# Code review: Mandelbrot set with CUDA.jl

**URL:** <https://discourse.julialang.org/t/code-review-mandelbrot-set-with-cuda-jl/68673>\
**Category:** GPU\
**Tags:** gpu, parallel\
**Created:** [September 24, 2021, 5:21am UTC](https://discourse.julialang.org/t/code-review-mandelbrot-set-with-cuda-jl/68673 "2021-09-24T05:21:26Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![vini-fda](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/vini-fda/32/29448_2.png) [@vini-fda](https://discourse.julialang.org/u/vini-fda)\
**Post date:** [September 24, 2021, 5:52am UTC](https://discourse.julialang.org/t/code-review-mandelbrot-set-with-cuda-jl/68673/3 "2021-09-24T05:52:02Z")

</div>

Oh it’s because I forgot the threads argument, correct? I guess that’s a silly mistake, haha. I’ve just now run with threads=1024 and it ran in about 13 seconds. Thank you.

Another question: I’ve seen the [post on @cuda threads](https://discourse.julialang.org/t/cuda-threads-and-blocks-confusion/54816/4) and the new occupancy API, but I’m not sure I understood how to use it. Could you expand on that?

---

_[View the full topic](https://discourse.julialang.org/t/code-review-mandelbrot-set-with-cuda-jl/68673)._
