Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!samsung!caen!hellgate.utah.edu!dog.ee.lbl.gov!ux1.lbl.gov!osborn From: osborn@ux1.lbl.gov (James R Osborn) Newsgroups: comp.sys.mac.programmer Subject: Soft Fonts Summary: Are Soft Fonts here yet? Keywords: fonts, greyscale Message-ID: <13376@dog.ee.lbl.gov> Date: 19 May 91 17:09:35 GMT Reply-To: osborn@ux1.lbl.gov (James R Osborn) Distribution: all Organization: Lawrence Berkeley Laboratory Lines: 42 X-Local-Date: Sun, 19 May 91 10:09:35 PDT I ran across a (very old) posting that I saved from a few years ago by - Oliver Steele - at Apple. The post was about Soft Fonts. For those of you who don't know what these are, essentially they are antialiased fonts. They use two or more bits of screen depth to represent characters which are claimed to be vastly more legible and comfortable on the eyes - especially with small sizes (like 9 pt). Supposedly the soft fonts increased reading speed (even over hardcopy) and reduced fatigue - I believe it. There were two technologies mentioned for generating these. First, you could convolve them from bitmaps with higher resolution. For example you could take the bitmap for an 18 pt font and apply a process to create a 2 bit deep 9 pt greyscale soft font from this. The second approach was to use a scan conversion algorithm directly with the outline for the font. Well, the latter course is preferred because it will generate more acurate representations of the font, and because it is probably faster. Well? We *HAVE* TrueType now! Wouldn't it be neat if you had an INIT (ah...ahem...extension) that would patch into the text rendering routines in the system and substitute a soft font when drawing to a color screen? This hypothetical INIT could use TrueType outlines and generate the soft version on the fly using scan conversion and then perform the usual CopyBitsing to the screen. My eyes would thank this INIT if it existed! Anybody working on such a beast? I'd pay money for it...hell I'd even *WORK* on it, it sounds so neat! -- James .------------------------------.--------------------------------------. | James R. Osborn | It just goes to show you it's always | | Lawrence Berkeley Laboratory | something. Either it's unclear tech | | osborn@ux1.lbl.gov | notes or your mac is smoking. It's | | (415) 548-8464 | always something... | '------------------------------'--------------------------------------'