Every year just sounds like “Nvidia’s new consumer GPUs are adding new features, breaking previous performance ceilings, running games at huge resolutions and framerates. Their datacenter cards are completely sold out because they can spin straw into gold, and Nvidia continues to develop new AI and graphics techniques built on their proprietary CUDA framework (that no one else can implement). Meanwhile AMD has finally sorted out raytracing, and their consumer GPUs are… well not as good as Nvidia’s but they’re a better value if you’re looking for a competitor to one of Nvidia’s 60 or 70 line GPUs!”
I'm unsure why you're criticizing the Efficient Markets Hypothesis or even using it here, but you need to also analyze this with some time horizon because the market and marketplaces are not static.
https://www.investopedia.com/terms/e/efficientmarkethypothes...
> The efficient market hypothesis (EMH), alternatively known as the efficient market theory, is a hypothesis that states that share prices reflect all information and consistent alpha generation is impossible.
Best I've got is "central planning". One firm being able to handily out perform others despite them also being both motivated and well capitalised lends itself pretty heavily towards markets being good, but I hardly think they were referring to "central planning" when they wrote "efficient market hypothesis".
If it's so obvious to you that you're dropping ellipsis, care to clue us in?
If there was any, the prices would go down as we have seen a billion times.
This is a textbook situation that would be perfect for a competitor to come in and undercut. However not only is that not happening, nobody is even trying.
Making the “theory” pretty worthless if it’s not even applicable in cases that would naturally produce this market entrant.
The reality is that private equity does not actually want to compete with large global brands.
Is it? A competitor can enter the market and undercut by producing a cheaper and otherwise undifferentiated commodity-type product. Nvidia's focus is adding moats that prevent competing on pure specs such as CUDA, design, and so on.
Irrespectively though I don't see how a competitor can replicate the context and tacit knowledge associated building something like CUDA for close to 20 years without putting in a similar amount of time.
Nvidia also made sure that CUDA runs on gaming GPUs and supports Windows. This is why its tools are so good. You don't need to buy a datacenter GPU, no need to mess with Linux. Just buy any gaming GPU, install CUDA SDK and you're good to go.
AMD wasn't like that. Their alternative - ROCM didn't even work on gaming GPUs. Their datacenter GPUs didn't even support Windows. Basically the opposite approach of NVIDIA. Now AMD is rushing to add Windows and consumer GPU support to ROCM, but it's a bit too little too late.
They already are undercutting.
Some business that are really into hyper-scaling are already pouring man-hours into making those "undercut" alternative products work. Specifically because sometimes you can't affort Nvidia at that scale. Or they have a large enough scale that they can make it work with suboptimal tech.
It's just that for most business the man-hours required to make "undercut" products work still aren't cheap enough to win in cost-benefit.
A lot of companies are trying. It's just that nvidia is really good and significantly ahead.
Do you have any idea how much R&D it takes to make a GPU, let alone something that could possibly compete with NVIDIA?
This is not something a dozen people in a large garage are going to do.
> However not only is that not happening, nobody is even trying.
AMD has been trying to compete for years. If you're a gamer that doesn't care about getting the top performance and just wants to optimize price/performance, their offerings aren't bad.
> The reality is that private equity does not actually want to compete with large global brands.
Because the barrier of entry into the GPU market is probably at least a couple billion dollars.
There are $BN Private Equity deals done everyday, there's plenty of money to fund a competitor.
No, I think people are just scared and don't care about the existence of monopolies.
You mean a theory that no economic school actually believes? Not even Austrians would sign that.
I think the principle is called perfect market or perfect competition, but it's only a theoretical concept. However, it's certainly possible that the market is less perfect than it could be due to interventions and regulations.
People talk about AMD being competition - but from most stats I've seen, they're ~10% of dGPU sales, with Nvidia being the other 90% (with new Intel offerings being pretty much noise now). That means that if they invest the same proportion into development, NVidia nearly have 10x the resources.
It may be that tech companies like this would "naturally" form a monopoly without outside (IE government) interference, as the only reason that multiplier of development resources doesn't completely crush new entrants is rather extreme mismanagement, or a new segment is created where the design resource don't really cross over that much.
I don't see anything like that happening in the short term, if anything there seems to be more opportunity for cross pollination of development within these corporations, as there's a fair bit of design similarities between various silicon (GPUs, CPUs, accelerators for the current ML techniques etc.) that may encourage more consolidation in the whole semi market to take advantage of that, not less. But again the only thing stepping in the way of that seems to be governments trying to keep national interests, like the blocking of NVidia buying ARM to pull in one of the big CPU players. Plus all their other IP that they may benefit from, like low power GPU designs or other accelerators ARM have designed.
This means that nvidia capital is spend on testing/development infrastructure and creative labor.
So, you don’t need monopoly breaking to handle nvidia if it keeps growing, but instead rethink IP laws.
Testing infrastructure is less capital investment than production manufacture. (See ASML and TSMC beeing booked), and humans can be persuaded to work elsewhere.
This means Nvidia cannot fall a sleep, even if they keep snowballing. In a few years a rival can always arrive, or current ones snag key people or have dev/test infra breakthrough, or IP law could change as its critics rise every year.
