|
Re: TTC fonts: msg#00409text.unicode.devel
On Wednesday, April 30, 2003, at 02:27 AM, Raymond Mercier wrote: At least this seems to settle in the negative the question as to whether they all have the same unicode blocks. TTC fonts and their rationales are discussed at some length on Microsoft's typography Web site (<http://www.microsoft.com/typography/>, specifically <http://www.microsoft.com/typography/otspec/otff.htm>). Theoretically, a TTC font can contain an indefinitely large number of fonts, since the numFonts field is 32 bits. In practice, more than three or four is unlikely. As explained in the spec, the purpose is to allow large fonts to overlap their data; if you have two Japanese fonts, say, one with proportional Roman and one with monospaced, but identical kanji. Then the huge 'glyf' table can be shared by the two with the smaller tables (such as 'cmap' and 'name') used to distinguish them. The result is a single file which contains two fonts but is considerably smaller than the two individual fonts would be as two files. One could also use TTC files to store multiple related fonts even without overlap. If I had a roman, bold, italic, and bold-italic, I could put all four into a single TTC file and have them therefore ship as a single file, rather than four individual files. ========== John H. Jenkins jenkins@xxxxxxxxx jhjenkins@xxxxxxx http://www.tejat.net/
|
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: Private Use Area, Peter_Constable |
|---|---|
| Next by Date: | Re: Private Use Area, Jungshik Shin |
| Previous by Thread: | Unicode 4.0 and SQL, Ake Persson |
| Next by Thread: | C string literals with 16-bit Unicode, Markus Scherer |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |