Paraview does not correctly render AMReX plotfiles with large physical geometry #2637
Replies: 3 comments 18 replies
-
|
Could you try reading them with amrvis, yt or visit to make sure the
plotfiles themselves aren’t somehow corrupted? We use paraview regularly
with no problem.
…Sent from my iPhone
On Feb 27, 2022, at 4:49 PM, Ben Wibking ***@***.***> wrote:
Hi,
Opening a set of plotfiles produced with my code, I only get a black
surface rendering for any of the fields, e.g.:
[image: Screen Shot 2022-02-28 at 11 47 01 AM]
<https://user-images.githubusercontent.com/1781919/155907666-10e62df6-6ba2-4cef-80f1-0a92e4233fb0.png>
Is this a known issue, or should I open a bug with Paraview for this?
—
Reply to this email directly, view it on GitHub
<#2637>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACRE6YWJNH6TT53FLQLAMBLU5LA77ANCNFSM5PPWFYLA>
.
Triage notifications on the go with GitHub Mobile for iOS
<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675>
or Android
<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
I can read them in with Visit with no problem. It seems to only happen when the grid sizes exceed 512^3 cells per rank. I've uploaded such a plotfile that exhibits this problem here: https://cloudstor.aarnet.edu.au/plus/s/pM333ijq9pr1QMQ |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
|
I can confirm I was getting strange output for both Paraview 5.9.1 and 5.10. Also values in the spreadsheet view seemed off. This is the output I got for Amrvis, does it look correct to you? |
Beta Was this translation helpful? Give feedback.
-
|
@etpalmer63 @asalmgren The issue is that Paraview can't handle large physical box geometry. If I rescale the geometry so that the RealBox coordinates go from [0, 1] in each dimension, it works fine. I've opened a bug on the Paraview bugtracker for this: https://gitlab.kitware.com/paraview/paraview/-/issues/21371 |
Beta Was this translation helpful? Give feedback.
-
|
Hi @BenWibking, great info! I'm planning to look into the Paraview code myself and see if I can submit a few fixes. In the meantime, thanks for all your detailed reports on the issues! |
Beta Was this translation helpful? Give feedback.
-
|
It turns out this is also an issue in Visit's volume rendering (though not slice or surface rendering): visit-dav/visit#17602 |
Beta Was this translation helpful? Give feedback.
-
|
Hey @BenWibking, since the issues appear to be similar I am wondering whether this comes from the VTK backend. I know ParaView has this, and it does look like VisIt has some of its code. In either case, I went ahead and let both ParaView and VisIt know that the issue occurs in both places since their engineers probably know better what aspects these codes share in common and maybe that will help them troubleshoot. Erik |
Beta Was this translation helpful? Give feedback.
-
|
That makes sense, thanks. |
Beta Was this translation helpful? Give feedback.
-
Hi,
Opening a set of plotfiles produced with my code, I only get a black surface rendering for any of the fields, e.g.:
Is this a known issue, or should I open a bug with Paraview for this?
Beta Was this translation helpful? Give feedback.
All reactions