<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Stephane <br>
<br>
<br>
<blockquote
cite="mid:OF9E08EC86.099FE2FA-ONC12573FE.005675F7-C12573FE.00570B27@notes.edfgdf.fr"
type="cite"><font face="sans-serif" size="2">Do you have an easy way
to measure the
performance in ParaView ? (is there a way to set up a timer? )</font>
<br>
<font face="sans-serif" size="2">By the way, I had another question
on
stream tracing : do you know why can't we use cell-based velocity field
for stream tracing ? </font>
<br>
</blockquote>
<br>
I'm not really sure how reliable the render timing on the help menu
would be, to display this. We could add a timer.start and timer.elapsed
call at the begining and end of RequestData in vtkStreamtracer which
will output the time in the output window of paraview, but that would
mean temporarily adding this to cvs. No real problem there. I was
hoping someone else would know a good way. <br>
<br>
Cell based velocity is not used because the interpolated value of
velocity remains constant within the cell, so you'd get small line
segments that were straight as the particle moved through the cell and
into the next. It could be changed to find the neighbour cells and do a
kind of on the fly cell-point interpolation and use those values, but
frankly it'd be inefficient and a bit messy and life is too short when
you can use cell-point filter before hand...<br>
<br>
JB<br>
<br>
</body>
</html>