Chinese, again

gabriel00's picture

Hello, this is my third post about chinese fonts, sorry about that, but I've just worked on a project in mandarin and, although the project is finished, am still looking into it.

I had some problems embedding Mac OSX fonts into PDF so I looked around a lot and found several complaints concerning the Heiti fonts (SC and TC) that are now standard in OS 10.6. This link is the best example

Does anyone have an opinion on this? I can't read a word (or ideogram) in chinese so I really don't see the flaws. I am concerned that the font may be flawed and am trying to find out if it is safe to use it in text meant for chinese readers.

Your input will be greatly appreciated.

Gabriel Attuy

ahyangyi's picture

According to Chinese Wikipedia, Heiti TC is fixed by Mac OSX 10.6.3.

gabriel00's picture

Very true, this site sheds some light on the subject [translated from chinese]

wongxiao's picture

lol, "black body!"

Anyway, there are some very noticeable differences in the example you posted. But first, let me clarify that I am NOT a native Chinese speaker. Nor am I anywhere near fluent.

In the second link you posted:

- The stroke width consistency seems improved (look at the bottom of "zhe," 這 the third character from the left).
- Also, the top stroke of the speech radical 言 now stands upright (left hand side of the first character, also seen in the third character).
- The bottom radical of zero 零 (I forget what this radical is called) also has a more modern form.
- The radical 示 now takes on a different form when inside a character (see the left hand side of the second character).
- The silk radical 糸 is now also different when positioned in the left of a character (see the seventh character).
- The food radical 食 is also different (see the sixth character).
- If you look at the feather radical 羽, it is more modern.

There other differences as well. I'm fairly certain that any person fluent in Chinese will be able to read either version, but if they're picky the second version is probably better. I find it more aesthetically pleasing, in any case.

I think the most offensive problem is that there seem to be inconsistent radicals. Chinese characters are made up of "Radicals," which are smaller components. Some radicals have different ways of being written, and some of them change when they become part of another character, as opposed to standing alone. Some radicals have changed shape in the past few decades.

If you look at 請, 情, 清, 啨, and 晴, the right half of these should all be the same. If you look at the lower-right hand side, its different from character to character in the previous version. The first link you provided highlights this (and other problems) nicely.

Also, note that a lot of these are only for traditional characters; I'm not sure what changes would have been made to the simplified character set. Many of the problems wouldn't apply (for example, the bottom three strokes of the silk radical become a single line: 纟)

ahyangyi's picture

Heiti TC is traditional. The simplified version is Heiti SC.

lunde's picture

What is being discussed are what can be best described as Pan-Chinese fonts that were introduced in Mac OS X 10.6. They are Pan-Chinese because they are TTC (TrueType Collection) fonts that advertise two font instances, one for Simplified Chinese, and a second for Traditional Chinese. The same fonts, though with Japanese and Korean fonts instances, are included with the iPhone, and those are Pan-CJK fonts.

These two STHeiti fonts, Light and Medium, stem from two similarly-named Simplified Chinese fonts. The glyphs for the Simplified Chinese font instances follow GB 18030, which is effectively the URO and Extension A in terms of the ideographs. There are also 6,000 or so glyphs for Extension B code points, mainly for Hong Kong SCS coverage.

I know the source of the glyphs, and I also know many of the issues surrounding the design and development of Pan-CJK fonts. One design issue that affects any Pan-CJK (or Pan-Chinese) font implementation is how to handle the locale-specific glyphs code points that do not have a source for that locale, meaning that the ideograph is not used. There are two choices. One is to simply not design a glyph for that locale, meaning that if used, the design may be inappropriate for that locale. The other choice is to design a glyph for that locale, and depending on whether there is an adequate source or reference for that ideograph for that locale, the designer may need to extrapolate.

In any case, I consider the ST Heiti typeface design to be first and foremost a Simplified Chinese one. The Traditional Chinese glyphs needed to be designed. Of the 30K or so CJK Unified Ideograph code points that are covered, approximately one-third have forms that are specific to Traditional Chinese.

I could go on and on...

Syndicate content Syndicate content