrected many other mistakes.

I was wrong not to include ligatures, just because I don't use them, there are many characters in this font that I will never use. I have added the glyphs for the ligatures and I will add them as open type ligatures when I become competent at driving Open Type Designer (it really needs some decent documentation).

There may still be mistakes in there, either in the glyphs themselves or in the metadata, if you notice anything wrong please let me know and I will do my best to correct it.

Also are there any characters which should be in there but are missing?

If anyone doesn't have 7-zip then let me know and I will post the uncompressed files.

Thanks for all your help.