MantisBT - ParaView
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0004371ParaView(No Category)public2007-01-25 19:492009-12-09 14:50
ReporterAlan Scott 
Assigned ToAmy Squillacote 
PriorityhighSeveritymajorReproducibilityalways
StatusclosedResolutionno change required 
PlatformOSOS Version
Product Version 
Target VersionFixed in Version 
Project
Topic Name
Type
Summary0004371: SNL: GLYPH filter can't set scale factors correctly - 2.6
Description* The GLYPH filter has lost the ability to set the scale factors correctly.
  Here are details on how to replicate this bug.
  * Windows XP machine.
  * Debug build (probably not important)
  * Run ParaView 2.6, from this Monday or Tuesday (probably not important).
  * Open can.ex2 (We have numerous versions of can running around here. We believe that they are exactly the same. The extension has been changed to make sure ParaView uses the correct reader for multiple extensions.)
  * Node variables - All On.
  * Accept.
  * Play the animation.
  * Go to the start of the animation. (left button on the animation panel).
  * Glyph filter.
  * Change Orient/Scale, Vectors to ACCL.
  * Read Scale Factor - BAD.

  Here are details to hid the bug.
  * Run ParaView 2.6, from this Monday or Tuesday (probably not important).
  * Open can.ex2 (We have numerous versions of can running around here. We believe that they are exactly the same. The extension has been changed to make sure ParaView uses the correct reader.)
  * Node variables - All On.
  * Accept.
  * Play the animation.
  (* Go to the start of the animation. (left button on the animation panel).) - DON'T do this.
  * Glyph filter.
  * Change Orient/Scale, Vectors to ACCL.
  * Read Scale Factor - GOOD.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2009-12-09 14:50Berk GeveciProject@3@ => ParaView
2011-06-16 13:10Zack GalbreathCategory => (No Category)

Notes
(0006272)
Amy Squillacote   
2007-01-26 10:37   
Hi Alan,

Utkarsh and I looked into this issue more closely this morning, and it turns out that the behavior in 2.6 is actually correct. The glyph filter uses the timestep selected on the Parameters page for the reader, and uses only the vector magnitude computed from the array for that timestep in computing the scale factor.

In 2.4, for some reason ParaView was using the *last* timestep, even when you reset the animation to the first timestep. This is incorrect behavior. It may have *appeared* that ParaView was somehow using the range of the array over all the timesteps in the dataset, but this is not the case. It just happens that the last timestep gives you reasonbly-sized vectors for playing through the whole animation.

The logic to actually use the vector range over all the timesteps would be pretty complicated; according to Berk, such functionality is not even planned for ParaView 3.
(0006273)
Alan Scott   
2007-01-26 11:50   
I really don't like this behavour, but consider it acceptable.

Thanks for looking at this one.

Alan