Hacker News new | ask | show | jobs
by janim007 4 days ago
That’s probably because only a handful of companies manufacture GPUs, and they’re still expensive. I think that will change over time as competition increases. LLMs are also still in a relatively early stage. We’re already seeing models become both smaller and more capable—for example, GPT-4 compared to Qwen3-30B, which can outperform GPT-4 in many tasks while using significantly less compute. So if this trend continues, they will be making good profits on your $100
2 comments
LaurensBER 4 days ago
Not to mention the rise of Chinese chips.

GPUs that can run everything from Crysis to CUDA are a harder engineering problem to solve than creating a chip that's optimized for inference. Not to mention that inference is an excellent first step towards a full, competitive GPU as well.

link
throwaway63467 4 days ago
Which chips are these? It seems the main challenge is data bus speed and memory capacity now and it seems no one can really compete with NVIDIA now? And i doubt NVIDIA is still optimizing for anything expect LLMs/AI, their last keynote had less than two minutes of gaming related content and they even canceled their new generation of gaming cards for now.

And it seems all of these advanced chips rely on the most advanced lithography which is tightly guarded and supply locked by a few companies.

link
LaurensBER 4 days ago
I'm not nearly as knowledgeable about chips as I would like to be but I'm seeing lots of hype around the new Huawei Ascend 910C-Chips. These are in no way competitive with Nvidia for training but they're cheap for inference and seems to be winning market share inside China.

China doesn't access to any of the latest chips technology but Huawei seems to have a roadmap to work around this by focussing on "3D chips" (vertical stacked). It's unclear if they can pull this off but if they can it might be a huge boost and allow them to further drive down inference prices.

link
throwaway63467 3 days ago
Interesting, always found it would be cool to have this stuff integrated e.g. into server hardware for specialized AI services
link
tim-tday 4 days ago
There are no Chinese gpus. The one thing they announced is equivalent to middle of the range of consumer cards four generations old and it’s not even available yet.
link
spiderxxxx 4 days ago
That may change over time, but it's not going to change anytime soon. All the fabs are sold out and companies like Nvidia and AMD are buying up all their capacity until at least 2028, and new fabs aren't going to be coming online in a short time. So, no, there won't be a lot more competition until at earliest 2032.
link