mdeck@sybil.cs.Buffalo.EDU (Mary Deck) (08/18/90)
Ok. So here I am, plugging along, building all sorts of fonts. I get
to the AMS symbol fonts. The first several go without a hitch. Then
I get to MSBM5 and MSBM6. They won't work. They bomb on "The Letter
M" (which is defined in XBCAPS.MF). I would have though that I had
made a mistake somewere, except that the very-similarly-defined MSBM7,
MSBM8, MSBM9, and MSBM10 fonts went through Metafont without incident.
This bothers me, though, because I haven't heard anyone else complain
of this problem. ?!?!
Following is a chunk of the .lis file....
This is METAFONT, Vax/VMS Version 1.7 (preloaded base=cmxerox 90.8.2) 17 AUG 1990 18:59
**&cmxerox \blacker:=0.5; fillin:=0.0; input msbm5
(TEX_ROOT:[MF.AMS.SYMBOLS]MSBM5.MF;1 (TEX_ROOT:[MF.AMS.SYMBOLS]AMSYB.MF;1
(TEX_ROOT:[MF.AMS.SYMBOLS]BSYMBOLS.MF;1
...
(TEX_ROOT:[MF.AMS.SYMBOLS]XBBOLD.MF;1 (TEX_ROOT:[MF.AMS.SYMBOLS]XBBASE.MF;1)
(TEX_ROOT:[MF.AMS.SYMBOLS]XBCAPS.MF;1
...
The letter M
! The paths don't intersect.
<to be read again>
;
intersectionpoint->...s don't intersect");
origin.else:0.5[point.x_.of...
<to be read again>
;
l.379 ...rsectionpoint ((x3l,0)--(x3l,h));
?
Any suggestions?
Also, the SliTeX ILASY8 font is giving me problems. It runs through
Metafont OK, but when I try to run gftopk on the resulting file, I get
an error like "Empty GF file." I realize that this is an invisible
font, but how can I convince gftopk to give me an invisible pk file
(like with the other invisible fonts)?
By the by, I got the mf sources from ymir.claremont.edu. The mode I'm
using is this:
mode_def xrx =
proofing:=0;
trtaceingtitles:=0;
fontmaking:=1;
pixels_per_inch:=300;
o_correction:=.5;
blacker:=0.5;
fillin:=0.0;
enddef;
mode=xrx;
Loads of tanks....
.....Mary
--
Mary M. Deck, Student Assistant, Software Development
University Computing Services, State University of New York at Buffalo
mdeck@cs.buffalo.edu [Unix]
acsgmmd@ubvms.cc.buffalo.edu [VAX/VMS]edgar@function.mps.ohio-state.edu (Gerald Edgar) (08/18/90)
In article <33055@eerie.acsu.Buffalo.EDU> mdeck@sybil.cs.Buffalo.EDU (Mary Deck) writes: > >Ok. So here I am, plugging along, building all sorts of fonts. I get >to the AMS symbol fonts. The first several go without a hitch. Then >I get to MSBM5 and MSBM6. They won't work. They bomb on "The Letter >M" (which is defined in XBCAPS.MF). Just hit return and go on. The font seems OK. These are VERY tiny letters, so how can you tell if they are a bit weird? -- Gerald A. Edgar Department of Mathematics Bitnet: EDGAR@OHSTPY The Ohio State University Internet: edgar@mps.ohio-state.edu Columbus, OH 43210 ...!{att,pyramid}!osu-cis!shape.mps.ohio-state.edu!edgar
mdeck@acsu.buffalo.edu (Mary Deck) (08/21/90)
In article <1990Aug18.125641.12466@zaphod.mps.ohio-state.edu> edgar@function.mps.ohio-state.edu (Gerald Edgar) writes: > >Just hit return and go on. The font seems OK. These are VERY tiny letters, >so how can you tell if they are a bit weird? That will work, but it's only a fudge, and I don't like fudges. (Well, this kind, anyway. :) Being a perfectionist, if it's not right, it isn't ok. The fact that I even get this error indicates that something is amiss somewhere. Who do I talk to about this? Is there a real solution? ....Mary -- Mary M. Deck, Student Assistant, Software Development University Computing Services, State University of New York at Buffalo mdeck@cs.buffalo.edu [Unix] acsgmmd@ubvms.cc.buffalo.edu [VAX/VMS]
wilcox@blob.cis.ohio-state.edu (Patricia P Wilcox) (08/21/90)
