Lightnews — Scholar-powered news

Kostas Anagnostou

@kostasanagnostou.bsky.social

1.9K followers 150 following 200 posts

Lead Rendering Engineer at Playground Games working on Fable. Always open for graphics questions or mentoring people who want to get in the industry. I tweet about graphics mostly. Views my own. Blog: https://interplayoflight.wordpress.com/

interplayoflight.wordpress.com

Posts Media Videos Starter Packs

Pinned

Kostas Anagnostou @kostasanagnostou.bsky.social · Oct 23

I realised that I jumped straight into posting without an intro first: Hi all, I'm a rendering engineer of 20+ years, currently leading the awesome Fable rendering team. I like my pixels both pretty and fast. I enjoy tinkering with graphics techniques in my spare time and sometimes I blog about it.

3 1 49

Kostas Anagnostou @kostasanagnostou.bsky.social · 8h

Quick experiment in Compiler Explorer to observe how VGPR allocation differs between wave32 and wave64 on RDNA: with wave32 it appears to be in batches of 8 while with wave64 it is in batches of 4 VGPRs. godbolt.org/z/W6ee8MhMx

Compiler Explorer - HLSL (RGA 2.6.2 (DXC trunk))

// The entry point and target profile are needed to compile this example: // -T ps_6_6 -E PSMain RWBuffer<float> output; cbuffer SomeData { float dataArray[10]; int index; } //Undefin...

godbolt.org

Kostas Anagnostou @kostasanagnostou.bsky.social · 11h

Really good, maths-free, introduction to the Fourier transform from this year's Siggraph, recommended watch: dl.acm.org/doi/10.1145/...

Introduction to the Fourier Transform | Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Courses

dl.acm.org

2 7 29

Kostas Anagnostou @kostasanagnostou.bsky.social · 2d

Quick tip, add the "-fspv-target-env=vulkan1.1" command line argument to Compiler Explorer's RGA to get it to compile HLSL shaders with wave intrinsics: godbolt.org/z/PKoh5d51K

Compiler Explorer - HLSL (RGA 2.6.2 (DXC trunk))

// The entry point and target profile are needed to compile this example: // -T ps_6_6 -E PSMain struct PSInput { float4 position : SV_Position; float4 color : COLOR0; }; float4 PSMain(PS...

godbolt.org

2 11

Kostas Anagnostou @kostasanagnostou.bsky.social · 7d

"Inside NVIDIA GPUs: Anatomy of high performance matmul kernels", includes a great intro to GPU architecture and PTX/SASS: www.aleksagordic.com/blog/matmul

Inside NVIDIA GPUs: Anatomy of high performance matmul kernels - Aleksa Gordić

From GPU architecture and PTX/SASS to warp-tiling and deep asynchronous tensor core pipelines.

www.aleksagordic.com

12 55

Reposted by Kostas Anagnostou

Arseny Kapoulkine @zeux.io · 8d

New blog post! In "Billions of triangles in minutes" we'll walk through hierarchical cluster level of detail generation of, well, billions of triangles in minutes. Reposts welcome!

zeux.io/2025/09/30/b...

1 53 160

Reposted by Kostas Anagnostou

Martin Fuller @martinfuller.bsky.social · 8d

Jobs available with Microsoft graphics: devblogs.microsoft.com/directx/sigm...

The Silicon, Graphics, Media and AI (SiGMA) team is hiring! - DirectX Developer Blog

Do you want to be a part of the Silicon, Graphics, Media and AI (SiGMA) platform team? We are hiring for multiple positions – if you’re interested in knowing more about opportunities in SiGMA, please ...

devblogs.microsoft.com

1 7 7

Reposted by Kostas Anagnostou

ceruleite.bsky.social @ceruleite.bsky.social · 9d

#SIGGRAPH2025 Advances in Real-Time Rendering in Games course talks slides posted: "Adaptive Voxel-Based Order-Independent Transparency" shipping in the upcoming Call of Duty: Black Ops 7 by Michal Drobot from Activision Central Tech is now online: advances.realtimerendering.com/s2025/index....

21 81

Reposted by Kostas Anagnostou

Martin Fuller @martinfuller.bsky.social · 12d

A little something I've been meaning to write for a while martinfullerblog.wordpress.com/2025/09/26/a...

Thank you to the awesome people at Machine Games and id Software!

