Monday, June 23rd 2025
Researchers Unveils Real-Time GPU-Only Pipeline for Fully Procedural Trees
A research team from Coburg University of Applied Sciences and Arts in Germany, alongside AMD Germany, introduced a game-changing approach to procedural tree creation that runs entirely on the GPU, delivering both speed and flexibility, unlike anything we've seen before. Showcased at High-Performance Graphics 2025 in Copenhagen, the new pipeline utilizes DirectX 12 work graphs and mesh nodes to construct detailed tree models on the fly, without any CPU muscle. Artists and developers can tweak more than 150 parameters, everything from seasonal leaf color shifts and branch pruning styles to complex animations and automatic level-of-detail adjustments, all in real-time. When tested on an AMD Radeon RX 7900 XTX, the system generated and pushed unique tree geometries into the geometry buffer in just over three milliseconds. It then automatically tunes detail levels to maintain a target frame rate, effortlessly demonstrating stable 120 FPS under heavy workloads.
Wind effects and environmental interactions update seamlessly, and the CPU's only job is to fill a small set of constants (camera matrices, timestamps, and so on) before dispatching a single work graph. There's no need for continuous host-device chatter or asset streaming, which simplifies integration into existing engines. Perhaps the most eye-opening result is how little memory the transient data consumes. A traditional buffer-heavy approach might need tens of GB, but researcher's demo holds onto just 51 KB of persistent state per frame—a mind-boggling 99.9999% reduction compared to conventional methods. A scratch buffer of up to 1.5 GB is allocated for work-graph execution, though actual usage varies by GPU driver and can be released or reused afterward. Static assets, such as meshes and textures, remain unaffected, leaving future opportunities for neural compression or procedural texturing to further enhance memory savings.
The key to this achievement is work graphs, which can orchestrate millions of tasks without exploding dispatch counts. Traditional ExecuteIndirect calls would struggle with trees that can have up to 128^4 leaves (around 268 million), but work graphs handle it with ease. Widespread adoption will take time since current support is limited to AMD's RDNA 3+ and NVIDIA's 30-series and newer GPUs. Full game-engine integration and console support are still on the horizon. Looking forward, the researchers are exploring how to extend this flexible, GPU-driven pipeline into ray tracing, possibly by building on-GPU bounding volume hierarchies with the same work-graph framework.
Source:
HPG Paper
Wind effects and environmental interactions update seamlessly, and the CPU's only job is to fill a small set of constants (camera matrices, timestamps, and so on) before dispatching a single work graph. There's no need for continuous host-device chatter or asset streaming, which simplifies integration into existing engines. Perhaps the most eye-opening result is how little memory the transient data consumes. A traditional buffer-heavy approach might need tens of GB, but researcher's demo holds onto just 51 KB of persistent state per frame—a mind-boggling 99.9999% reduction compared to conventional methods. A scratch buffer of up to 1.5 GB is allocated for work-graph execution, though actual usage varies by GPU driver and can be released or reused afterward. Static assets, such as meshes and textures, remain unaffected, leaving future opportunities for neural compression or procedural texturing to further enhance memory savings.
The key to this achievement is work graphs, which can orchestrate millions of tasks without exploding dispatch counts. Traditional ExecuteIndirect calls would struggle with trees that can have up to 128^4 leaves (around 268 million), but work graphs handle it with ease. Widespread adoption will take time since current support is limited to AMD's RDNA 3+ and NVIDIA's 30-series and newer GPUs. Full game-engine integration and console support are still on the horizon. Looking forward, the researchers are exploring how to extend this flexible, GPU-driven pipeline into ray tracing, possibly by building on-GPU bounding volume hierarchies with the same work-graph framework.
34 Comments on Researchers Unveils Real-Time GPU-Only Pipeline for Fully Procedural Trees
By the way, I still don’t get this, well school taught me this so nothing else is possible attitude. The point of basic 101 anything is to dumb things down to get a wide audience interested in something. Not define the length and breath of what’s possible.
That is why you need to go back to "how code works 101" and learn the most basic "dumbed down" version of the subject so you gain the slightest idea of what you are talking about. Hence the 101 reference.
There is a means of having generic performant hardware... they are called FPGAs, and it requires insane coding expertise to make them efficient and they still aren't a best choice for many applications.
If you want to know about the direction of compute, read up on silicon photonics, and all the competing fabrics that will eventually use it, nvlink, UAlink, Ultra Ethernet.
The reason the industry moved to dedicated silicon chunks for specialized acceleration is efficiency... you can have specific tasks handled and powered down when not needed.
There are certainly things that can be combined, FP32 and Int32 in gpus have been separate and combined back and forth from generation to generation, and you get... a tradeoff, it can do more FP32 all at once, but worse if both are needed at the same time.
You struck the 3 of us with profound ignorance and want to wave inapplicable credentials while claiming to want to learn while not listening.