Sam Martin 🇺🇦 retweeted
ARM is talking about their neural upscaler in Moving Mobile Graphics. Great that some HW vendors are starting to be more open about their AI upscalers.
Sam Martin 🇺🇦 retweeted
Got accepted to SIGGRAPH 2026! My talk is titled:
Reducing graphics API complexity: A clean slate API design for modern hardware
I am an ARM Ambassador nowadays. I am going to be talking in their track:
developer.arm.com/community/…
Moving Mobile Graphics has been accepted again to SIGGRAPH! More neural graphics, GPU-driven rendering, API futures and real time GI from speakers across the mobile industry. Details here: developer.arm.com/community/… cc. @SebAaltonen @l_j_oneil
During the summer, I had the pleasure of talking about our work on ML in GPUs at SIGGRAPH. “How GPUs must adapt in the Age of ML/AI”. Recording now up! youtube.com/watch?v=lshI6PbX…
Turns out I’m doing a keynote at GPC in a month, talking about what I see as the eras past, present and future in mobile graphics. Spoiler: AI is involved. Looking forward to it! graphicsprogrammingconferenc…
1
2
339
Sam Martin 🇺🇦 retweeted
If you are interested about neural temporal AA/upscaling, here's a great paper from ARM. I wish other GPU vendors would document their approaches as openly as ARM. AMD FSR to be fully open source too, but FSR4 isn't open source anymore.
huggingface.co/Arm/neural-su…
Sam Martin 🇺🇦 retweeted
GPU Work Graphs SIGGRAPH 2025 course notes: coburggraphicslab.github.io/…
I swear I must own ~30 UK-north America adapters now. Can I find even one of them? No chance. Where do they go?!
2
279
Going to SIGGRAPH and interested in ML? I’m going to be talking about what we’ve been working on for the last 5 years. If you are interested in or working on real time neural graphics, seriously please say hi! s2025.conference-schedule.or…
1
5
427
Sam Martin 🇺🇦 retweeted
Are you coming to #Siggraph? Here! You're welcome. Please share!
docs.google.com/document/d/1…
Sam Martin 🇺🇦 retweeted
We're delighted to welcome Professor Hannah Fry to Cambridge!
Mathematician and award-winning science presenter @fryrsquared will join us as Professor of the Public Understanding of Mathematics @FacultyMaths, following an announcement @NewtonInstitute 👉 cam.ac.uk/research/news/awar…
Sam Martin 🇺🇦 retweeted
Woooooooo, our tiny game about real-time GI & procgen and shipping Rust & Vulkan is finally out! 🎉🖖🦀
Tiny Glade is out NOW on Steam (Win & Linux) 💗🎉
Thank you so much for following the development and being so supportive of our little diorama builder.
We hope you have fun, and I can’t wait to see what you make 🤩 If you post your glades, add #TinyGlade so I can find 'em :D
Sam Martin 🇺🇦 retweeted
Once @Microsoft 's Shader Model 7 is released, #DirectX 12 will accept shaders compiled to SPIR-V. Microsoft is working with the Khronos #SPIR and #Vulkan Working Groups to ensure that this transition benefits the whole development ecosystem.
devblogs.microsoft.com/direc…
Microsoft
Stockholm friends! We are in you! Done Djurgården, Gamla Stan, Ice Bar. But what next? Any recommendations?
2
2
532
Some slides for Moving Mobile Graphics course now up! Including neural super res from @l_j_oneil, Tencent's SmartGI from Shun Cao and Supercell's rendering from @heinapurola #SIGGRAPH2024 community.arm.com/arm-commun…
Some slides for Moving Mobile Graphics course now up! Including neural super res from @l_j_oneil, Tencent's SmartGI from Shun Cao and Supercell's rendering from @heinapurola #SIGGRAPH2024 community.arm.com/arm-commun…
Sam Martin 🇺🇦 retweeted
ARMs neural upscaler presentation was awesome. Significantly improved quality over FSR2. They said 1W / 10 TOPS neural unit should be able to run it. They also said that low latency GPU-NPU interop (and resource compatibility) is super important for graphics. I agree 100%.
Sam Martin 🇺🇦 retweeted
We finally get to share our neural materials research.
We created custom highly-realistic reference materials, we invented new neural models, and we built new graphics systems.
End result is these neural materials run in a real-time path tracer.
research.nvidia.com/labs/rtr…












