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!
Having just got done listening to the Micron quarterly presentation, it seems we will see "meaningful" production of the 3dXpoint memory before the end of the year. Although they used different nomenclature, it seemed they were talking about 3dXpoint calling it 1D memory and referring to it as their advanced memory. I'll listen to the call again when I'm not as tired and give an update if needed. This sounds like Intel/Micron ran into some serious problems either as to production, costs or some technical issues that they feel they have somewhat under control. Only time will tell. It would be nice if both Intel and Micron had more transparency before making premature announcements.
While doing my morning reading on MU, I ran across some interesting information, Intel may end up losing ownership of 3dXpoint completely at the beginning of 2018, with MU having buy out rights. With the delays, MU could take control as the market really starts to take off. I wish the facts around this where more transparent, but I'll have to do some more research.
Just read through the MU conference call transcript... it was like analysts were not even allowed to ask questions about 3D Xpoint. Instead, there were some strange allusions to the "mix" in 2018, and any of those questions around product "mix" were given non answers. Other analysts might have been trying to get a sense of where Xpoint was without asking directly by asking about capex in 2018, and those questions also received non answers.
IMO, the lack of transparency here is borderline an SEC issue, where something could be very wrong with 3D Xpoint and MU could be failing to disclose it to investors, or worse, selectively disclosing to analysts and telling them not to ask about it.
Arthur, I'm not sure what you mean by 1D nomenclature. I thought they were referring to 1x DRAM, which is their DRAM on what sounds like a ~10nm range node.
While doing my morning reading on MU, I ran across some interesting information, Intel may end up losing ownership of 3dXpoint completely at the beginning of 2018, with MU having buy out rights. With the delays, MU could take control as the market really starts to take off. I wish the facts around this where more transparent, but I'll have to do some more research.
From your Seeking Alpha link, 2014 Micron Annual Report (page 80): "Since inception in 2006 through August 28, 2014, we have owned 51% of IMFT, a venture between Intel and us to manufacture NAND Flash memory products and certain emerging memory technologies for the exclusive use of the members. IMFT is governed by a Board of Managers and the number of managers appointed by each member to the board varies based on the members' respective ownership interests, which is based on cumulative contributions to IMFT. The IMFT joint venture agreement extends through 2024 and includes certain buy-sell rights with an Intel put right, commencing in January 2015, and our call right commencing in January 2018, pursuant to which Intel may elect to sell to us, or we may elect to purchase from Intel, Intel's interest in IMFT. If Intel elects to sell to us, we would set the closing date of the transaction within two years following such election and could elect to receive financing of the purchase price from Intel for one to two years from the closing date."
This is really a bit strange that only Micron has the right to buy Intel's 49% ownership in the joint venture and Intel only has the right to initiate a sale of its 49% ownership in the joint venture to Micron. If Intel sells its share, Intel is obligated to finance it if Micron asked. Is Intel too generous or Intel has no other choice?
Haha, IoT means literally everything. Just a giant, multi trillion dollar market just waiting to be disrupted, and our digital transformation consultants are eager to help your company take part! We can start with a small, multimillion dollar project with vague goals no measurable outcomes, and scale it from there.
From my latest reading it seams that wide distribution will not begin until 2018, which leads me to believe they are having production yield problems they are in the process of solving. Any other thoughts or views on this? They are selling some it seems on a limited basis for commercial use, but not large scale yet. Has anyone got actual hands on that can give us some real world information?
I agree with the downbeat assessment on 3D Xpoint. I think 3D NAND will come out on top in mainstream applications, with 3D Xpoint regulated to niches - and that's only if Intel is able to solve manufacturing and technical issues.
Here is what I consider to be the definitive answer to this question by someone who understands the full picture from the technical standpoint, including business applications and economic impact. I must thank the SemiWiki community for the leads that enabled me to get this answer. This is why I value SemiWiki and the people behind it so highly. I especially want to give Thanks to Daniel Nenni for his support and encouragement.
Here is what I consider to be the definitive answer to this question by someone who understands the full picture from the technical standpoint, including business applications and economic impact. I must thank the SemiWiki community for the leads that enabled me to get this answer. This is why I value SemiWiki and the people behind it so highly. I especially want to give Thanks to Daniel Nenni for his support and encouragement.
There is so much wrong with this rebuttal, for starters, they outright say they haven't even read the full article they are rebutting. Second, there is the line "Our disagreement is less about challenging any technical data,..." followed by a handful of examples of use cases that may or may not benefit from 3D Xpoint, and most of these use cases are either low value or niche. And third, the company that published this rebuttal has a Micron VP on it's advisory board.
Here is what I consider to be the definitive answer to this question by someone who understands the full picture from the technical standpoint, including business applications and economic impact. I must thank the SemiWiki community for the leads that enabled me to get this answer. This is why I value SemiWiki and the people behind it so highly. I especially want to give Thanks to Daniel Nenni for his support and encouragement.
I feel Micron might be developing this at a different rate for a significant part of the research effort might be integrating it into an Automata processor. In other words Micron and Intel might be looking at a longer game of a massive breakthrough. This is a long shot and just speculation, but who knows?
