Introducing Gouda: Write CUDA Run Anywhere How it works: Not a compatibility layer Conway's Game of Life Life in C Porting to Gouda Getting started with Gouda Welcome back to our series on GPU and parallel programming. In our last two posts, we compared the code required to run a simple CUDA parallel program on an NVIDIA GPU, vs an OpenCL program on any other GPU or CPU. After writing a few CUDA…
OpenCL 101 - Write an OpenCL program from scratch In this article we'll write parallel code that can run on a CPU or GPU, for both AMD and NVidia devices. This is the OpenCL version of our CUDA 101 article. Even if you plan on writing code for CPU or an AMD GPU, it you might find it helpful to first skip the CUDA article above. The OpenCL interface is unfortunately much more verbose, so it can be…
CUDA 101 - Write a simple program from scratch Splitting a workload : grid and blocks and threads Starting a task What is a CUDA kernel? A Simple CUDA Kernel Host vs Device Memory Putting it all together: Output: Experiment ideas Next up: Basic GPU Architecture In this article we'll look at writing a CUDA Program from scratch. As we learned in the previous article, CUDA allows us to write programs…
GPU Lab Launched We've just launched our GPU lab, which allows you to practice GPU programming like CUDA in your browser. GPU Programming requires both an expensive GPU, and one compatible with your preferred language. If you've been wanting to experiment with CUDA, but don't have an NVidia GPU, you can now do so with the GPU Lab. The first version allows you to edit and run your programs. We'll…
Intro to GPUs The GPU A Problem for GPUs: Matrix Multiplication Matrix Multiplication Refresher Implementation note: packed row-major matrices. Python Matrix Multiplication: Javascript Matrix Multiplication: C Matrix Multiplication CUDA Matrix Multiplication Performance Comparison Up next: Basic GPU Programming with CUDA and OpenCL In this article we'll introduce the GPU, and compare it with…
Try CUDA Online in your Browser Want to learn CUDA and parallel programming, but lack a compatible GPU? Or perhaps you're looking for an easier environment for practice, without worrying about setup. We have just launched out GPU Lab, which allows you to write and edit parallel programs in your browser, then run against our cloud GPUs. GPUs can be quite expensive, and this approach allows us to…
PicoNES Plus for Android Launches! We have just launched PicoNES Plus, an update to our earlier NES emulator for Android. This new app brings improved accuracy and performance. If you're looking for an easy to use emulator that just works, with no confusing setup process, PicoNES Plus is a great solution. Updating from PicoNES Classic PicoNES Plus is incompatible with saves from PicoNES classic.…
New Newsletter! We've recently migrated to a new newsletter provider. If you're getting this in your inbox, this is the new format. A few people had complained about the format / ads injected by the old newsletter provider, and our new one has none of those issues. While you're here, if you could reply to this email, it would help ensure that you're able to see new updates in the future.
PicoNES for Web Updated! We've just updated our web version of PicoNES, our NES emulator. We finally have sound working for this classic system in the browser, and we've also improve accuracy for a number of games. Try it today! PicoNES Web - NES in your browser