Anisotropic Scaling in Indiana Jones and the Great Circle and DOOM: The Dark Ages

While working on Indiana Jones and the Great Circle and later DOOM: The Dark Ages, it became apparent that both titles are particularly sensitive to anisotropic sampling cost on the Xbox Series con…

martinfullerblog.wordpress.com

19 63

Kostas Anagnostou @kostasanagnostou.bsky.social · 15d

I expect it does, but this shouldn't affect potential upstream bottlenecks (ISBE/PE/TRAM).

Kostas Anagnostou @kostasanagnostou.bsky.social · 16d

Nsight Graphics' GPU Trace/Trace Analysis often provides more low level hardware information, in the form of tooltips and performance advice, than the documentation available online does, worth exploring a few captures to understand the GPU architecture better.

Kostas Anagnostou @kostasanagnostou.bsky.social · 16d

Thanks, it doesn't seem like it has.

2 2

Kostas Anagnostou @kostasanagnostou.bsky.social · 17d

Continuing on the topic of GPU utilisation and performance, as a practical example, I looked a bit deeper into the impact of vertex shader exports on the cost of a drawcall and wrote another blog post with some observations interplayoflight.wordpress.com/2025/09/21/t...

The performance impact of vertex shader exports

Following up on the previous post on GPU utilization and performance, and to provide a practical example, I expanded a bit on a topic discussed in brief: vertex shader exports and their impact on p…

interplayoflight.wordpress.com

2 13 50

Reposted by Kostas Anagnostou

mjp @mjp123.bsky.social · Sep 8

I've got a new blog post for all of you fine folks! It runs through the additions to D3D12 since it was released, and finishes up with some of the things that have changed for me personally in my code.

(And yes it's really been 10 years 👴).

therealmjp.github.io/posts/ten-ye...

Ten Years of D3D12

For those of us that have been using it from the start, it can be hard to believe that Direct3D 12 has been around for nearly ten years now. Windows 10 was released on July 29th 2015, and D3D12 has be...

therealmjp.github.io

7 52 150

Kostas Anagnostou @kostasanagnostou.bsky.social · Sep 3

Brief introduction to a number of OIT techniques: "Advances in Order Independent Transparency for Real-Time & Virtual Production Workflows" www.youtube.com/watch?v=wXSJ...

Advances in Order Independent Transparency for Real-Time & Virtual Production Workflows - P. Kakkar

YouTube video by Academy Software Foundation

www.youtube.com

11 47

Kostas Anagnostou @kostasanagnostou.bsky.social · Aug 30

I'm glad you found it useful. Async compute was mentioned in passing in that article, I have looked into it much deeper here: interplayoflight.wordpress.com/2025/05/27/a...

Async compute all the things

GPUs make work parallelism very easy by design: each drawcall/dispatch shader instruction operates on batches of vertices, pixels, threads in general at the same time automatically. On the other ha…

interplayoflight.wordpress.com

Kostas Anagnostou @kostasanagnostou.bsky.social · Aug 29

New blog post discussing a few approaches to bottleneck reduction and GPU utilisation and performance increase interplayoflight.wordpress.com/2025/08/29/g...

GPU utilisation and performance improvements

Drill deep into a GPU’s architecture and at its heart you will find a large number of SIMD units whose purpose is to read data, perform some vector or scalar ALU (VALU or SALU) operation on i…

interplayoflight.wordpress.com

1 24 100

Kostas Anagnostou @kostasanagnostou.bsky.social · Aug 27

It is interesting that both RTGI presentations at Advances in R-T Rendering this year (both great reads!) introduce it as a solution to scalability and baking size/time issues, due to the size/dynamic nature of the world, more than a visual improvement advances.realtimerendering.com/s2025/index....

Advances in Real-Time Rendering in Games, SIGGRAPH 2025 - Celebrating 20 years!

advances.realtimerendering.com

2 18

Kostas Anagnostou @kostasanagnostou.bsky.social · Aug 27

Detecting reads from uninitialised heap memory in C++ programs at runtime www.forwardscattering.org/post/71

Forward Scattering - The Weblog of Nicholas Chapman

First off, most malloc implementations in debug mode will fill the memory with a special byte pattern like 0xCDCDCDCD. Unfortunately this doesn't directly help you detect reads of such patterns.

www.forwardscattering.org

4 22

Kostas Anagnostou @kostasanagnostou.bsky.social · Aug 22

Worth re-sharing this oldish but still great presentation as an example of how perf should be viewed holistically, maximising all GPU units' utilisation even if it means making a particular drawcall's execution slower to achieve this. s3.amazonaws.com/nd.images/re... www.youtube.com/watch?v=CvS6...

s3.amazonaws.com