[Matplotlib-users] Saving image on Windows problem

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
  <title></title>
</head>
<body bgcolor="#ffffff" text="#000000">
<br>
Has anyone else noticed a problem when saving an image (png) on windows.<br>
Hard to describe, .. but it seems that if there is lots of data to plot
and window size too big, or maximized, it cannot save the full plot
data stopping half way. (plot look fine on window). ....&nbsp; Interesting
reducing the window size can get to the point where can save the full
image.<br>
<br>
Good plot (small window size)<br>
<a class="moz-txt-link-freetext" href="http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/12p5KInternalIR_InputGndGood.png">http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/12p5KInternalIR_InputGndGood.png</a><br>
<br>
Failed plot (larger window size)<br>
<a class="moz-txt-link-freetext" href="http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/12p5KInternalIR_InputGndFailed.png">http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/12p5KInternalIR_InputGndFailed.png</a><br>
<br>
I get the same problem on two different systems:<br>
<b>Windows XP &amp; Python 2.4 &amp; matplotlib &amp; numarray 1.5.2</b><br>
matplotlib installed from matplotlib-0.87.6.win32-py2.4.exe<br>
matplotlibrc has <br>
<blockquote><i>backend&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : TkAgg<br>
numerix&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : numarray&nbsp; # numpy, Numeric or numarray<br>
interactive&nbsp; : False&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; # see <br>
  </i></blockquote>
<br>
<b>Windows 2000 &amp; Python 2.5 &amp; matplotlib &amp; numpy 1.0</b><br>
matplotlib installed from matplotlib-0.87.7.win32-py2.5.exe<br>
matplotlibrc has <br>
<blockquote><i>backend&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : TkAgg</i><br>
  <i>numerix&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : numpy&nbsp; # numpy, Numeric or numarray</i><br>
  <i>#interactive&nbsp; : False&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; # see <a class="moz-txt-link-freetext" href="http://matplotlib.s">http://matplotlib.s</a></i><br>
</blockquote>
<br>
find my data here<br>
<a class="moz-txt-link-freetext" href="http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/ScanDataInternalIR_GndInput.log">http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/ScanDataInternalIR_GndInput.log</a><br>
<br>
and the script here (only tested with numpy and numarray)<br>
<a class="moz-txt-link-freetext" href="http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/irplot24.py">http://members.optusnet.com.au/~steve_geo/MatPlotLibTests/irplot24.py</a><br>
<br>
the command to reproduce the test is <br>
<font face="Courier New, Courier, monospace">irplot24.py --second=fft
--target=600000 ScanDataInternalIR_GndInput.log<br>
<br>
=================<br>
</font>Also it's also interesting to note the difference in load time
(sec) of the data between numarray and numpy for the same data.<br>
<b>Windows </b><b>2000 </b><b>&amp; Python 2.4 &amp; matplotlib &amp;
numarray</b><br>
X:\srg\MatPlotLibTests&gt;c:\python24\python ..\irplot24.py
--second=fft --target=600000 ScanDataInternalIR_GndInput.log<br>
Importing pylab library ........ 1.685<br>
Opening file and loading array .<b> 3.999<br>
</b><br>
<b>Windows 2000 &amp; Python 2.5 &amp; matplotlib &amp; numpy</b><br>
X:\srg\MatPlotLibTests&gt;c:\python25\python ..\irplot24.py
--second=fft --target=600000 ScanDataInternalIR_GndInput.log<br>
Importing pylab library ........ 0.882<br>
Opening file and loading array . <b>10.687</b><br>
<br>
<font face="Courier New, Courier, monospace"><br>
<br>
</font>Thanks for any comments<br>
Steve<font face="Courier New, Courier, monospace"><br>
</font>
</body>
</html>