Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!usc!brutus.cs.uiuc.edu!apple!sun-barr!newstop!sun!angel!henry
From: henry@angel.Sun.COM (Henry McGilton -- Software Products)
Newsgroups: comp.lang.postscript
Subject: Re: 72.27! (was Re: ruler.ps - an inch/point ruler of your very own)
Summary: Laser Writer Is A Consumer Product
Message-ID: <130414@sun.Eng.Sun.COM>
Date: 18 Jan 90 02:03:35 GMT
References: <8696@dime.cs.umass.edu>
Sender: news@sun.Eng.Sun.COM
Lines: 50

In article <8696@dime.cs.umass.edu>, rudenko@cs.umass.edu writes:


    *  I tried drawing a vertical line 10 inches long on a
    *  (300 dpi) DEC LN03R via:

    *  	0 setlinewidth   % thinnest possible line
    *  	newpath
    *  	100 10 moveto
    *  	0 720 rlineto stroke    % 10 inches * 72 pts per inch

    *  I measured the length of the resultant line with a
    *  steel rule and got a length of about 9 31/32 inches.

    *  The resultant line turned out to be just a hair over 10
    *  inches long!  Sending the file to the printer a second
    *  time produced lines of 9 61/64 and a hair short of 10
    *  inches, respectively.  Sending the file to the printer
    *  a third time yielded lines a hair short of 9 31/32 and
    *  as close to 10 inches as I could measure, respectively. 

    *  This experiment seems to indicate that the size of
    *  PostScript point is apparently equal to the printer's
    *  point, 1/72.27 inch, and not 1/72 inch as claimed by Adobe!?

1/32 of an inch in 10 inches is an error of 0.3 per cent.  The paper
feed mechanism on the printer, the laser tracking machinery,
the stretching of the paper due to the heat of the fusing and the
compression of the feed rollers, plus the humidity, plus probably
the phase of the moon and the colour of the printer room door,
can easily account for an error of 0.3 per cent.  Laser printers
of the DEC and Apple type are not precision devices.

I just took a page off a LaserWriter IINTX that has a
registration error of around 1/16 inch at the moment -- it's
definitely in need of repair.

I suggest repeating the experiment on a 2540 dpi
Linotronic.  But first, replace your  0 setlinewidth  to
something more rational. At 2540 dpi, the thinnest line is
4/10,000 inch, and you probably won't be able to see it.
The 0 setlinewidth construct to get the thinnest line is
definitely a very device dependent thing to do and creates
inherently non portable PostScript.

	........... Henry
+-------------------+---------------------------+---------------------------+
| Henry McGilton    | I'll bet those people who |                           |
| Sun Microsystems  | put control-D characters  | arpa: hmcgilton@sun.com   |
| 2550 Garcia       | in PostScript files also  | uucp: ...!sun!angel!henry |