NVIDIA discussion [2024]

  • Thread starter Deleted member 2197
  • Start date
NVIDIA today reported revenue for the third quarter ended October 27, 2024, of $35.1 billion, up 17% from the previous quarter and up 94% from a year ago.

Data Center:
Third-quarter revenue was a record $30.8 billion, up 17% from the previous quarter and up 112% from a year ago.

Gaming:
Third-quarter Gaming revenue was $3.3 billion (the highest ever?), up 14% from the previous quarter and up 15% from a year ago.

Professional Visualization:
Third-quarter revenue was $486 million, up 7% from the previous quarter and up 17% from a year ago.

Automotive and Robotics:
Third-quarter Automotive revenue was $449 million, up 30% from the previous quarter and up 72% from a year ago.

Outlook:
NVIDIA’s outlook for the fourth quarter of fiscal 2025 is as follows: Revenue is expected to be $37.5 billion, plus or minus 2%.


19 billions of profit this is crazy
 
They beat all the estimates, and the market still wasn't happy as the stock was down a bit! I think more to do with QoQ growth slowing but that was bound to happen at some point.
 
They beat all the estimates, and the market still wasn't happy as the stock was down a bit! I think more to do with QoQ growth slowing but that was bound to happen at some point.

Yes very much expected since all of that was already priced in. They would have to blow away estimates for any upward movement. It’ll dip a little then continue the steady climb just as it did for the past few earnings.
 
They beat all the estimates, and the market still wasn't happy as the stock was down a bit! I think more to do with QoQ growth slowing but that was bound to happen at some point.

Meh. Buy on the rumour, sell on the news.

Honestly the analysts who come out with the estimates and predictions are not dissimilar to pundits who predict the outcome of horse races. What they do isn't that hard, and if they really knew what they were doing they'd be day traders.
 
John Carmack:

Reading The Nvidia Way book brings back tons of memories — the misguided NV1, the fast-but-flawed RIVA128, the fantastic TNT, and the path to dominance.One underrated crucial thing Nvidia did was take over driver writing from the board vendors. In The Bad Old Days, each board vendor tried to tweak the drivers to get an edge, and almost none of them had the skill to do it responsibly. Quake 3 was written so defensively because there were so many crap drivers to work around. Nvidia centralizing all the work was a huge boon.Similar situation to Apple and Google taking mobile phone software development away from carriers and vendors.

 
Man, the RIVA128 does indeed bring back good memories. It was my very first 3D accelerator, and the OpenGL rendering of Quake was really interesting when comparing to software rendering of the day. I had completely forgotten board manufacturers were building drivers at the time; damn how crazy WAS that era?
 
Man, the RIVA128 does indeed bring back good memories. It was my very first 3D accelerator, and the OpenGL rendering of Quake was really interesting when comparing to software rendering of the day. I had completely forgotten board manufacturers were building drivers at the time; damn how crazy WAS that era?
riva tnt was my first, before that i had a tseng labs et4000 svga thing which was a good 2d card for sure
 
My first "aftermarket" 2D card was an S3 864p 2MB VLB card in my 486 SX25 rig, which eventually got a bump to DX33, and then later one of the AM5x86 DX3/120 chips. Later when I moved to the Pentium MMX 166 era, I had a Matrox MGA-G200 of some flavor, I think it was a 4MB card. I yanked it for the RIVA128, but later tossed it back into another machine with the Voodoo card. Man I've gone through so many video cards since then :D
 
My first "aftermarket" 2D card was an S3 864p 2MB VLB card in my 486 SX25 rig, which eventually got a bump to DX33, and then later one of the AM5x86 DX3/120 chips. Later when I moved to the Pentium MMX 166 era, I had a Matrox MGA-G200 of some flavor, I think it was a 4MB card. I yanked it for the RIVA128, but later tossed it back into another machine with the Voodoo card. Man I've gone through so many video cards since then :D
my first cpu was a pentium 100 (pentium c? with mmx? who can remember) with EDO ram i think, those were the days
 
My first 3D accelerator was the rendition verite, not sure where that falls in the timeline although I had a 3dfx voodoo next and can't remember what my first all in one card was. I think the verite was in a pentium 90 and the voodoo in a celeron 100 oc'ed to 400, my first overclock.
 
TSMC N3E is most likely the process as it's too early for N3P. Will HBM4 be ready for mass production by Q4 2025 though?
as usual with Nvidia, it's a custom BEOL node. And yes HBM4 will be there. SK and Micron are competing hard. Samsung promises a lot but they are late
 
as usual with Nvidia, it's a custom BEOL node. And yes HBM4 will be there. SK and Micron are competing hard. Samsung promises a lot but they are late
All the processes used by major customers are more or less customized, others just don't bother giving them new names and so far nothing has indicated NVIDIAs is more customized than others
 
All the processes used by major customers are more or less customized, others just don't bother giving them new names and so far nothing has indicated NVIDIAs is more customized than others
Correct but note that TSMC N3 introduces FinFlex that allows more customization than before
 
Elon Musk is going bonkers now. Expanding his Colossus super cluster facility to 1 million NVIDIA GPUs. Colossus has 100K H100s right now, they are currently adding an additional 100K of H200, but now Musk is expanding that to 1 million, unknown whether they will be H100, H200 or GB200.

Elon Musk’s artificial intelligence start-up xAI has pledged to expand its Colossus supercomputer tenfold to incorporate more than 1mn graphics processing units, in an effort to leap ahead of rivals such as Google, OpenAI and Anthropic.

Work has already begun to increase the size of the facility in Memphis, Tennessee, according to a statement from the Greater Memphis Chamber on Wednesday. Nvidia, Dell and Supermicro Computer would also establish operations in Memphis to support the expansion, the chamber of commerce said, while it would establish an “xAI special operations team” to “provide round-the-clock concierge service to the company”.


Meanwhile, Meta just announced it will build a $10 billion artificial intelligence data center in northeast Louisiana, it will most probably be powered by NVIDIA as well.

 
Last edited:
Back
Top