# Limit of the data value's exponent

**URL:** https://discourse.paraview.org/t/limit-of-the-data-values-exponent/16919
**Category:** ParaView Support
**Tags:** vtk
**Created:** [July 9, 2025, 11:50am UTC](https://discourse.paraview.org/t/limit-of-the-data-values-exponent/16919 "2025-07-09T11:50:28Z")
**Posts on this page:** 2
**Page:** 1

<div class="post-metadata">

### Author: ![mdev](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/m/45deac/32.png) [@mdev](https://discourse.paraview.org/u/mdev)
#### Post date: [July 9, 2025, 11:50am UTC](https://discourse.paraview.org/t/limit-of-the-data-values-exponent/16919/1 "2025-07-09T11:50:28Z")

</div>

Hello,

I have a question about any limit of the exponent value in .vtk data. I have tried to open the attached .vtu files, but the first one ‘test.vtu’ triggers an error “cannot read point data array ~. The data array in the element may be too short.” Then, after I modified the exponent of a very small data value from -46 to -44, it is working properly. I am wondering if there’s any limit value for the exponent.

Thank you.

[test.vtu](https://discourse.paraview.org/uploads/short-url/3WjZh1PL2vzk62TEOMD2kl4WLAG.vtu) (4.0 KB)  
[test\_rev.vtu](https://discourse.paraview.org/uploads/short-url/iLu1wP1jGa7awzrhKIQqAyAgzNO.vtu) (4.0 KB)

 ![Bild1](https://discourse.paraview.org/uploads/default/original/3X/5/e/5e4521531c77fb301d30e44436457ce9c68d5907.png)

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [July 9, 2025, 11:59am UTC](https://discourse.paraview.org/t/limit-of-the-data-values-exponent/16919/2 "2025-07-09T11:59:47Z")

</div>

> I am wondering if there’s any limit value for the exponent.

Yes, read this: [https://stackoverflow.com/questions/7644699/how-are-floating-point-numbers-stored-in-memory](https://stackoverflow.com/questions/7644699/how-are-floating-point-numbers-stored-in-memory)

The solution is simple. you are using `float` (aka float32) data representation, use `double` (float64) instead.
