# Slow paint speed with large (10GB) volume

**URL:** <https://discourse.slicer.org/t/slow-paint-speed-with-large-10gb-volume/18195>\
**Category:** Support\
**Tags:** segmentation, performance\
**Created:** [June 17, 2021, 9:41pm UTC](https://discourse.slicer.org/t/slow-paint-speed-with-large-10gb-volume/18195 "2021-06-17T21:41:25Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![hherhold](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/hherhold/32/12199_2.png) [@hherhold](https://discourse.slicer.org/u/hherhold)\
**Post date:** [June 17, 2021, 9:41pm UTC](https://discourse.slicer.org/t/slow-paint-speed-with-large-10gb-volume/18195/1 "2021-06-17T21:41:25Z")

</div>

I have a colleague with a fairly large micro-CT scan (10GB), and he is experiencing sluggish response times when painting with a sphere brush in the 2D slices (Show 3D not enabled). His machine is windows with 128GB RAM. Not sure on specs for the CPU.

Seems to me this is a pretty large volume, and it’s just a lot of data to go through for painting with intensity range turned on? Any suggestions? I’m thinking the best way to proceed is probably to chop it into a few pieces and work on it in sections.

Any ideas would be most appreciated - thanks!!

-Hollister

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [June 18, 2021, 8:00pm UTC](https://discourse.slicer.org/t/slow-paint-speed-with-large-10gb-volume/18195/2 "2021-06-18T20:00:03Z")

</div>

This volume is about 30-50x larger than average volumes, which exposes performance bottlenecks that we normally don’t perceive at all. A 0.5s operation may turn into a 20s operation.

Potential solutions:

- crop and/or downsample: as you suggested, to work with smaller pieces at a time
- use larger hardware: if it’s just a one-off case then you can try to rent a very powerful computer from a cloud provider and see if it makes the workflow fast enough
- optimize performance: build Slicer in “release with debug info” mode and do the segmentation with a profiler attached; if you identify the performance bottlenecks then we can give advice about how to improve the update speed