# Numerics of "Gradient Of Unstructured DataSet" Filter

**URL:** https://discourse.paraview.org/t/numerics-of-gradient-of-unstructured-dataset-filter/638
**Category:** ParaView Support
**Created:** [September 26, 2018, 12:54pm UTC](https://discourse.paraview.org/t/numerics-of-gradient-of-unstructured-dataset-filter/638 "2018-09-26T12:54:58Z")
**Posts on this page:** 1
**Showing post:** 16

<div class="post-metadata">

### Author: ![chiarapalu](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/c/e47774/32.png) [@chiarapalu](https://discourse.paraview.org/u/chiarapalu)
#### Post date: [December 18, 2019, 8:50am UTC](https://discourse.paraview.org/t/numerics-of-gradient-of-unstructured-dataset-filter/638/16 "2019-12-18T08:50:58Z")

</div>

I’ve checked the scales of both velocity and size of the grid and they are both in meters (in paraview). Do you think that remeshing the velocity field in paraview could change something? My idea so far is that the computation of the derivatives in paraview is done by finite difference while fluent could work on finite volumes so probably is more accurate. Do you think is more correct computing the gradient on cell data or point data? Because also changing from point to cell the results are completely different…  
Thanks a lot!

---

_[View the full topic](https://discourse.paraview.org/t/numerics-of-gradient-of-unstructured-dataset-filter/638)._