## AMS Euler compared with Computer Modern

I've been meaning to put up this sort of comparison for a quite some time, so that it is perhaps clearer why I have been thinking about designing a symbol font to serve as a better companion to AMS Euler.

Above is an excerpt of a side by side comparison of AMS Euler and Computer Modern Italic (which is the Computer Modern math font) that I just prepared. For the most part it is hopefully clear that AMS Euler has a generally darker »color« and more »robust« stems. Now consider a comparison of an inference rule not unlike would be seen in many papers I read or write.

The important points to note here are that currently, many important symbols are used identically regardless of which typeface has been chosen. Therefore, despite AMS Euler's darker color the turnstile, equivalence symbol, up-arrow, and »forall« quantifier remain the same »weight«. It is interesting to note that the up-arrow is ever-so-slightly wider for AMS Euler though. Additionally, in this context it feels like the universal quantifier should have a wider spread when used with AMS Euler. The star probably is fine as is, but it might even be worth providing a heavier version.

Anyway, I hope this makes things clearer. However, if you disagree with my assessment, and feel that the differences are insignificant, let me know that too.

## kitby said,

December 4, 2006 @ 11:50 pm

Having seen both Computer Modern and AMS Euler in print and on screen (paper preview size and “presentation” size), I say the differences are noticeable, namely that Euler is easier to read. I more holistic comparison would also be useful, say, a paragraph with both plain text (including a pangram of some sort, even if it’s a meaningless phrase), inline math, and “display” math (to use LaTeX terminology). I guess if I were choosing what fonts to use, I’d be making a comparison based on some particular context. (At some point, I used to choose completely different fonts for presentations than I did papers.)

## oxlahun said,

December 5, 2006 @ 10:43 am

I had a pretty good idea what you were talking about, but the example definitely strengthens it.

Turnstile and equivalence bother me the least of the five imported symbols, possibly because it’s easiest for me to think of those as purely geometric. Yeah, they could use a little more weight, but it’s not as jarring as up-arrow and forall. The star, curiously, is the one that looks completely out of place to me.

So what’s the plan? Your post the other day expressed, if I may generalize, some uncertainty about the simple geometric forms. What are you doing about the complex geometrics (forall and up-arrow) and the nongeometrics like the star?

## Joshua Dunfield said,

December 6, 2006 @ 6:20 am

I’d definitely like to see this fixed. The anorexic turnstile, \oplus, etc. have bothered me enough that I’ve used “splat bold”, which just “splats” a second copy of the symbol shifted slightly from the first one:

\newlength{\zzsplatboldwidth}

\newcommand{\xsplatbold}[2]{\settowidth{\zzsplatboldwidth}{{#2}}{#2}\addtolength{\zzsplatboldwidth}{-#1}\hspace{-\zzsplatboldwidth}\raisebox{#1}{{#2}}}

e.g.:

\newcommand{\splatbold}[1]{\xsplatbold{-0.04mm}{#1}}

\newcommand{\splatoplus}{\splatbold{\oplus}}

This is pretty hideous, but hideous beats unreadable-from-the-back-of-the-room. Obviously I’d prefer to have properly designed symbols.

## wkpark's me2DAY said,

January 17, 2009 @ 12:57 pm

영구의 생각…AMS Euler compared with Computer Modern…