# PVD with files with different meshes
**URL:** https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946
**Category:** ParaView Support
**Created:** [July 24, 2020, 7:30am UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946 "2020-07-24T07:30:52Z")
**Posts on this page:** 9
**Page:** 1
### Author:  [@mrklein](https://discourse.paraview.org/u/mrklein)
#### Post date: [July 24, 2020, 7:30am UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/1 "2020-07-24T07:30:52Z")
Dear community,
I have a time series of node-centered data on rectangular grid with **changing geometry**. I create PVD file as follows:
```
```
Where x can be r, s, or u, since, I can save the data as RectilinearGrid, StructuredGrid, or UnstructuredGrid.
The problem is: when I step over data in ParaView, it does not update displayed geometry for cases of RectilinearGrid and StructuredGrid. Yet, ParaView re-reads files, since Information tab displays correct number of points/cells. In case of UnstructuredGrid geometry is updated. If I load individual VT(R|S|U) files, geometry is displayed correctly. The behaviour is the same for ParaView versions 5.6.2, 5.7.0, and 5.8.0.
I the behaviour a bug or a feature? I would prefer to save data as RectilinearGrid or as StructuredGrid, since they have implicit connectivity.
You can find example files in attached archive.
[vtk-example.tar.gz](https://discourse.paraview.org/uploads/short-url/piCv5mJ1wkWodY7Ch6s8misE0iX.gz) (1.1 KB)
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 1, 2020, 1:05pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/2 "2020-09-01T13:05:24Z")
The WholeExtent cannot grow over multiple timesteps. This is definitely a bug or at least a limitation of the implementation. It has already been reported sometime ago :
[https://gitlab.kitware.com/paraview/paraview/-/issues/12391](https://gitlab.kitware.com/paraview/paraview/-/issues/12391)
The simplest workaround is to add a “dummy” timestep at the begining that is the biggest possible in terms of extent.
---
### Author:  [@mrklein](https://discourse.paraview.org/u/mrklein)
#### Post date: [September 3, 2020, 8:08pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/3 "2020-09-03T20:08:23Z")
Dear Mathieu,
Thank you for the reply.
The solution with “dummy” time step is interesting, yet I was not able to find in PVD documentation ([https://www.paraview.org/Wiki/ParaView/Data\_formats](https://www.paraview.org/Wiki/ParaView/Data_formats)) a way to hide this time step (if it is possible at all).
Going from VTR/VTS to VTU solves the problem, increases file size, complicates code logic, but is acceptable solution.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 4, 2020, 7:35am UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/4 "2020-09-04T07:35:19Z")
> yet I was not able to find in PVD documentation a way to hide this time step
Not possible afaik.
---
### Author:  [@Nic](https://discourse.paraview.org/u/Nic)
#### Post date: [June 30, 2023, 1:55pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/5 "2023-06-30T13:55:24Z")
Sorry to bump this old thread, is there any update on that?
Fixing this bug would be **extremely useful**.
Thank you.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [June 30, 2023, 2:02pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/6 "2023-06-30T14:02:39Z")
> [@Nic](#):
>
> Fixing this bug would be **extremely useful**.
Which bug ? This one: [https://gitlab.kitware.com/paraview/paraview/-/issues/12391](https://gitlab.kitware.com/paraview/paraview/-/issues/12391)
---
### Author:  [@Nic](https://discourse.paraview.org/u/Nic)
#### Post date: [June 30, 2023, 2:21pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/7 "2023-06-30T14:21:08Z")
Hi Mathieu, yes this one.
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [July 3, 2023, 5:02am UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/8 "2023-07-03T05:02:17Z")
I’m afraid not yet.
Do not hesitate to reach out though!
> [@How are features and bug fixes incorporated into ParaView?](https://discourse.paraview.org/t/how-are-features-and-bug-fixes-incorporated-into-paraview/4735):
>
> As with all open source projects, adding features and fixing bugs takes developer effort. That effort may originate from volunteer community contributors, or through paid professional services offered by companies like [Kitware](https://www.kitware.com). Features and bug fixes are prioritized by the needs of contributing community members and by professional services customers. Note that contributions are vetted by a rigorous community review and testing process; therefore it is recommended that significant changes are d…
---
### Author:  [@Nic](https://discourse.paraview.org/u/Nic)
#### Post date: [July 4, 2023, 2:43pm UTC](https://discourse.paraview.org/t/pvd-with-files-with-different-meshes/4946/9 "2023-07-04T14:43:37Z")
Thank you @mwestphal,
I’m not sure I’m qualified to do that.
Nevertheless, people in my group have found a workaround by accident (trying every filter):
Applying a **RedistributeDataSet** on the series fixes the issue. 🤷♂️
Hope it’s helpful to someone.