[visit-users] SimplifiedDomainNesting

Siddharth Shankar funkysidd at gmail.com
Wed Apr 15 15:42:03 EDT 2009


Sending a VTK dump of data file may serve the purpose. Let me know if that
would be ok?

-Siddharth

On Wed, Apr 15, 2009 at 1:27 PM, Justin <luitjens at cs.utah.edu> wrote:

> I could send you a datafile but you would not have the appropriate reader
> as our reader is not part of visit and i'm not certain what would be the
> best way to get the plugin to you.
> The problem is we do not want to prune data out, which is what the
> simplified domain nesting appears to be doing. Look at the attached image.
>  The coarsest level (red) should have patches which overlap the finer levels
> but those patches are not being displayed because the domain nesting is
> removing those patches.  Just to clarify the red patches should be
> everywhere.
>
> Thanks,
> Justin
>
>
>
> Mark Miller wrote:
>
>> I'd have to get Hank to eye-ball this a bit but judging from the
>> pictures, I'd say you are asking for a GUI control to work around what
>> appears to be a bug in the simplified domain nesting class (perhaps an
>> off-by-one issue) as I don't see anything in your images that would
>> suggest that the missing parts is 'proper' behavior for that class.
>>
>> If I am correct, then I am very strongly against having a GUI control
>> for this. We simply need to fix the bug. Can you send (small) data file
>> example (say less than 10 megs).
>>
>> Mark
>>
>> On Wed, 2009-04-15 at 12:10 -0600, Justin wrote:
>>
>>
>>> Attached are 2 images showing the patch sets of a 3 level AMR dataset.
>>>
>>> The first image shows the artifact and the second shows the way we want
>>> to display it.  Essentially it appears that the coarser level patches are
>>> sometimes completely removed.  In this case we want to show each level
>>> entirely and have them overlap.
>>> Justin
>>>
>>> Mark Miller wrote:
>>>
>>>
>>>> If you can elaborate on the conditions under which simplified domain
>>>> nesting is yielding poor results, I'd bet we can add some cheaply
>>>> computable logic to automatically decide NOT to apply it. My guess is
>>>> that it is the case where the domains do not 'align' on their 'corners'.
>>>> Is that the case?
>>>>
>>>> I think finding some reasonable conditions under which to disable it is
>>>> probably a better route that providing a GUI control for it.
>>>>
>>>> Mark
>>>>
>>>> On Wed, 2009-04-15 at 11:40 -0600, Justin wrote:
>>>>
>>>>
>>>>> Hi,
>>>>>
>>>>> For some of our plots we do not want to have simplified domain nesting
>>>>> on as it causes odd visual artifacts.  One solution is to edit
>>>>> /src/plots/Subset/avtSubsetFilter.C and comment out
>>>>> TurnSimplifiedNestingRepresentationOn().  Is there another way to disable
>>>>> this without having to recompile VisIt?  Ideally it would be nice if there
>>>>> were an option in the gui somewhere to turn this on and off.
>>>>>
>>>>> Thanks,
>>>>> Justin
>>>>>
>>>>> --
>>>>> List subscription information: https://
>>>>> email.ornl.gov/mailman/listinfo/visit-users
>>>>> Searchable list archives: https://
>>>>> email.ornl.gov/pipermail/visit-users
>>>>> VisIt Users Wiki: http://  visitusers.org/
>>>>> Frequently Asked Questions for VisIt: http://  visit.llnl.gov/FAQ.html
>>>>>
>>>>>
>>>>>
>>>>>
>>>> --
>>> List subscription information: https://
>>> email.ornl.gov/mailman/listinfo/visit-users
>>> Searchable list archives: https:// email.ornl.gov/pipermail/visit-users
>>> VisIt Users Wiki: http:// visitusers.org/
>>> Frequently Asked Questions for VisIt: http:// visit.llnl.gov/FAQ.html
>>>
>>>
>>>
>>
>
> --
> List subscription information:
> https://email.ornl.gov/mailman/listinfo/visit-users
> Searchable list archives: https://email.ornl.gov/pipermail/visit-users
> VisIt Users Wiki: http://visitusers.org/
> Frequently Asked Questions for VisIt: http://visit.llnl.gov/FAQ.html
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://email.ornl.gov/pipermail/visit-users/attachments/20090415/e9c80e4c/attachment.html>


More information about the visit-users mailing list