Comment by seanc
1 day ago
If you're NVIDIA then open-weight models are a classic example of commoditizing your complements; cheaper models mean more people buying GPU's to run them. [1]
Your guess is as good as mine for China though.
[1] https://www.joelonsoftware.com/2002/06/12/strategy-letter-v/
Nvidia selling more GPU's at the cost of it's datacenter business is pretty close to Kodak selling digital cameras at the cost of film.
The data center side is so bloated anything that eats into it is a huge negative. Their data center business brings in 20x the gpu market. Local open weight models will be what pops the bubble and China will do anything in it's power to enable that pop.
I wonder what the thinking inside NVIDIA is at the moment. They have countless examples to learn from here, about the danger of not being willing to cannibalize your high end products. But, of course, there’s a reason that there are lots of examples of this sort of failure.
There’s plenty of competition that would be happy to attack them from below, though…
I also wonder what this moment will mean for chip design going forward. The fact that we are so constrained on the GPU side along with the success of unified memory with apple silicon does make me wonder what chips in 3-5 years will look like.
2 replies →
Make money while there's money to be made then pivot to the next scheme?
Gaming > Bitcoin > LLMs > Robotics
Jensen's job is to just be one step ahead of the market dynamics to keep the investor dollars flowing.
3 replies →
> Nvidia selling more GPU's at the cost of it's datacenter business is pretty close to Kodak selling digital cameras at the cost of film.
This assumes a) AI is a zero-sum game, and b) we're actually talking about on-prem AI will replace cloud-based AI. I think neither statements are true.
AI is like compute: we'll need all sorts of it, in various sizes, everywhere. I'm sure Nvidia whats to own all the workloads.
On the other hand, I do think open weight, like open source, will win in general.
It will take about 35 years for digital camera sales to reach Kodaks profit margins. The market growing and companies raising the economy are not zero-sum. Nvidia/Kodak killing the golden goose is zero-sum. When the difference is your valuation crashes that's zero sum for the company, just not for everyone else.
My only concern is if we can limit the economic impact from lowering investments and causing a 40% market collapse circa 2008/9.
But they won’t with the Chinese around long-term and other tech companies that have capabilities. This is a short term bottleneck.
> Nvidia selling more GPU's at the cost of it's datacenter business is pretty close to Kodak selling digital cameras at the cost of film.
You mean they resisted the idea trying to protect their legacy business, and it ended up all but killing them?
They resisted the idea because it did kill their legacy business. Kodaks peak was 16 billion in revenue in 1996. The digital camera market will be approaching that level of value inflation adjusted around 2030.
35 years to get back to status quo. Putting a bullet in a golden goose is often considered a bad idea. Everyone is happy they are dead, but if you can't understand why they might try to keep the corpse alive you have never looked at the numbers.
7 replies →
Kodak, Xerox, IBM, Motorola of Schaumburg, Illinois are perfect examples of companies that did not want to upset the apple cart, another example outside tech of tech that did not want to go into the future was US Steel, and while I’m at it to a lesser extent Ford and GM are teetering since 1973…
Hmm this sounds like an incumbent missing a paradigm shift because they didn't want it to disrupt their core (usually enterprise) business, although riding the shift would have ultimately delivered an order magnitude larger business.
Classical example is Microsoft actively undermining mobile because it threatened selling Windows or enterprise licenses.
Or Yahoo fighting Google's model because the latter model's didn't depend on taking enterprise deals to rank results.
Thanks generic LLM model response. Please upgrade to a higher tier model if you are going to waste our time.
1 reply →
Could you explain what you mean by this? I thought all of their insane profitability and returns are from crazy margins on their GPUs. I know they started/partnered/invested in some data center businesses, but I thought they were fledgling
Their data center business is selling gpus. They wrap them in a complete platform, but that's what they are selling.
They do report it separately from consumer and business sales of gpus used in PCs.
The data center business is selling Tensor core GPU's which are used for Large language models nearly exclusively. I don't think of them as GPU's anymore considering a GPU is a graphics processing unit and this is not the primary function of those cards anymore. A GPU will always be a card you put in your computer to play a game in my mind.
I know a tomato is a fruit but will still be annoyed when someone is pedantic about it because that's dumb.
When the mobile phone is eventually integrated into the human body or some other silly application in the future that will annoy me as well. Congrats on being pedantic.
3 replies →
Could you put that in units of HP Printers and Toner please?
You need 64 H200 super-node for inference for kimi k3. You will not do inference locally. What might pop the western hardware bubble is Chinese GPU, memory, networking companies. But even in China, these AI centric hardware is not cheap.
> But even in China, these AI centric hardware is not cheap
Definitely not cheap for individuals, but well within SME territory. There are countless small-town, family-owned businesses that had higher startup costs than a hypothetical Kimi-R-Us, Inc.
There’s export controls on nvidia. According to Jensen, they expect that Chinese models will start being optimized for Huawei
https://www.dwarkesh.com/p/jensen-huang
They already are.
No, Nvidia is worried about Chinese hardware stack. China is under sanction, what are they training and inferencing these models on? Even if this model might still be Nvidia chips, what about the next model. Everyone knows model and hardware companies are working together. There are a number of very competitive companies in China in this space. After they got this area sorted out, China will do training, inference and tokens entirely on their stack and export their entire stack.