# Clip closed surface filter over multiple open structures
**URL:** https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969
**Category:** ParaView Support
**Created:** [October 2, 2023, 2:39am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969 "2023-10-02T02:39:08Z")
**Posts on this page:** 7
**Page:** 1
### Author:  [@fashing](https://discourse.paraview.org/u/fashing)
#### Post date: [October 2, 2023, 2:39am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/1 "2023-10-02T02:39:08Z")
Hi Paraview,
I tested `clip closed filter` on a `superquadric` (donut) and I found something weird.
With default setting (Origin: 0, 0, 0; Normal: 1, 0, 0), one of the holes is not completely closed.
However, as I modify the parameters, sometimes it works and sometimes it doesn’t.
 
I can open a report if someone confirms this issue. Thank you!
[Update]
There’s another related issue, the quality difference between `clip filter` (left) and `clip closed filter` (right) is huge. Is this normal or can it be avoided with some additional setting?
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [October 2, 2023, 6:02am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/2 "2023-10-02T06:02:05Z")
> I can open a report if someone confirms this issue. Thank you!
Please do, looks like a bug
> the quality difference
Missing normals, can be recomputed with GenerateSurfaceNormals, but this should not be needed, no reason to remove the normals.
---
### Author:  [@fashing](https://discourse.paraview.org/u/fashing)
#### Post date: [October 2, 2023, 6:35am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/3 "2023-10-02T06:35:34Z")
The issue has been reported.
[Error in clip closed surface filter on multiple open structures (#22327) · Issues · ParaView / ParaView · GitLab (kitware.com)](https://gitlab.kitware.com/paraview/paraview/-/issues/22327)
> no reason to remove the normals.
I didn’t remove the normals, maybe this is another issue of the filter.
This quality drop doesn’t happen on my other testing though (only on `superquadric`).
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [October 2, 2023, 7:14am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/4 "2023-10-02T07:14:17Z")
> [@fashing](#):
>
> I didn’t remove the normals, maybe this is another issue of the filter.
Not reason for \*\*the filter \*\* to remove the normals, looks like a bug.
> This quality drop doesn’t happen on my other testing though
In which situtation does it not appear ?
---
### Author:  [@fashing](https://discourse.paraview.org/u/fashing)
#### Post date: [October 2, 2023, 9:02am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/5 "2023-10-02T09:02:13Z")
In my previous post, the quality did not drop.
[Clip closed surface filter over contour filter - ParaView Support - ParaView](https://discourse.paraview.org/t/clip-closed-surface-filter-over-contour-filter/12901)
I think this information suggests `clip close surface` filter use `clip` filter for contour output. 🤔
Should I open another issue about the normal removal, or append this in the previous issue?
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [October 3, 2023, 6:28am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/6 "2023-10-03T06:28:59Z")
Normals do not matter for a box.
Clip closed surface actually remove all arrays as it would need to compute them for new cells. That’s possible but more of a missing feature than a bug.
---
### Author:  [@fashing](https://discourse.paraview.org/u/fashing)
#### Post date: [October 3, 2023, 6:45am UTC](https://discourse.paraview.org/t/clip-closed-surface-filter-over-multiple-open-structures/12969/7 "2023-10-03T06:45:29Z")
Actually, it’s not a regular cuboid, but a round box from a contour filter.
And I was wrong, the image quality also drops in this case.
(vdb → contour filter)
(vdb → contour filter → clip closed surface)
