Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia announces financial results for second quarter fiscal 2024

nvidianews.nvidia.com

Post reply on HN
Print view View original

381–390 of 444 posts

Re: Nvidia announces financial results for second quarter fiscal 2024

#381

Post by FuriouslyAdrift »

PheonixPharts wrote: post #175
TheAlchemist wrote: post #124

What's also pretty interesting that they actually didn't sell more chips this quarter - they ... just pretty much doubled the prices (hence the huge margin). This is what having a monopoly looks like ! This is also why companies that manufacture their cards didn't report any uptick in profits. I'm wondering how this play out in some months ? Do they have any pricing power with respect to NVidia ? Or NVidia could just…

> This is what having a monopoly looks like ! As someone who has been in the AI/ML space for over a decade, and even had an AMD/Radeon card for more than half of that, I can't help but feel that this is partially AMD's own fault. For many, many years it seemed to me that AMD just didn't take AI/ML seriously whereas, for all it's faults, NVIDIA seemed to catch on very early that ML presented a tremendous potential mar…

AMD has an entire line specifically for AI/ML... https://www.amd.com/en/graphics/instinct-server-accelerators

They just don't have those capabilities in their consumer GPUs.

AMD is also nearly 50/50 with nVidia for supercomputers in the Top500 (and dominates at the top)

It took a few years after completeing the massive purchase of Xilinx to get going, but they are picking up speed rapidly.

Re: Nvidia announces financial results for second quarter fiscal 2024

#382

Post by girvo »

Dalewyn wrote: post #315
girvo wrote: post #299

Earlier quoted context omitted.

> could it be that a lot of companies only optimize for hardware that has the largest market share? Yes, but also Nvidia partners directly with companies too (money can/does change hands). Now the flip side is: so does AMD.

>Now the flip side is: so does AMD. And their antics shutting out Nvidia (eg: FSR only, no DLSS) aren't being received well, not the least because their offerings are objectively inferior to Nvidia's.

While Nvidia isn’t doing these silly antics today, they’ve absolutely done them in the past. None of the large consumer silicon companies have clean hands with respect to anticompetitive/anti-consumer behaviours. They’ve all got too much power frankly.

Re: Nvidia announces financial results for second quarter fiscal 2024

#383

Post by izacus »

zirgs wrote: post #344
izacus wrote: post #330

Earlier quoted context omitted.

And then there's the console market, with more units sold than PC gaming market, with 100% AMD GPUs.

The Switch is based on Nvidia hardware.

And the high budget consoles aren't.

Re: Nvidia announces financial results for second quarter fiscal 2024

#384

Post by gen220 »

saberience wrote: post #365
gen220 wrote: post #257

Earlier quoted context omitted.

The fundamental model of company valuation is that the company's equity is worth the accumulation of the company's future cash flows, typically discounted by some rate so that predicted cash flows 5 years from now are worth some % less than 100%, with the intent of pricing in uncertainty. Nvidia has had a watershed moment of revenue growth, because they're the only significant player in the space of top-of-the-line G…

So you think demand for AI driven software and tools is going to stop growing over the next few years? It’s a big call. I think it’s just the beginning personally but time will tell.

Note it isn’t the demand for AI driven software and tools that has to be sustained or increased, it’s the demand for massive GPUs that are used to train AI tools.

To make a flawed analogy, but if people want oil (AI), NVDA sells drills (means of AI production). Demand for each is obviously correlated, but the profile of demand for the two products is really different.

Re: Nvidia announces financial results for second quarter fiscal 2024

#385

Post by haldujai »

danielmarkbruce wrote: post #308
haldujai wrote: post #224

Earlier quoted context omitted.

> Customer support. I've never seen so many people get up to speed so fast on something. It's unlike anything ever. Are you saying that you've seen enterprises developing, training and running their own in-house LLMs from scratch directly on large (i.e. 100s to 1000s) GPU clusters, whether on-prem or cloud, for this to be relevant to Nvidia? Pardon my skepticism but it seems odd that a generic Fortune 500 co has the…

You seem to be moving the goalposts. First it was using in production, now it's building foundation models.. But, to answer: Not foundation models (although, Bloomberg did that for another use case). Fine tuned versions of falcon, now llama 2, open assistant and others. They are running both the fine tuning (not that it's much) and inference on A100s, inferentia and TPU. The skillset required to fine tune a model for…

> You seem to be moving the goalposts. First it was using in production, now it's building foundation models..

