Can I crunch numbers on my GPU from .NET? - Yes you can, and it's easy! - Tor Kristen Haugen

Nvidia's CUDA platform is the most popular framework for GPU-accelerated general computing (bar cloud solutions). But for a regular .NET developer, CUDA's C++-based libraries, PTX and nvcc compiler can be a somewhat daunting prospect to take on.
That's where ILGPU comes in. This all-.NET library, neatly nuget-packaged, does all the heavy lifting for you, letting you write not just the CUDA interface code, but even the kernel code, all from the comfort of your favorite .NET language.
The speaker will implement the iconic Mandelbrot render algorithm using ILGPU to generate those familiar, stunning images in an astonishingly few lines of code.
Check out our new channel:
NDC Clips:
‪@ndcclips‬
Check out more of our featured speakers and talks at
ndcconferences.com/
ndcoslo.com/

[@quachhengtony7651]

very interesting topic, thanks for sharing

[@Kefir0]

Awesome short talk! Can we get the demo code somewhere?

[@UllrichPraetz]

Great talk - nice Demo!

[@YM-kn1qz]

Thank you so much

[@dennistindall5487]

Great talk would really like to have the code. Thanks

[@MrArkien]

I've used it for rather compute intensive image processing for a hobby project, it rocks ! Not only all my code in C#, it runs concurrently on my GPUs and gives me results in less than 10 secs. Just what I needed for fast iterating on the code for dev & debug... And of course to use it !

[@juandaviddelgadohernandez4489]

Any tutorial to begin with?