<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1479" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2><FONT size=2>
<P>Dear Dr. Herbert Voss:</P>
<P>Caution, I am a novice in PSTricks with and/or without your -add package </P>
<P>which appears to be very novice-friendly. I had been playing with a log-ln
</P>
<P>graph example you had presented underthe name " doppellog.tex" As I </P>
<P>play with this example and succeeded in producing an .eps out of it, I </P>
<P>found something surprising. The y-axis in your example .pdf as well as in
</P>
<P>my .dvi, .ps and .eps output files seems to be seriously wrong. The
length </P>
<P>of the y-axis between successive units in a log or ln cycle should become
</P>
<P>progressively smaller as we go from 1, 2, 3, .... 9, 10. The distance between
</P>
<P>1 and 2 is larger than that between 2 and 3, etc. so that in ten steps, </P>
<P>10^(-n) becomes 10^(-[n-1]). The scale in the figure resulting from the
doppellog </P>
<P>example code implies that 10 times 10^(-n) = 10^(-[n+1]), which is obviously
</P>
<P>wrong.</P>
<P>Can you point to me where in the program is the mistake? Or, is it that I am
</P>
<P>altogether in the far-left-field?</P>
<P>Sincerely, Murty Kanury </P></FONT></FONT></DIV></BODY></HTML>