I'm not moving the goal posts but maybe I wasn't clear. When we talk about corporations adopting fine-tuning and inference that's not especially relevant to H100 sales which is the main cash-cow (~10B of revenue at 80-90% margins) and Nvidia's massive market cap growth. What is relevant is corporations like Inflection AI building 22k H100 clusters.

I work in academia so PyTorch is more common but are people in industry fine-tuning LLMs actually working directly with CUDA that much for this to be a big moat?

Re: Nvidia announces financial results for second quarter fiscal 2024

#386

Post by solardev »

paulmd wrote: post #318
solardev wrote: post #259

Earlier quoted context omitted.

Why do GPUs get so much more expensive compared to the rest of the system? CPUs, hard drives, prebuilt systems, laptops etc. haven't gotten ridiculously more expensive. It used to be that $200 would get you a low-range GPU, $400 a midrange, and $600 a pretty darned good one. The 1070 launched at $379 (or $480 in 2023 dollars), the 2070 at $499, the 3070 at $499, then all of a sudden the 4070 now is $599. That's a big…

It’s because CPUs tend to be fundamentally limited in the ways they can efficiently utilize transistors to scale performance, by things like cache latency, reorder depth, branch prediction, etc. While gpus have always been god’s strongest soldier for putting transistors on silicon scalably. They were the perfect machine for a world where transistors per dollar doubled every 18 months. On the other hand now that it’s…

So if they're running into hardware fab limits, how does running deep learning on that limited hardware equate to a doubling of price? I don't quite follow the logic there?

Yeah the upscaling is nice, but I dunno about $1200 nice...

It was cool when they instead focused on things like creating mobile smaller versions of the cards with laptop level power draws, or better cooling systems that weren't as noisy etc. While staying within reason price points. Just the mundane stuff, not necessarily real time raytracing (gimmicky and not very noticeable IMO, even as someone who uses 4080s on Geforce Now). Something like the Steam Deck package with its integrated console-like APU is much more consumer friendly, but also much much weaker than a 4090 I guess. Different priorities that Nvidia might've reconsidered if not for the AI supersampling stuff also bleeding into crypto and ML, justifying their bet.

Edit: Yeah their AI profits are skyrocketing, while gaming is a has-been. Good bet for the company, bad omens for gamers :/

https://www.pcgamer.com/nvidias-record-breaking-profits-are-...

Re: Nvidia announces financial results for second quarter fiscal 2024

#387

Post by redox99 »

Gareth321 wrote: post #372
redox99 wrote: post #333

Earlier quoted context omitted.

They are not similar at all. Tesla P/E is 67, Nvidia's P/E is 244.

Which is - excuse my French - fucking insane. The current PE ratio of 244 is pants on head ridiculous. That's 244 years worth of exceptionally high profit just to break even. Your great great great great great great great great grandchildren could take over your shares at the same value as today and the company would still be overpriced. That's centuries worth of technological change and high volatility baked into th…

Also Tesla a lot more room to grow in term of revenue (they're still a small % of cars solds) than nvidia (already dominates gaming and datacenter).

I admit I'm biased in favor of Tesla however.

Re: Nvidia announces financial results for second quarter fiscal 2024

#388

Post by danielmarkbruce »

haldujai wrote: post #385
danielmarkbruce wrote: post #308

Earlier quoted context omitted.

