Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!ll-xn!mit-eddie!apollo!roc
From: roc@apollo.uucp (Richard Curtis)
Newsgroups: comp.text
Subject: dimpress
Message-ID: <374c81ab.1054@apollo.uucp>
Date: Tue, 15-Sep-87 17:30:00 EDT
Article-I.D.: apollo.374c81ab.1054
Posted: Tue Sep 15 17:30:00 1987
Date-Received: Thu, 17-Sep-87 06:19:54 EDT
Distribution: na
Organization: Apollo Computer, Chelmsford, Mass.
Lines: 50

My version of dimpress (the thing that AT&T supplies with DWB 2.0 to convert
device independent troff output to imPRESS) does not
output a real inch of vertical space when troff asks for one.

When I run this file

.\"----- start ---
.de T
\l'0.5i'
..
.po 0
.nf
.wh 0i T
.wh 1i T
.wh 2i T
.wh 3i T
.wh 4i T
.wh 5i T
.wh 6i T
.wh 7i T
.wh 8i T
.wh 9i T
.wh 10i T
.wh 11i T
.\"----- end ---

through device independent troff, it should produce a series of horizontal
lines, one inch apart. The raw troff output actually includes the 
right stuff; e.g. a series of vertical moves of exactly the number
of units in an inch, but when I postprocess through dimpress (at 
either 240 or 300 dpi), I get output in which the lines are only .993
inches apart. This may not sound like much but over the length of a page,
it adds up to over 1/16" of an inch.

There is a similar quantum of error in horizontal measurements.

Running the same troff output through Adobe's 
psdit postprocessor yields (utterly) correct spacing.

Any ideas what or where the problem is?

-r

Richard Curtis
Apollo Computer Inc.
330 Billerica Rd., Chelmsford, MA 01824

617-256-6600 x7852

{umix,mit-eddie,uw_beaver}!apollo!roc