Tell me more ×
TeX - LaTeX Stack Exchange is a question and answer site for users of TeX, LaTeX, ConTeXt, and related typesetting systems. It's 100% free, no registration required.
up vote 10 down vote favorite

This may be a little bit off-topic. However I think it's still TeX related.

Recently I decided to move my personal style (a class file) to be unicode-math and fontspec based, so I can freely use some commercial fonts I have and more important, to take advantage of some opentype features.

However, as a Statistics PhD student, Math font is surely the first thing I have to decide. And after some trying I have decided to stick with Cambria math for now. Though I like euler, I don't think my supervisor will like to see it in my thesis. And the design of STIX is just too "Times Roman" and I simply don't like Times all the time. And the design of ASANA math, in my opinion still requires some improvement, especially the weights do not blend very well.

So I have to decide which text font to use with Cambria math. Though I fell Cambria math is acceptable to me, I don't like its text font.

Any suggestions are more than welcome. I have both commercial and free fonts for choice (more specific those in Adobe Font folio 11 and some others).

I would prefer a fonts with at least the following features, smcp, case, cpsp, lnum, onum, tnum, pnum, and of course kern, liga

In addition, I don't think the text fonts have to be similar to Cambria Math in design. This is neither a sufficient nor a necessary condition for the text and math to blend well.

Just to give a example. Arrus, designed by Lipton. If one set Arrus in body, and need some large size text for headings, then Cataneo will be a far more better choice than just scale up Arrus. If similarity is a criterion, then scaled Arrus has exactly the same shape as non-scaled Arrus. The problem is that scaling changes weight, and weights is one of the criterion for matching fonts.

I am not saying that similarity is something bad. It is just not necessary. And sometimes not sufficient. For example, Cambria and Cambria Math, they bare the same name but they do have different designs. And simply don't like Cambria's small caps.

share|improve this question
3  
Well, I'd still suggest you to use Cambria with Cambria Math. – Leo Liu Jun 10 '11 at 17:34
2  
If, in your opinion, design similarity is not a criterion for "matching", please explain what is. – lockstep Jun 10 '11 at 17:42
@lockstep For example, Concrete Roman and Euler, they are not similar, but they blend very well. A match between fonts do not mean they have to look the same. Instead, their weights, slope, etc need to mix good. – Yan Zhou Jun 10 '11 at 17:48
@YanZhou: Concrete Roman and Euler don't blend well at all (they only match in weight, not in style, which is the most important criterion). That's not a surprise since the only reason Knuth used Concrete with Euler is because he wanted to try modifying Computer Modern to see how flexible his meta-design was (he explains so in Typesetting Concrete Mathematics, page 32). An interesting read on font matching is Thierry Bouche's Diversity in Math Fonts. – Philippe Goutet Aug 23 '12 at 7:31

3 Answers

up vote 6 down vote

As you'll probably want a font with at least a bit of similarity to Cambria, I suggest to give Linux Libertine a try.

share|improve this answer
3  
Linux Libertine has a great binding for LaTeX. Check out the libertine package! It works best with XeTeX, and also provides an organic grotesk sans serif font with the same metrics called Biolinum. – FUZxxl Jun 10 '11 at 18:00
up vote 3 down vote

Old question, but I wanted to suggest Marion by the Typodermic Foundry. Unfortunately at least the version which came with OS X doesn't have smallcaps. Here is an example (Marion/Cambria Math, 10/15*27):

enter image description here

(The content is from “Log canonical models for $M_{g,n}$” by Han-Bom Moon, available from arxiv)

share|improve this answer
up vote 3 down vote

Constantia works particularly well, and its small caps & old-style figures are quite nice too. See this document of mine or this detail from it:

constantia plus cambria math

share|improve this answer
You might want to add that we assume a(.,.,.) is locally Lipschitz and h(0,0)=0,g(0,0)=0 :P – percusse Sep 11 '12 at 20:51
:Þ indeed, @percusse. If I ever get back to working on that thesis, maybe. – J. C. Salomon Sep 11 '12 at 21:59

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.