Viewing profile — msond

We'd love to know which libraries your production code falls over on! Feel free to ping us on discord ( https://discord.com/invite/KNpgGbTc38 ) or you can find my e-mail in my prof…

We haven’t yet released support for tcgen05, but we’ll deal with it the same way we deal with other inline PTX: parsing it and converting it to target-appropriate instructions toge…

We actually support NVIDIA hardware, too. In some benchmarks, SCALE beats nvcc, and we have compiler optimizations in the pipeline that will improve those numbers over time. > If a…

We have a comparison page: https://docs.scale-lang.com/stable/manual/comparison/#zluda

A guess would be some time next year — since our public launch our focus has generally been on API coverage and increasingly recently, on performance. While performance improvement…

Actually we launched in 2024 and the last message in our discord is definitely not that: https://discord.gg/KNpgGbTc38

We're actually targeting all of it, and not just CUDA C++.

Hey there! Would love to connect and hear more about what you like about SCALE and where you'd like it to go. AMD is a part of our strategy, but it's not the end-game - we envision…

With the latest release of SCALE, wave64 AMD Instinct GPUs are supported

Very interesting, thank you for sharing! We do believe in open source software and we do want to move the GPGPU market away from fully closed languages. The future is open for disc…

We're still thinking about our approach but this is a nice suggestion, thank you. I'm curious, for what reasons are you interested in the source code yourself?

SCALE is not a "translation layer", it's a full source-to-target compiler from CUDA-like C++ code to AMD GPUs. See this part of the documentation for more details regarding warp si…

We have not reverse engineered any compiled code in the process of developing SCALE. It was clean-room implemented purely from the API surface and by trial-and-error with open CUDA…

We're going to be publishing more details on later blog posts and documentation about how this works and how we've built it. Yes, we're not open source, however our license is very…

Actually, we use mkdocs and the excellent material for mkdocs theme: https://squidfunk.github.io/mkdocs-material/

We're putting together benchmarks to publish at a later time, and we've asked some independent third parties to work on their own additionally.

You can learn more about us on https://spectralcompute.co.uk