I would be in favor of either taking the last time step, or preferably using 
all of them.  If you take a subset of any chaotic event, you will miss spikes.

Alan

________________________________
From: [email protected] [mailto:[email protected]] On 
Behalf Of Renato Elias
Sent: Thursday, May 14, 2009 10:29 AM
To: Moreland, Kenneth
Cc: [email protected]
Subject: Re: [Paraview] Color bar rescaling for transient datasets

What about interpolating the ranges in a regular number of time steps? For 
instance, let's say we have a dataset with 100 time steps. Instead of loading 
all files to find the correct range, we could have an estimative by computing 
the maximum and minimum values using only 20 or 50 time steps. Maybe the user 
could choose how many time steps should be skipped.

more two cents :oP

Renato.



On Thu, May 14, 2009 at 1:15 PM, Moreland, Kenneth 
<[email protected]<mailto:[email protected]>> wrote:
I am told that EnSight find the range by loading the range from the last time 
step, which is where we got the idea.  It might be more to it than that.  
Perhaps if the file format stores ranges over times it will load that and you 
data follows that?

-Ken



On 5/14/09 9:00 AM, "Renato Elias" 
<[email protected]<http://[email protected]>> wrote:

I wonder know what's the magic behind Ensight. It loads the transient data 
range almost instantaneously. We just set a variable to paint the model and the 
color bar appears with a reasonable range for the transient data.

Ken, setting the range by the last time step is a bit dangerous isn't it? For 
example, in my case, I'm coloring a dam break fluid flow model by the velocity 
magnitude. In this problem we stop the simulation after the fluid rests in a 
static position. Therefore the last time step velocity should be almost zero.

Renato.

On Thu, May 14, 2009 at 10:33 AM, Moreland, Kenneth 
<[email protected]<http://[email protected]>> wrote:
This issue has come up quite a bit.  We are trying to think of a way to set the 
range reasonably while avoiding expensive loading and maintaining consistent 
behavior.  One idea we are considering is to set the range based on the last 
time step, which often has a range close to the full range.

-Ken



On 5/14/09 7:28 AM, "Chris Kees" 
<[email protected]<http://[email protected]> 
<http://[email protected]> > wrote:

I would be interested in a simple way  to do this too.  It seems like a 
potentially expensive choice since it requires loading every time step, but it 
would be worth having it as a simple option.

On May 14, 2009, at 8:13 AM, Renato Elias wrote:

Ok, I've just found a way using the "Temporal Statistics" filter to get the 
maximum and minimum values. After that we just need to fix the color bar range 
using these values.

Just to keep the question:

Is there any simpler or automatic way to do that?

Cheers

Renato.

On Thu, May 14, 2009 at 10:05 AM, Renato Elias 
<[email protected]<http://[email protected]> <http://[email protected]> > wrote:


Does anyone know if it's possible to rescale the color bar considering all 
values of a transient dataset? I just need to get the maximum and minimum 
values across all time steps.

thanks for any help.

 Renato.


 _______________________________________________
Powered by www.kitware.com<http://www.kitware.com> <http://www.kitware.com>  
<http://www.kitware.com>

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview




   ****      Kenneth Moreland
    ***      Sandia National Laboratories
***********
*** *** ***  email: [email protected]<http://[email protected]> 
<http://[email protected]>
**  ***  **  phone: (505) 844-8919
    ***      web:   
http://www.cs.unm.edu/~kmorel<http://www.cs.unm.edu/%7Ekmorel> 
<http://www.cs.unm.edu/%7Ekmorel>





   ****      Kenneth Moreland
    ***      Sandia National Laboratories
***********
*** *** ***  email: [email protected]<http://[email protected]>
**  ***  **  phone: (505) 844-8919
    ***      web:   
http://www.cs.unm.edu/~kmorel<http://www.cs.unm.edu/%7Ekmorel>


_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: 
http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Reply via email to