You seem to be moving the goalposts. First it was using in production, now it's building foundation models.. But, to answer: Not foundation models (although, Bloomberg did that for another use case). Fine tuned versions of falcon, now llama 2, open assistant and others. They are running both the fine tuning (not that it's much) and inference on A100s, inferentia and TPU. The skillset required to fine tune a model for…

> You seem to be moving the goalposts. First it was using in production, now it's building foundation models.. I'm not moving the goal posts but maybe I wasn't clear. When we talk about corporations adopting fine-tuning and inference that's not especially relevant to H100 sales which is the main cash-cow (~10B of revenue at 80-90% margins) and Nvidia's massive market cap growth. What is relevant is corporations like…

Companies are using both A100 and H100 for inference. The datacenter numbers include both, and I don't believe they break it down further. And no, by and large, large enterprises are not building out large clusters - but they are much of the demand for all those cloud provider build outs.

No, no one is using CUDA directly. But if you are a vendor with a new equivalent, it's no small feat integrating it with a framework like PyTorch. There is a lot of work required by various parties.

Re: Nvidia announces financial results for second quarter fiscal 2024

#389

Post by andromeduck »

Panzer04 wrote: post #310
andromeduck wrote: post #139

Earlier quoted context omitted.

The problem with Intel is: 1. They don't pay - Nvidia/Google/Apple easily pays 1.5-2x Intel before appreciattion. 2. They're cheap/beaurcratic. The office sucks, your laptop sucks. 3. They suck at software. https://pharr.org/matt/blog/2018/04/18/ispc-origins 4. They can't develop/retain talent. Half the ML-HW/FW teams at AMD/Google/Nvidia/Apple are ex-Intel.

How is that indicative of not developing talent? Sounds like they have a fairly robust pipeline if so many end up working at other companies... (Retention notwithstanding :P)

They lost most of their talent in the early/mid 2010s but you're right to an extent for non-software. For software they're not a terrible place to accumulate domain specific knowledge but don't really retain people long enough or promote quick enough to make a difference.

Re: Nvidia announces financial results for second quarter fiscal 2024

#390

Post by paulmd »

highwaylights wrote: post #326
paulmd wrote: post #312

Earlier quoted context omitted.

In reality it literally is as simple as “intel says it’s 7nm so it’s 7nm”, but density is what determines the node number, insofar as anything determines it beyond marketing. So, if intel 7 is a 10nm class node then so is tsmc N7. But the whole industry cooked their node names a decade ago and intel is allowed to do the same terminology. To do anything otherwise would be a needless marketing disadvantage. If someone’…

This, but also if we took it at face value (we shouldn't for the reasons in the reply above), then this means Intel is even further ahead when adjusting for the difference in node sizes. i.e. This would mean Intel will be getting a four-node-scale jump in performance (10nm->7nm->5nm->3nm->1.8nm) from where they are with Raptor Lake by 2025 if they can stay on track. I don't have any insight into this but I would hope…

> I don't have any insight into this but I would hope Pat Geslinger is righting the salary, perks and incentive structure as a matter of priority to stop the shedding of talent.

Absolutely not and this is one of the strongest headwinds intel faces. They just did a big round of layoffs and then everyone who didn’t get cut got a surprise 20% pay cut after, and retention bonuses for seniors was typically low-3-digits to high-2-digits. Principle engineers were solidly 3 digits though. And then they did stock buybacks that exceeded the size of the layoffs and pay cuts.

Intel is fucking broke and they’re gonna have to cut every cost they can to focus on their key markets, and do it with personnel that were already underpaid and have better opportunities elsewhere. Not that labor is the straw that’s breaking intel’s back here but the situation is really AMD-in-2012 level dire, they have an incredibly long and expensive road to return to profitability and they will have to do it with whatever they can scrape up out of a staff that was already abused even in the “good times”.

Oh and they want you to move from desirable west-coast locations (well, except Arizona) to Ohio. And they’re massively behind on their tech stack (fighting uphill is not what everyone wants) in a failing org with a ton of middle-management rot and neglect. And they’re still failing to execute a huge amount of the time, causing massive delays, etc.

It’s really really dire and just because intel has staunched the worst of the financial bleeding doesn’t mean there’s much light at the end of the tunnel. Like Boeing I don’t think they will be allowed to go under, they’re strategically too important, but they’re gonna have a bad time for a while and probably legitimately do need a bunch of subsidy and “pity contracts” for national labs HPC computers to stay on their feet - just like AMD did.

Post reply on HN