Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : AMD, ARMH, INTC, NVDA -- Ignore unavailable to you. Want to Upgrade?

To: Vattila who wrote (33781)	10/31/2019 9:33:19 PM
From: neolib	Read Replies (1) \| Respond to of 73147

Yeah, IMHO, Nvidia has a problem because they have huge dies at the top end. Those could not yield on 7nm, which I think is partly why Nvidia has stayed stuck on 12nm. BTW, Jensen at one of the previous 2 CCs (forget which one) made the same statement that Lisa did more or less: Arch matters more than process. He was reacting to AMD's 7nm rollout and was saying that Nvidia has higher performing parts with better perf/W than AMD despite being a node back, and that this was due to their superior arch. Ok, fine, but if you took that same arch and had it a node ahead would it not be even better? Should be, but if the die size is too large, then it won't yield at the leading edge, and Nvidia has had humongous die sizes for their largest parts (like 800mm2 IIRC). I think the fine grained parallelism in GPUs makes chiplet approach take a hit, but you would have to have someone who knows more about that than me comment on it. I wish someone would post AMD vs Nvidia GPU benchmarks normalized to die size. I think that might be interesting....