Discussion exciting new features, research & advancements in gaming (graphics & adjacent software)

[marees]

thread to discuss all new features to gaming GPU architectures & graphics APIs

idea behind this thread is to discuss "Real time graphics, GPGPU, ML and adjacent (physics, audio etc) software research"

(suggestion: if you are posting videos then post text links also)

separate thread below for hardware features

Discussion - General GPU µArch Research Thread

Hi all, thought I'd start some new threads just for research on µArchs, with this one specific to GPUs alone.

forums.anandtech.com

reference 1 — GPU blogs

1. nvidia GeForce — https://blogs.nvidia.com/blog/category/gaming/
2. amd GPUOpen — https://gpuopen.com/
3. intel Arc — https://game.intel.com/in/stories/?category=intel-arc

reference 2 — Gaming APIs

• OpenGL — Cross-platform GAPI. Use if uncertain.
• Vulkan — OpenGL but low-level. High-performance, cross-platform.
• MoltenVK — Enables Vulkan on macOS/iOS.
• DirectX 11 — Microsoft's GAPI for Windows. Use if Windows-only.
• DirectX 12 — Improved, lower-level for Windows. Use for more features.
• WebGL — FLASH. Web standard for 3D graphics. Use for web rendering.
• WebGPU — Successor to WebGL. Better, broader support. Use for modern web graphics (many functions not supported on newer webb updates).
• Metal — Apple's low-level GAPI. Use for macOS/iOS development.

reference 3 — history of GPUs

https://www.techspot.com/article/659-history-of-the-gpu-part-4/

https://www.techspot.com/article/2702-gpu-compute-ai/

 

[marees]

"Work graphs completely changes how a GPU is programmed and what a GPU even is. Think of it as another GPGPU or Programmable shader moment."

AMD blog on this — for those who want text version

GDC 2024 Work graphs and draw calls – a match made in heaven! - AMD GPUOpen

Introducing "mesh nodes", which make draw calls an integral part of the work graph, providing a higher perf alternative to ExecuteIndirect dispatches.

gpuopen.com

& follow up demo paper/pdf here

https://twitter.com/x/status/1813318845769097417

extracts from blog & paper (as screenshots)
View attachment 136538

View attachment 136541


View attachment 136540

That's the boring paper for mesh nodes launch.

At HPG 2025 they went full on crackpot with GPU tree gen: https://gpuopen.com/download/Real-Time_GPU_Tree_Generation.pdf

AMD's solution to the VRAM problem & memory price issues ??

https://twitter.com/x/status/1937338942157898113

 

[marees]

removed speculative info on patents

 

[marees]

removed post on CCU as it is a hardware feature

 

[MrMPFR]

The most impressive Work Graphs paper to date from HPG 2025:

GPUOpen download: https://gpuopen.com/download/Real-Time_GPU_Tree_Generation.pdf

HPG talk:

Bastian Kuth's YT demonstration:

Features:

• 100% GPU driven SOTA procedural tree generation using GPU Work Graphs API and Mesh Nodes extension
• Per frame changes based on 150 parameters: "Our model supports procedural displacement, seasonal changes, complex pruning, animation, culling, continuous LOD, and intuitive artistic control with real-time edits."
• Based on 55kB of generation code vs 34.8GB for pregenerated assets (static)
• Allocates a 1.5GB scratchpad buffer for work graph vs Execute Indirect's combinatorial explosion that makes this impossible on consumer grade HW
• Rendered in real-time with 3.13ms total overhead (G-buffer + PCG) on RX 7900 XTX
• Auto LOD geometry overhead to ht certain FPS or ms target.
• Nanite like continuous LOD
• Responds to wind in very convincing way (rebuild for every single frame)

 

[marees]

this is very old (from 6 years ago)

but I am curious. whatever happened to CPU based ray tracing for directX11 games ?

World of Tanks Implements Ray Traced Shadows for ANY DX11 GPU​

Meant to post this yesterday, but I forgot. The video link below goes through how the implemented it, and its an interesting watch. For anybody wondering why its only shadows, its because each tank has in excess of 50k polygons, and you can have up to 50 tanks on screen at any given time. Currently no hardware exists that can handle that kind of load with RT reflections/GI. Wargaming has always tried to make their engine run on even low end hardware, and not require special hardware.

However, with how they implemented the RT Shadows, with help from Intel, it can run on any GPU. I play the game a lot, so I downloaded their test client, with the game settings at Ultra, and RT Shadows at HIGH, my system with a 4.5GHz 4690K and an RX480 4GB ran it really quite well at 1080P. It did drop below 60fps a few times. As a note, this does not use any specialized RT hardware, even if the GPU has it. This uses a combination of CPU and GPU to handle the calculations, and they use only 1 ray per pixel, with a custom dithering filter that they came up with which is shown in the video.

Details: https://worldoftanks.com/en/news/general-news/ray-tracing/

Developer Video: