[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: more TeX font stuff
On Tue, 5 Feb 1991 18:08 PST Don Hosek said:
>Regarding mode_defs, I've been thinking about putting together a
>mode_def subdirectory on the ymir archive. I'd be glad to have
>you do the work of creating the files. My initial thought was to
>include each mode_def twice: once in its own file in a format
>suitable for smode (i.e., sans the mode_def..enddef wrapper) and
>once in a collected file. The former would be useful for the poor
>sods who don't have the necessary privs to add a mode_def to the
>On write_white devices, multiple mode_defs seem to be the best
>way to handle getting the best output. Some work was done at
>Texas A&M a few years back (in conjunction with the TeXrox
>development) but those who did the work are all gone to
>countries bordering the Indian Ocean and haven't been heard from
>in quite a while.
Not entirely true: I got a letter from Glenn Vanderburg the very day you sent
this note. ;-)
> Tom Reid should be able to mail something on
>this topic to us so I've included him in my list of recipients.
The work that Don refers to consisted of adjusting the blacker and fillin
values for each of the Computer Modern fonts. This task was done by Medhi
Widjaja (who has since graduated and returned to Indonesia and now works
in Jakarta). Medhi came up with a list of blacker/fillin pairs for nearly
all of the CM fonts at magstep0. Experiments showed that the same values
applied rather well for non-zero magsteps. He actually ran two sets of
blacker/fillin pairs; one for our Xerox 4050 printer, and another for our
9700s. Both are writes-white engines, but they use different fusing
technologies and have different speeds (50 ppm v. 120 ppm).
Glenn Vanderburg's (who has since graduated and moved to Perth, Australia)
part in this was to take the list of pairs which Medhi came up with and
define a set of modedefs. Of the complete set of pairs, Glenn defined ten
to twelve different modedefs; altering some values slightly to "round" to
the nearest selected modedef. We then regenerated the complete set of fonts
using these modedef values.
Medhi subsequently wrote a UNIX shell script and a VM/CMS EXEC file (based
on Don Hosek's MFTOOL EXEC, I think) to read a list of font names, magsteps,
blacker, and fillin values then run METAFONT to build the fonts. This
process uses a single modedef with fairly generic values since the actual
values for blacker and fillin are supplied by the script/EXEC.
The is work presently being done at the State University of New York at
Stony Brook to tune the individual fonts for the Xerox 4650. This is a
50 ppm printer with the same fusing technology as the 4050, but with 600
dots per inch. I have added Len Boyle to the list of recipients to comment
on the progress on that project.
If anyone is interested in seeing any of the work, let me know and I'll
see what files are still around.