A WebGPU implementation of Augmented Vertex Block Descent • Hacker News Board

Re: A WebGPU implementation of Augmented Vertex Block Descent

#21

Post by jackling » Fri, Apr 10, 2026, 2:23 PM UTC

fc417fc802 wrote: post #20

10000truths wrote: post #17

Earlier quoted context omitted.

One or more of: * Low gravity * Slow timestep * Unrealistic mass ratios of colliding bodies * Incorrect, or lack of, drag

The timestep effect is remarkable. I was quite surprised with a basic PBD simulation when I lowered the timestep into the nanosecond (IIRC, anyway it was really small and no longer ran in real time) range just to see what would happen and got lots of high frequency shivering effects that looked exactly like what happens IRL when metal objects are fed through a shredder.

Never tried with that low of a timestep, wonder if that could start causing floating point issues which will lead to more instability.

Re: A WebGPU implementation of Augmented Vertex Block Descent

#22

Post by jackling » Fri, Apr 10, 2026, 2:25 PM UTC

Stevvo wrote: post #9

mjmdavis wrote: post #6

This is really awesome. Great work. I aspire to build cool stuff like this in WebGPU. Very excited for the future of the web.

It can definitely be achieved with Claude. Even with no experience in graphics progammming, I've been able to replicate results of several papers in related to fluid simulation.

Curious, when you do this, do you understand the math/reasoning of the paper and just have claude as do the coding? Not saying that matters if you just care about the end result, but I'm curious how much using an agent affects your understanding of what the papers are proving.

Re: A WebGPU implementation of Augmented Vertex Block Descent

#23

Post by fc417fc802 » Fri, Apr 10, 2026, 2:32 PM UTC

jackling wrote: post #21

fc417fc802 wrote: post #20

Earlier quoted context omitted.

The timestep effect is remarkable. I was quite surprised with a basic PBD simulation when I lowered the timestep into the nanosecond (IIRC, anyway it was really small and no longer ran in real time) range just to see what would happen and got lots of high frequency shivering effects that looked exactly like what happens IRL when metal objects are fed through a shredder.

Never tried with that low of a timestep, wonder if that could start causing floating point issues which will lead to more instability.

Yeah I did have to refactor slightly to minimize precision related impacts.

Re: A WebGPU implementation of Augmented Vertex Block Descent

#24

Post by Stevvo » Fri, Apr 10, 2026, 9:55 PM UTC

jackling wrote: post #22

Stevvo wrote: post #9

Earlier quoted context omitted.

It can definitely be achieved with Claude. Even with no experience in graphics progammming, I've been able to replicate results of several papers in related to fluid simulation.

Curious, when you do this, do you understand the math/reasoning of the paper and just have claude as do the coding? Not saying that matters if you just care about the end result, but I'm curious how much using an agent affects your understanding of what the papers are proving.

I went in with limited understanding, gained some more as the agent worked through it. What can appear complex in the paper often turns out to be far more simple and elegant once you see the code written.

Re: A WebGPU implementation of Augmented Vertex Block Descent

#25

Post by madbo1 » Tue, Apr 14, 2026, 6:53 AM UTC

The direction seems very interesting indeed. Using optimization algorithms such as Augmented Vertex Block Descent in WebGPU marks a step towards doing proper numerical computations right in the browser and not just visualizations and/or ML inferences.

Perhaps, more importantly, this development makes optimization approaches more accessible by eliminating the need for special environments (CUDA, native compilation, and alike).

I am very interested in comparing the performance of these optimizations in comparison with native GPU implementations, especially in terms of memory throughput and precision limits.