3 ms·

>there's no guarantee that they could match the software side of it. How many times in history has this ever happened? That a product was so good no competito

>there's no guarantee that they could match the software side of it. How many times in history has this ever happened? That a product was so good no competitors could match it, forever?

xenospn 2y ago
Apple has been doing this for decades with multiple product lines.
- matthewdgreen 2y ago
  Apple controls the customer interface and has a huge network effect. Consumers of AI services (when they exist in financially significant numbers) don’t care which hardware processes their inference (or trains their network), they only care about speed and price. A better example might be Intel, but Intel at least had a unique instruction set. And of course Intel is more of a cautionary tale than a story of eternal market dominance.
mgh2 2y ago
Is Nvidia's technology so advanced that others can't copy them? Is their hardware (GPU+CPU chips) [1], software (CUDA) or its combination their moat? Analogy: Apple's hardware + software combination as competitive advantage Are they just the beneficiaries of inside (R&D) or outside (AI) market forces? History has shown that the latter is not sustainable (dot-com, nanotech, covid, crypto, AI) TLDR: Shovels eventually become commodities [1] https://www.youtube.com/watch?v=Z1RrVgI_HQE https://www.youtube.com/watch?v=Z1RrVgI_HQE
- lokimedes 2y ago
  They snatched Infiniband a couple of years ago. The deep integration of low latency fabric with their GPUs is basically Nvidia owning the stack that makes GPU compute scalable beyond the single chassis. There’s a long road to commoditization.
  - paulmd 2y ago
    AMD has spent years developing MCM/multi-chip packaging to solve exactly this same problem. And they have Xilinx too, and Infinity Link (which is different from infinity fabric), and all these other things that help them build systems too. They aren't helpless babes unaware of the problem either. (although I agree with the point that yes, NVIDIA was thinking about systems engineering much sooner and more comprehensively than AMD was. And not just multi-node scaling, but also things like the programming model (CUDA vs OpenCL/AMD APP/HSA Framework), and they also seem to have correctly identified that ethernet allows higher bandwidth-density than pcie as a transport for this interconnect, giving them an implementation advantage. Similarly, despite all the praise for AMD's packaging, it's often come with downsides and caveats, and NVIDIA usually isn't far behind with one that doesn't. Eg Fury X/Vega/Radeon VII vs GP100/GV100/GA100, or MI300X vs B200. And NVIDIA has explicitly been much earlier-to-market on the systems-scaling stuff like NVSwitch or pre-specced high-performance systems modules like DGX, which AMD still doesn't really have an answer for despite NVSwitch being over a decade old at this point etc.) Tesla built a whole interconnect system for Dojo too, iirc, that was an explicit focus in their design. Not surprising for Jim Keller. But Jensen Huang nailed that requirement too.

>there's no guarantee that they could match the software side of it. How many times in history has this ever happened? That a product was so good no competitors could match it, forever?

xenospn 2y ago

Apple has been doing this for decades with multiple product lines.

matthewdgreen 2y ago
Apple controls the customer interface and has a huge network effect. Consumers of AI services (when they exist in financially significant numbers) don’t care which hardware processes their inference (or trains their network), they only care about speed and price. A better example might be Intel, but Intel at least had a unique instruction set. And of course Intel is more of a cautionary tale than a story of eternal market dominance.

mgh2 2y ago

Is Nvidia's technology so advanced that others can't copy them? Is their hardware (GPU+CPU chips) [1], software (CUDA) or its combination their moat? Analogy: Apple's hardware + software combination as competitive advantage Are they just the beneficiaries of inside (R&D) or outside (AI) market forces? History has shown that the latter is not sustainable (dot-com, nanotech, covid, crypto, AI) TLDR: Shovels eventually become commodities [1] https://www.youtube.com/watch?v=Z1RrVgI_HQE https://www.youtube.com/watch?v=Z1RrVgI_HQE

lokimedes 2y ago
They snatched Infiniband a couple of years ago. The deep integration of low latency fabric with their GPUs is basically Nvidia owning the stack that makes GPU compute scalable beyond the single chassis. There’s a long road to commoditization.
- paulmd 2y ago
  AMD has spent years developing MCM/multi-chip packaging to solve exactly this same problem. And they have Xilinx too, and Infinity Link (which is different from infinity fabric), and all these other things that help them build systems too. They aren't helpless babes unaware of the problem either. (although I agree with the point that yes, NVIDIA was thinking about systems engineering much sooner and more comprehensively than AMD was. And not just multi-node scaling, but also things like the programming model (CUDA vs OpenCL/AMD APP/HSA Framework), and they also seem to have correctly identified that ethernet allows higher bandwidth-density than pcie as a transport for this interconnect, giving them an implementation advantage. Similarly, despite all the praise for AMD's packaging, it's often come with downsides and caveats, and NVIDIA usually isn't far behind with one that doesn't. Eg Fury X/Vega/Radeon VII vs GP100/GV100/GA100, or MI300X vs B200. And NVIDIA has explicitly been much earlier-to-market on the systems-scaling stuff like NVSwitch or pre-specced high-performance systems modules like DGX, which AMD still doesn't really have an answer for despite NVSwitch being over a decade old at this point etc.) Tesla built a whole interconnect system for Dojo too, iirc, that was an explicit focus in their design. Not surprising for Jim Keller. But Jensen Huang nailed that requirement too.