Duncan

Got my first game credit!!!

22.03.2025 22:21 — 👍 5    🔁 0    💬 0    📌 0

www.youtube.com/watch?v=SV-n...

While I certainly wouldn't recommend open source tools, this is an otherwise excellent talk on basic FPGA GPUs, and the amazing things you can do with a modest board. Really great to see other people doing talks like this, and I really hope it inspires more people

18.03.2025 04:57 — 👍 24    🔁 5    💬 1    📌 0

Taster image of trees and sky, from the Work Graphs presentation by AMD at GDC 2025 Advanced Graphics Summit.

Join us & the Microsoft DirectX12 team at #GDC2025 for exciting sessions that push the boundaries of #gamedev 🛠️

📈 2.40pm today: our in-depth look at how Work Graphs are redefining GPU programmability!

👉 Learn more about all the DirectX sessions: devblogs.microsoft.com/directx/dire...

17.03.2025 19:19 — 👍 12    🔁 4    💬 0    📌 0

I feel the need to shout out this awesome D3D12 feature table that a few people from the DirectX discord put together: d3d12infodb.boolka.dev/FeatureTable...

18.11.2024 02:23 — 👍 122    🔁 38    💬 1    📌 2

Work Graph Playground: a learning framework for GPU Work Graphs Read about our latest sample for D3D12 GPU Work Graphs. We're making Work Graphs more accessible with a tutorial framework.

Signal boosting from last week
gpuopen.com/learn/work-g...

Provides a great introduction to GPU work graphs.

18.11.2024 16:35 — 👍 3    🔁 1    💬 0    📌 0

Black cat lying down ontop of a cat tower

Black cat staring down ontop of a cat tower

Gecko pictures for your timeline.
You're welcome :)

17.11.2024 16:57 — 👍 1    🔁 0    💬 0    📌 0

Having ever delved into raytracing in any form in d3d12 before, but quite happy that inline is so much easier to perform than the standard dxr pipeline.

17.11.2024 16:55 — 👍 0    🔁 0    💬 0    📌 0

Basic compute inline raytracing implemented.

Next jumping onto compute shader execution using work graphs.

17.11.2024 16:55 — 👍 0    🔁 0    💬 1    📌 0

Quick and dirty post on implementing hot shader reload with DXC: wallisc.github.io/rendering/20...

16.11.2024 19:13 — 👍 46    🔁 9    💬 2    📌 0

Nice summary of occupancy as a way to hide mem latency, but it is not the only (and sometimes not the best) way. Helping the compiler to add more instructions between the mem read issue and the data use by for eg partially unrolling a loop might help as well (from gpu-primitives-course.github.io)

17.11.2024 12:46 — 👍 81    🔁 10    💬 2    📌 1

hello bluesky

16.11.2024 11:44 — 👍 1    🔁 0    💬 0    📌 0