# "Temporal Particles To Path-lines" and "Temporal Interpolation" filters...do they require particles to be defined for every time step?
**URL:** https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322
**Category:** ParaView Support
**Created:** [May 30, 2021, 12:47am UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322 "2021-05-30T00:47:13Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@datathing](https://discourse.paraview.org/u/datathing)
#### Post date: [May 30, 2021, 12:47am UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322/1 "2021-05-30T00:47:13Z")
The output from my modelling software doesn’t define every particle at every time step and it seems to be causing havoc with two filters:
- **Temporal Particles to Path-lines filter** : No path-lines drawn for particles if they aren’t defined for consecutive time steps, even if the particles new position is defined in a subsequent time step.
- **Temporal Interpolator** : No interpolation, for the same cases as above.
e.g. for a very simple case with three particle and five time steps where particle 1 and 3 are defined at all time steps, but particle 2 is only defined at the first and last time step.
What I get for timestep 1, 2 and 3:
Particles 1 and 3 interpolate and have have path-lines, but particle 2 does not.
Is the simple answer to this that those filter really require every particle to be defined at every time step? Or is there a way to make those filters work under these circumstances?
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [May 31, 2021, 7:48am UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322/2 "2021-05-31T07:48:17Z")
I’m afraid this is the expected behavior of both these filters. That being said, if you have a way to identify which particle is which, it should be trivial to implement the filter you are looking for in a python programmable filter.
---
### Author:  [@datathing](https://discourse.paraview.org/u/datathing)
#### Post date: [May 31, 2021, 8:52pm UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322/3 "2021-05-31T20:52:48Z")
Thanks! I can code a script to populate interpolated particle locations for all the missing time steps, so that should do the job. That said, the python programmable filter sounds like it might be a better option? What would the general goal/output of the filter be such that it overcomes this problem? I currently only have in mind a bit of a brute force pre-processing method (i.e. generating interpolated particle locations for all time steps).
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [May 31, 2021, 9:24pm UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322/4 "2021-05-31T21:24:35Z")
Both solution are ok.
The only complexity is that you need to keep a cache of previous timesteps in order to identify missing particles and compute the interpolation.
---
### Author:  [@datathing](https://discourse.paraview.org/u/datathing)
#### Post date: [June 1, 2021, 2:00am UTC](https://discourse.paraview.org/t/temporal-particles-to-path-lines-and-temporal-interpolation-filters-do-they-require-particles-to-be-defined-for-every-time-step/7322/5 "2021-06-01T02:00:19Z")
That makes sense. I will attempt to pre-process my data for now, but will look at the programmable filters as soon as I can. Thanks!