I noticed that in Glyphs 2.2 the TTFautohinting looks different when exporting from Glyphs Export or from a Project. Actually, direct from Glyphs Export it looks the same as in Glyphs 2.1, but from a Project it looks different, even with the same setting. Is it possible that Projects includes already the new TTFautohint 1.4, but Glyphs Export doesn’t?
interested to know the answer to this.
No, both use the same new version. I’ll had a look and the normal and project export uses the same parameters. Can you send me the fonts and HTML text documents?
Just sent you the files.
Thanks for the files. I figured it out. You need to add a “Save as TrueType” custom parameter, otherwise it would be CFF based.
ok, good to know. Thanks!
BTW: Which Version of TTFautohint is Glyphs 2.2 using?
PS: CFF based Webfonts seem to look quite good by now. Tested on Win 7.
the latest version 1.4
If you don’t look at really mall sizes. And it is even better in Windows 10. And on older windows versions, it looks even better in bigger sizes.
Also in small I think they are not that bad anymore. Similar to Mac, because bolder. I wonder if it would be a strategy to make Webfonts thinner and in CFF, so they would look quite equal on Mac and PC!?
If you find the result exactable that could work. But remember that windows users are used to the crisper rendering. When Safari came out for windows it used the same rendering as on the mac and a lot people where annoyed by that.
When generating CCF based Webfonts it decomposes the components, right? Wouldn’t it be better to keep the components to get a smaller file size? Or is it generally a bad idea to have CCF based fonts with components? (Or maybe does Postscript hinting not work in composites?)
There are no components in CFF (there are but they can’t be used for most glyphs). And the subroutines compress the font quite nicely.