[net.micro] TEX on PC

jds@duke.UUCP (Joe D. Sloan) (07/09/85)

	The July 1985 Dr. Dobb's Journal has an ad for TEX for
the IBM-PC?  Has anyone used it or does anyone know more?  

				Joe Sloan
				duke!jds

nather@utastro.UUCP (Ed Nather) (07/11/85)

> 	The July 1985 Dr. Dobb's Journal has an ad for TEX for
> the IBM-PC?  Has anyone used it or does anyone know more?  
> 
> 				Joe Sloan
> 				duke!jds

There are two different implementations of TeX for the IBM PC -- the one
advertized in DDJ and one implemented by David Fuchs, available from
Addison Wesley for $495 (less 10% educational discount to schools). I
have this version running on my PC XT.  I have not tested it thoroughly,
and the "System Guide" manual is stamped "Preliminary Uncorrected Proofs"
so it appears to be a "pre-release version."  It works, and provides
eye-popping output on an FX 80+ dot matrix printer (at 240 dots/inch).
First quality output is very slow (6 passes of the print head per printed
line) but the results are spectacular, particularly for complex equations.

The program requires 512K or memory (640K is better -- the manual implies
it's a tight fit in 512K) and a hard disk (fonts take up about 4 MB).  TeX
loads in about 10 seconds, and chews through a demo math page in about 5.
According to Fuchs, an XT can set the whole TeXbook (500 pp) in a bit over
4 hours (an AT takes less than 2).  A driver is included in the package
that runs IBM Matrix printers, IBM Graphics printers, or the Epson FX, MX 
and RX printers.  The output files from the program (called MicroTeX) are
TeX-standard .dvi files than can be typeset on any systems you have a
driver for.

The Epson driver permits a "quality" adjustment -- lower quality output for
proofreading is quite respectably fast -- and the "plain" TeX macros
are pre-loaded with the TeX processor.  Outstanding quality is possible
with a bit of fiddling:  if you magnify your document by 40% before
processing, then photo-reduce your resulting printout by about the 
same amount, the effective resolution is improved enough so it rivals
output from a LaserWriter.

I have no experience with the other TeX implementation, and probably
won't ever have -- I'm sold on this one.

Usual disclaimer: I've never met David Fuchs and wouldn't know Addsion
Wesley if I met him in a parking lot.

-- 
Ed Nather
Astronomy Dept, U of Texas @ Austin
{allegra,ihnp4}!{noao,ut-sally}!utastro!nather
nather%utastro.UTEXAS@ut-sally.ARPA