Thought for the day: Doing Everything Right

Throwing out a thought for the day here:

Lisa Su is doing everything right and still can’t win.

Winning against Intel required Intel to make a series of major mistakes. Nvidia is not making major mistakes. What remains is the opportunity they can get by pushing uphill against an enormous force… they can grow into a growing market but I don’t see how they can displace Nvidia in many opportunities.

Likewise, in CPUs they now face Intel starting (barely) to get their act together, and the Arm CPUs are taking more and more of the hyperscaler footprint.

Not sure immediately what this means for the current share price. Just thoughts about the opportunity in front of AMD.

I can’t immediately see a 5x from here, and I can’t imagine there aren’t more probable paths to a 2x, so I am thinking to pull some money out. Been losing too much lately.

1 Like

My impression is all of them are priced for significant improvement and growth. Even though AMD exceeded expectations we saw a significant drop due to the forecast. The market is very nervous, just like you are.
Alan

Nvidia is not making major mistakes.

NVIDIA has issues with blackwell. Their large parts are not yielding. Hardware wise AMD is in great shape

Nvida earning are due Feb 26. Then we might learn a lot. Until then its probably silent period for them. They cannot respond to comments. Rumors are free to fly. Shorts and hedge funds find it an exciting time.

2 Likes

I thought the yield issue was resolved long since.

Nvidia’s yield-killing design flaw in its Blackwell GPU was fixed months ago, and a refined version of the B100/B200 processors is about to enter mass production. Jensen Huang, Nvidia’s CEO, admitted this week that the flaw was entirely caused by Nvidia and said that the company’s production partner TSMC helped fix it in a timely manner, according to Reuters.

“We had a design flaw in Blackwell, it was functional, but the design flaw caused the yield to be low,” Huang said. “It was 100% Nvidia’s fault.”

Nvidia’s Blackwell B100 and B200 GPUs link their two chiplets using TSMC’s CoWoS-L packaging technology, which relies on an RDL interposer equipped with local silicon interconnect (LSI) bridges (to enable data transfer rates of about 10 TB/s). The placement of these bridges is critical. However, a supposed mismatch in the thermal expansion properties between the GPU chiplets, LSI bridges, RDL interposer, and motherboard substrate caused the system to warp and fail, and Nvidia reportedly had to modify the top metal layers and bumps of the GPU silicon to enhance production yields. While the company did not disclose specific details about the fix, it did mention that new masks were required.

The now-fixed Blackwell GPUs for AI and supercomputers will enter mass production in late October and should start shipping early next year (which will still be Nvidia’s fiscal year 2025).

2 Likes