problem with imshow and ps+xpdf backend

Hi All,
I have a problem with the output of the ps backend when I use the xpdf
distiller rather than ghostscript and try to save a figure from
imshow. When I run the script below I get a blue square with
ghostscript (as expected), but grey dots and lines with xpdf. With
non-image based plot commands I get normal (actually very nice)
results with xpdf. Any ideas? Thanks for any help!

Script:

···

=============
import scipy
import matplotlib
matplotlib.use('PS')
matplotlib.rc('ps', usedistiller='ghostscript')
#matplotlib.rc('ps', usedistiller='xpdf')
import pylab

a = scipy.ones((50,50))
pylab.imshow(a)
pylab.savefig('tplot.eps')

Machine details:

xpdf version 3.02
GPL Ghostscript SVN PRE-RELEASE 8.61 (2007-08-02)
scipy 0.5.2
python 2.5.1
matplotlib 0.90.1
x86_64 linux
Ubuntu Gutsy (beta) AMD64 version

These reports almost always turn out to be a problem with one of the external
dependencies. I see no difference between the two results on my machine.

pdftops-3.00
GPL Ghostscript 8.60
python-2.5.1
svn mpl
Gentoo ~amd64

···

On Tuesday 16 October 2007 10:00:55 am John Travers wrote:

Hi All,
I have a problem with the output of the ps backend when I use the xpdf
distiller rather than ghostscript and try to save a figure from
imshow. When I run the script below I get a blue square with
ghostscript (as expected), but grey dots and lines with xpdf. With
non-image based plot commands I get normal (actually very nice)
results with xpdf. Any ideas? Thanks for any help!

Script:

import scipy
import matplotlib
matplotlib.use('PS')
matplotlib.rc('ps', usedistiller='ghostscript')
#matplotlib.rc('ps', usedistiller='xpdf')
import pylab

a = scipy.ones((50,50))
pylab.imshow(a)
pylab.savefig('tplot.eps')

Machine details:

xpdf version 3.02
GPL Ghostscript SVN PRE-RELEASE 8.61 (2007-08-02)
scipy 0.5.2
python 2.5.1
matplotlib 0.90.1
x86_64 linux
Ubuntu Gutsy (beta) AMD64 version

You are right. I removed xpdf-utils from my machine and installed
poppler-utils instead (which on Ubuntu conflicts with xpdf-utils). The
result is now correct.
Thanks for your help!!
John

···

On 16/10/2007, Darren Dale <darren.dale@...163...> wrote:

These reports almost always turn out to be a problem with one of the external
dependencies. I see no difference between the two results on my machine.

pdftops-3.00
GPL Ghostscript 8.60
python-2.5.1
svn mpl
Gentoo ~amd64