Non-standard Numerals and Letters

I am new to designing fonts and was wondering if anyone out there might be able to answer my question: when designing a font what are the typical characters that are included in the superscript, subscript, numerators, denominators and ordinals and scientific inferiors?

Typically Adobe include the characters included in the image attached (where red text shows the text that changed when applying that style), however, I am wondering if there are any more that should be added? Also what is the difference between Subscript and Inferior and Superscript and Superior? I would be grateful for any feedback. Thank you.

AttachmentSize
Numerals and Letters.png85.54 KB
Karl Stange's picture

Looking at Adobe Caslon Pro Regular (v2.092) it looks like they include a large part of the Superscripts and Subscripts Unicode codepage (these links to Unicode codepages from the Consortium's web site may trigger a PDF download depending on your browser, OS and settings) as well as those superscript characters that appear in the Latin-1 Supplement Unicode codepage. Other Unicode codepages, such as the Phonetic Extensions and Phonetic Extensions Supplement contain examples of other sub- and superscript characters.

For more advanced mathematical support there may also be additional sets of non-Unicode specific characters to support, for example, a complete set of sub- and superscript alphanumeric characters in Latin and non-Latin scripts. If you have access to the Cambria Math font this contains a vast range of both kinds.

I am not sure about the differences, but I think that Inferior and Superior refer more specifically to usage in scientific, technical and mathematical typesetting and composition. The Wikipedia page might be worthwhile reading on the subject.

peterfwyang's picture

Hi Karl,

Thanks for the Unicode pages, v. helpful. The wikipedia page is what was confusing me. I took a look at Cambria (TTF) compared to Adobe Caslon Pro (OTF) and there are differences in how things are encoded see below:

TTF Lookup // OTF Lookup
SUPS // SUPS * (See note below)
SINF // ------
NUMR // NUMR
DNOM // DNOM
ORDN // ------

Opentype SUPS has two lookups, one with a subtable containing a-z lowecase and the other with a subtable with all the symbols i.e. 0-9 -$c(),.

I'm trying to work out how the encoding happens for Inferiors and Ordinals (you see opentype has no SINF or ORDN lookups)

Karl Stange's picture

Those two fonts would have been engineered very differently. To the best of my knowledge both Ross Mills and John Hudson at Tiro worked on Cambria Math and would be able to give you insight into how those features were programmed. I know based on comments made in the Build section and elsewhere that John uses VOLT to develop OpenType features and so it is possible that this workflow had a completely different impact than one using AFDKO as likely used by Adobe.

With regard to your original question, unless you are developing a font that will likely be used in technical composition contexts then you would only really need to include those characters seen in the standard Latin-1 Supplement codepage. Personally I would probably also include those characters in the Superscript and Subscript codepage as well and leave it at that.

Karl Stange's picture

Considering the nature of the question and assuming that you are still able to do so, this might be better placed in the Build forum.

Syndicate content Syndicate content