In article <33186@eerie.acsu.Buffalo.EDU> mdeck@acsu.buffalo.edu (Mary Deck) writes: >In article <1990Aug18.125641.12466@zaphod.mps.ohio-state.edu> edgar@function.mps.ohio-state.edu (Gerald Edgar) writes: >> >>Just hit return and go on. The font seems OK. These are VERY tiny letters, >>so how can you tell if they are a bit weird? > >That will work, but it's only a fudge, and I don't like fudges. >(Well, this kind, anyway. :) Being a perfectionist, if it's not >right, it isn't ok. > >The fact that I even get this error indicates that something is amiss >somewhere. Who do I talk to about this? Is there a real solution? The "strange turning angle" message means that Metafont is having to go clockwise instead of counterclockwise around some part of a filled path. We used to get several of these messages when generating write-white fonts, before I fine-tuned our printer mode definitions. An overly large value of "blacker" might, for example, cause the edge of a stroke to move from the right to the left side of some critical intersection point of other strokes in the character. Usually you're just getting tiny clockwise loops at the corners of larger filled areas. A cw loop is erased instead of filled, so nothing much will show up if it's 1) tiny or 2) not superimposed on a previously- filled area. The ultimate solution might be to recode the .mf character definitions for write-white devices (changing mode definitions is only an approximation to a correct fix for printer characteristics -- for some letters on some printers, it's not a very good approximation). For example, math characters drawn with lines of width "rule-thickness" are uniformly too light when everything else is perfect on my printer, and I may one of these days throw "define_blacker_pixels(rule_thickness);" into a copy of cmbase.mf and see whether that helps. None of this stuff is magic; it just seems that way. 8-) --Pat Wilcox (wilcox@cis.ohio-state.edu)
dhosek@hmcvax.claremont.edu (Hosek, Donald A.) (08/21/90)
In article <33186@eerie.acsu.Buffalo.EDU>, mdeck@acsu.buffalo.edu (Mary Deck) writes... >In article <1990Aug18.125641.12466@zaphod.mps.ohio-state.edu> edgar@function.mps.ohio-state.edu (Gerald Edgar) writes: >>Just hit return and go on. The font seems OK. These are VERY tiny letters, >>so how can you tell if they are a bit weird? >That will work, but it's only a fudge, and I don't like fudges. >(Well, this kind, anyway. :) Being a perfectionist, if it's not >right, it isn't ok. >The fact that I even get this error indicates that something is amiss >somewhere. Who do I talk to about this? Is there a real solution? The fact of the matter is that many MF programmers, such as they are, get their code working for their output device at maybe one size and consider their work done. I have a mental list of some not-so-borderline cases that break a lot of common fonts (including some really odd cases, e.g., one font works at 300dpi but breaks at 2470). Knuth's CM is one of the more robust faces and even that has problems. In the case of the blackboard bold letters dying on your MF run, I'd bet large quantities of cash that the problem is due to the largish blacker value of 0.5. The best solution is to grab the new VMS MF change file (yes, Mary, it's FINALLY ready) and run the files using MF/BATCH, which will use \batchmode when processing so those errors will just be skipped over. and if you really want to complain, write to tech-support@math.ams.com -dh --- Don Hosek TeX, LaTeX, and Metafont support, consulting dhosek@ymir.claremont.edu installation and production work. dhosek@ymir.bitnet Free Estimates. uunet!jarthur!ymir Phone: 714-625-0147 finger dhosek@ymir.claremont.edu for more info