You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
Last quarter we all read that there was a design flaw in Blackwell and most folks, a number of contributors here included, assumed it was with packaging. However, on the call Jenson clarified that there was a change to the mask and no functional changes necessary.
Vivek Arya -- Analyst
Thanks...
One major problem with making general statements about a particular culture is, generally speaking, they are not general. If we were to compare software development, for example, then the supposition would be flipped. Move fast and break things and all that.
It seems the media have latched on to the notion that the Blackwell delay has all but been confirmed by Nvidia themselves. Analysts seem to quote timelines between 3 weeks and 3 quarters and technical experts (these forums no less) agree that there is a plausible explanation for the issues...
I appreciate the response. Jensen did not refer to Moore's law in any way. I suppose it was more the notion that its the natural progression in expanding the capabilities of compute; as the physical limitations of increasing chip density raise costs the industry seems to have moved to, as you...
During last nights conference call for Q4 earnings Jensen stated that there were two fundamental changes in compute. Obviously one of those changes was Generative AI but his second pivotal change, according to Jensen, was that compute has now moved from General to Accelerated computing...
They mentioned VMware and by extension Broadcom so often on the last conference call that it just slipped out. To answer your question, I take it back, I do not think they are top two.
Couple question for the community:
#1
I'm sure many of you saw that the Biden Administration confirmed their expansion of export restrictions. https://www.reuters.com/technology/biden-cut-china-off-more-nvidia-chips-expand-curbs-more-countries-2023-10-17/
Given that NVDA worked around the...
Digitimes reports that TSMC needs to up their game in packaging and is looking towards memory vendors as a...source of inspiration.
https://www.digitimes.com/news/a20220601PD215.html?mod=2
I'm not sure I see this as any kind of event for NAND memory producers but I'm curious what the take is...
I would have to respectfully disagree with some of the negative sentiment here about Pat and the comments he has made.
The comment about subsidies indeed included South Korea. If you do a quick google you can see that they are in fact offering to subsidize up to 50%. However, if memory serves...
What is the scope of the current META activities as it relates to EPYC servers? Is Facebook going to migrate everything to AMD(sorry but i throw up a little bit when I say Meta - can't put Stephenson and Zuckerberg in same sentence)? Or is the pledge to host META operations on EPYC processors...
How many EUV lithography machines does Intel need? Numbers out of ASML suggest minimal expenditure from North America at roughly 1/4 of Taiwan or S. Korea and a backlog of over 40 units. If they are using TSMC for 2023 high end chips where does that leave them from a timing perspective? With...
