Nvidia Again Crushes Earnings!

[Daniel Nenni]

Was there really any doubt?

Nvidia Crushes Earnings on Soaring AI Chip Demand. The Stock Is Still Slipping

Nvidia’s revenue was up 206% in its latest quarter, beating Wall Street’s expectations. The chip maker also provided a revenue forecast for the January quarter handily above estimates.

But the stock’s huge 2023 gains left investors wanting even more. Shares wavered in after hours trading.

The semiconductor company reported October quarter adjusted earnings per share of $4.02, compared to Wall Street’s consensus estimate of $3.37, according to FactSet. Revenue of $18.1 billion was above analyst expectations of $16.2 billion. Data center revenue surged in the October quarter to $14.5 billion, up 279% from the prior year and up 41% from the prior quarter.

The outlook was also robust. For the current quarter, Nvidia provided a revenue forecast range that, at the midpoint, was $20 billion. That’s above the consensus of $18.0 billion.

“Our strong growth reflects the broad industry platform transition from general-purpose to accelerated computing and generative AI,” Nvidia CEO Jensen Huang said in the press release. “NVIDIA GPUs, CPUs, networking, AI foundry services and NVIDIA AI Enterprise software are all growth engines in full throttle. The era of generative AI is taking off.”

Nvidia's AI Chip Demand Is Strong. But Supply Counts.

Nvidia stock has surged 245% this year. The big question is what management will say about Nvidia's progress in alleviating supply constraints.

www.barrons.com

 

[hist78]

At this speed, by the end of current quarter Nvidia's annual revenue $58.8 billion will surpass Intel's $54 billion. The semiconductor market is changing rapidly.

B=Billion

Nvidia Fiscal 2024 Revenue
2024-01-31 $20B (Guidance)
2023-10-31 $18.1B
2023-07-31 $13.507B
2023-04-30 $7.192B


Intel 2023 Revenue
2023-12-31 $15.1B ($14.6B ~ $15.6B Guidance)
2023-09-30 $14.158B
2023-06-30 $12.949B
2023-03-31 $11.715B

 

[nghanayem]

It impresses me that in spite of the consistently raising bar and how impossibly high Nvidia already seems that they are still getting big QoQ improvements on the top line. I am still getting intel in 1999 energy as it is hard to believe that this huge AI DC build out will last for all of time. I don't remember who I heard this quote from, but it goes something like this: "When your uncle is asking about investing in it, it has probably already peaked.". One thing is for sure though, when we look back in 20 years even if the training boom tempers, and domain specific chips and integrated AI engines end up being the main vehicles of running AI models I don't think NVIDIA will be the pets.com or IBM of AI.

 

[hist78]

It impresses me that in spite of the consistently raising bar and how impossibly high Nvidia already seems that they are still getting big QoQ improvements on the top line. I am still getting intel in 1999 energy as it is hard to believe that this huge AI DC build out will last for all of time. I don't remember who I heard this quote from, but it goes something like this: "When your uncle is asking about investing in it, it has probably already peaked.". One thing is for sure though, when we look back in 20 years even if the training boom tempers, and domain specific chips and integrated AI engines end up being the main vehicles of running AI models I don't think NVIDIA will be the pets.com or IBM of AI.

IBM has been working on AI and AI related chip designs for a long time. So far we haven't seen any huge AI success coming out from IBM yet. I'm talking about something like ChatGPT, AI enabled Bing or Google search, H100, or even MI300 types of rollout and market acceptance. Do you have any thoughts on it?

 

[nghanayem]

IBM has been working on AI and AI related chip designs for a long time. So far we haven't seen any huge AI success coming out from IBM yet. I'm talking about something like ChatGPT, AI enabled Bing or Google search, H100, or even MI300 types of rollout and market acceptance. Do you have any thoughts on it?

I didn’t really mean IBM’s AI stuff. More so that I don’t think we will be looking back at Nvidia as something similar to what happened to IBM with the PC and server (at least not on Jensen’s watch).

As for their AI stuff I am not really a great person to ask given I don’t do design, software, or ML stuff. As an outsider who doesn’t really pay attention to IBM’s chip designs looking in I look at it like their mainframes. Niche, well integrated, high margin, low volume. But as I said I am FAR from an authority on this subject.

 

[Jumper]

Interesting thing is Nvidia is also very well positioned in other segments. So even if the demand for AI softens it can just shift reaources to other segments.