Sure, if Nvidia keeps its good game it will keep on growing and get even bigger share, but if it doesn’t, it will happen what happened with intel, intel got greedy on its position, and the company got fat.
As long as nvidia keeps its game, it’ll be good for customers even if they swallow more market share, as prices are always limited by the value business customers get from AI, and the investment needed for a competitor is not even close to infrastructure or resource extraction stuff like high end chip fabs, oil extraction, energy grids, telecom grids.
Again, Nvidia does not produce any physical goods, only designs.
With Nvidia and other GPU competitors being IP-focused, effectively outsourcing all this "manufacturing stuff" (to the same 3rd party much of the time), that's one less thing for them to keep up with, and one less think that'll hurt if they do start "falling asleep". I can't see this happening to Nvidia in quite the same way right now. My point was that not having manufacturing makes advantages of consolidation larger, not smaller.
I wonder what would have happened if Intel realized it's manufacturing wasn't hitting targets and "quickly" added TSMC as an option, would AMD even have had a chance with ryzen? There was clearly a time when AMD had superior manufacturing processes through them, if Intel's designs of the time were on the same process would they have managed to grab the headlines?
And no, I Strongly disagree that NVidia running unchecked over the entire market being "Good for consumers", and not sure if the capital expenditure of getting over this moat is really much smaller than things like resource acquisition or infrastructure, they have $billions in current software ecosystems and hardware designs. Those $billions probably could buy you a fair bit of infrastructure investment on the scale you mentioned. Look how much Intel is burning right now just to get a toe into the market and not laughed out the door - and they're still clearly behind their competitors right now. Their chips aren't anywhere near competitive from a performance-per-area point of view, and their software is rather poor for the vast majority of use cases.
AMD compute is nowhere compared to NVIDIA. NVIDIA wanted to buy ARM, has got its finger in RISC-V, but apart from that, they don't really care. To be fair AMD has done decent with GPUs, but never enough to dethrone NVIDIA, whose playbook for the past few gens is "just make everything bigger than last gen and increase the frequency." Surely AMD could have chosen the same lazy approach to surpass the 4090 only just, but instead they didn't, so it's still NV undefeated in its space because AMD forgot to squeeze the last 1% out of their card.
The market is powerless if the competitors aren't really competing. Intel is the only chance, unless they manage to get their own Taiwanese CEO somehow related to Huang and Su.
There are in fact startups, also doing what they can (and probably not trying to go head on against the most productive competitor they can find.) And it has been reported countless times that some of the biggest customers of Nvidia are actually trying to design their own.
If you want to point out a market with broken competition, this isn't it.
The situation is created by artificial restrictions on free market (namely state enforced monopolies on "IPR", or as some call it, imaginary property).
Yes, there has been repeated efforts to chip at Nvidia's market share, but there's also a graveyard full of AI accelerator companies that fail to find product market fit due to lack of software toolchain support - and that applies even for older Nvidia GPUs and their compatible toolchains, let alone other players like AMD. This isn't a hit on Nvidia, I'm just saying things move so quickly in the space that even the only-game-in-town is trying to catch up.
Nvidia is also leading by being one or two hardware cycles ahead of their competition. I'm pretty confident AI workloads in enterprise is their next major focus [1]. I think this more than anything else will accelerate AI adoption in enterprise if well executed.
To your point, I think the industry needs to focus more on the toolchains that sit right between the deep learning frameworks (PyTorch, Tensorflow etc.) and hardware vendors (Nvidia, AMD, Intel, ARM, Google TPU etc.) Deep learning compilers will dictate if we allow all AI workloads run on just Nvidia or several other chips.
[1] - https://www.nvidia.com/en-us/data-center/solutions/confident...
"The Render Network® Provides Near Unlimited Decentralized GPU Computing Power For Next Generation 3D Content Creation."
"Render Network's system can be broken down into 2 main roles: Creators and Node Operators. Here's a handy guide to figure out where you might fit in on the Render Network:
Maybe you're a hardware enthusiast with GPUs to spare, or maybe you're a cryptocurrency guru with a passing interest in VFX. If you've got GPUs that are sitting idle at any time, you're a potential Node Operator who can use that GPU downtime to earn RNDR."
EDIT: It was just named Salad. https://salad.com/ https://salad.com/download
Or their dominance leads to competition throwing in the towel and investing resources in a market with less stiff competition.
I wouldn't be surprised to see AMD start to pair back ivnestment on high-end GPUs if things continue down this path. I would say Intel likely keeps pushing, but I'm less convinced they can actually make much headway in the near future.
It seems that Intel is making great headway on their fabs and may somehow pull off 5 nodes in 4 years. Intel 3 is entering high volume production soon and according to Gelsinger 20A is 6 months ahead of schedule and planned for H2 2024.
If they do pull this off and regain leadership that would change outlook.
https://aws.amazon.com/blogs/aws/new-amazon-ec2-p5-instances...
3.2 terabits.
The main reason why you need massive ammounts of fast VRAM in the first place is that the main limitation of AI is memory bandwidth. Can't simply distribute an algorithm that is already throughput limited by memory bandwidth and distribute it with awful latency and bandwidth and hope for any improvement.
Good luck with that.
In the current state of things, Nvidia is like a car manufacturer that exclusively owns the concept of tires.