Re: Font problem or not?

[George White]

On Thu, 31 Oct 2002, Staszek Wawrykiewicz wrote:
 
 George N. White III [EMAIL PROTECTED] writes:
  I find the current map file mechanism problematic, especially when using
  non-std fonts, because you can't tell from the TeX source what fonts are
  really being used.
 
 It is not TeX problem but final output/device dependent. For TeX we use
 only .tfm files.

So you can have dvips fail to find the font that corresponds to the 
.tfm file (missing entry) or end up with different fonts being used
on different systems (e.g., Adobe vs URW, MathTime vs Belleek, etc).

 If (for some reason) we have two different map files, we can always declare
 them in config.ps and pdftex.cfg, respectively (p +mymap.map).

You can, but if you take the file to another system or mail it to someone
else, will you remember to change the .map files?  If you use pdftex's
ability to load map files specified in the .tex file, it is at least clear
which fonts you intended to use, and it will be obvious if the required
map files aren't present.

--
George White [EMAIL PROTECTED] 
Head of St. Margarets Bay, Nova Scotia

new dvips

[Dimitri Antoniou]

 Hi,

 I downloaded and compiled version 5.92 of dvips.
 When we say that the character shifting issue has been corrected,
 do we mean that
  dvips -Ppdf ...   without either -G or -G0, 
 now works for computer modern, adobe postscript, and combinations of them?
 Is that right? 

 Dimitri

Re: new dvips

[Thomas Esser]

  When we say that the character shifting issue has been corrected,
  do we mean that
   dvips -Ppdf ...   without either -G or -G0, 
  now works for computer modern, adobe postscript, and combinations of them?
  Is that right? 

It means that -G1 (as set e.g. via -Ppdf) does no longer produce
incorrect output.  The remapping of characters works for computer modern,
but it does not work for a lot of other fonts. The new dvips decides on
a per-font basis whether to remap it or not and it only remaps if that
won't mess up the result.

Thomas