home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.palmtops
- Path: sparky!uunet!spool.mu.edu!agate!ames!nsisrv!nssdcs.gsfc.nasa.gov!williams
- From: williams@nssdcs.gsfc.nasa.gov (Jim Williams)
- Subject: Re: Whats the smallest (readable) 24x80
- Message-ID: <1993Jan21.200756.28157@nsisrv.gsfc.nasa.gov>
- Sender: usenet@nsisrv.gsfc.nasa.gov (Usenet)
- Nntp-Posting-Host: nssdcs.gsfc.nasa.gov
- Organization: NASA Goddard Space Flight Center, Greenbelt, Maryland
- References: <1993Jan20.030014.9216@crl.com>
- Date: Thu, 21 Jan 1993 20:07:56 GMT
- Lines: 46
-
- [Apologies if you see this twice. It doesn't seem to have gotten out
- the first time.]
-
- In article <1993Jan20.030014.9216@crl.com> jc@crl.com (Jim Couch) writes:
- >Subject line says it all. The hp95 looks like it would
- >be awfully crammed to squeze in 24x80. Is there anything
- >between the smallest full size laptop and the hp which
- >would be both small and readable?
-
- "Readable 24x80" is a matter of some opinion. Others have posted
- articles here describing highly squeeed text (*three* pixels/character
- wide!) that manages to get an 80x24/25 display onto the hp95.
- Personally, I just can't read it. My own minimum is a 5x7 character
- in a 6x8 cell (just what the hp95 uses). Thus, you would need at
- least a 480x200 display to get 80x25 characters. The old, standard,
- CGA display, at 640x200 was able to do slightly better by using an 8x8
- cell. An 8x10 or 8x12 cell makes underlining much nicer and keeps the
- descenders on one line from touching the tops of tall letters below.
- This happens on the hp95 a lot. I don't know of any machine with a
- display that is significantly better than the hp95 (in terms of size),
- but is still not a full CGA. I personally wouldn't bother with
- anything less than CGA if you really want 80x25 text screens.
- Machines like the ZEOS use CGA and the screen on that is quite
- readable, although I think the hp95 screen is better.
-
- If I was going to cram the maximum text onto an hp95 screen, I'd use a
- carefully crafted, san serif variable width font, with each character
- as narrow as possible, while still having a unique bit pattern. I
- even considered how this could be done "adaptively". Rather than
- having a width table for each font, you take each character in turn
- and create a mask that is the bitwise OR of all the rows. The
- leftmost and rightmost 1's in this mask will give you the left and
- right most pixels in that character. You only have to do this once
- per character. The mask, combined with some bit-shifting, can be used
- to render the character onto the screen so that it comes right up
- against the previous character, with just a single pixel between them.
- Even more elaborate algorithms can save pixels via kerning. But I
- suspect the horsepower needed for any of this adaptive stuff would
- make text rendering too slow. I may try it at some point...
-
- Jim
- --
- Spoken: James W. Williams Company: Hughes STX
- Internet: williams@nssdcs.gsfc.nasa.gov Phone: +1 301 286-1131
- USPS: NASA Goddard Space Flight Center; Code 633; Greenbelt, MD 20771
- Maryland Car Tags: DEV CAR
-