though the citie analyst had some technical points that are valid I believe KLAC's outlook over next 12 to 18 months has improved relative to LRCX and the stock is beaten down leaving more room to appreciation while paying excellent divy.
Lam raised expectations while KLAC did the opposite though there are opportunities coming up - either EUV or multi patterning.
One reason KLAC missed out is because TSMC and Samsung 16nm and 14nm FinFet is really based on 20nm backend which allowed both to reuse equipment from 20 nm node - but this will change @ 10nm.
It does not seem there are issues for KLAC in regards to loosing share
If you read the "Purple Swan" article again,....
you would think one of those exotic "know it all" boutique brokerges like BlueFin would jump on this kind of stuff....
for sure you won't find it in Tiermans Barron's blog....
actually I think they know very well.....but the fact they ignoring it speaks for itself
The same assessment was also presented by Intel’s Jeff Groff from his synopsis of Intel’s Q2 call: “In summary, it seems that Intel is executing fairly well on the process technology side of the business considering the ever increasing difficulty of pushing forward with Moore’s Law. We can expect exciting new structures and materials (just maybe not at 10nm) and an increasing importance of 3-D structures in both logic and memory fabrication.” This resonates with our blog Intel Calls for 3D IC, and was recently voiced by Intel process guru Mark Bohr: “Bohr predicted that Moore’s Law will not come to an abrupt halt, but will morph and evolve and go in a different direction, such as scaling density by the 3D stacking of components rather than continuing to reduce transistor size.” Bohr’s ISSCC slide from earlier this year reasserts this:
The key two concerns regarding 3DIC stacking using TSV are (a) Cost, noted in the slide above “Poor for Low Cost,” and (b) Vertical connectivity, as voiced by Mark Bohr: “Intel’s Bohr agrees that 3D structures will become more important. He said the kind of through-silicon vias used for today’s chip stacks need to improve in their density by orders of magnitude.”
System Plus Consulting, sister company of Yole Développement (Yole), released this month its new reverse costing report, Samsung 3D TSV stacked DDR4 DRAM. In August 2014, Samsung announced the mass production of the first analyzed 3D TSV technology based DDR4 modules for enterprise servers. According to Samsung, this new module, because of its high density and high performance will play a key role in supporting the enterprise servers’ development and cloud-based applications, as well as further diversification of data center solutions.
prime example for street's stupidity ..remember G61 raving about TSMC capex exceeding Intel...
here's reality check out of the horse's mouth...that's typical TSMC "culture" big mouth.....
However, uncertainty over the timing and magnitude of leading edge capacity ramps for the two major foundries continues to create headwinds for foundry CapEx in the remainder of calendar year 2015.
Don't worry about the "interface" ....Intel already implemented stacked DRAM connected via TSV.....
Micron is top notch battling successfulyl Samsung and Hinix - ARM just does not have the knoweledge ....
here you have Intel and Micron's brains combined
As one of the largest bottlenecks in modern computing is the slow transfer between processors and memory, TSV would make for an enormous increase in overall system performance, likely even greater than the inclusion of 80 cores on a single die. Intel actually manufactured memory for a period in the early 80’s, but abandoned the business amid the boom in processor sales. Rather than manufacturing their own memory, however, Intel would likely partner up with memory manufacturers and sell the memory as a package deal with processor. Said Intel CTO Justin Rattner:
Finally analysts begin to see the light as to what is really the bottleneck - Intel already implemented stacked DRAM on high end server chips
There are immediate uses in replacing DRAM, write the authors, and also in solving processor “bottlenecks” for Intel:
While the companies do not expect XPoint to be a replacement technology for DRAM or NAND, but rather a product gap filler between the two, we believe initial applications will likely focus on replacing DRAM in enterprise applications such as in-memory database, where bringing massive amount of data closer to the processor could enable much faster analyses. We also think Intel is incentivized to enable the ecosystem for faster memory/storage to remove system bottlenecks and increase the appeal of high-end server processors. Over time, XPoint could also be used in PCs, although we think the cost could be a limiting factor.
a shrink - probably one reason why Intel pushed out 10nm.
Too complicated for A.E.
comment from Barron's
as mentioned TSV!!!
By the way, this new memory technology could also put ARM into jeopardy since Intel can put this memory stack on top of its CPU chip, since it does not generate as much heat as older DRAMs, so Intel’s SOC, CPU etc. will have CPU and memory combined as one unit, much lower power consumption than any ARM with DRAM combine for mobile and server.
I think it's closer to a laptop because of the larger screen and it could favor Intel silicon (NO WINOWS RT for phablet folks !!!). And you probably wonder why this story (IC Insight is fairly reputable) ) never made the headlines...because "the street" controlls the financial media - but you can still beat them.
Even AMD yu aka pumper K missed that completely - sorry AMD you and your P...poor assessment about Greece - Merkel sucked it to you
The phablet segment is expected to account for 17 percent of total smartphone shipments in 2015, which are forecast to be about 1.5 billion handsets. The Update report shows phablets representing 21 percent of the 1.7 billion smartphones that are forecast to be shipped in 2016. Phablet sales are projected to reach 30 percent of the nearly 2 billion total smartphones shipped in 2018, according to the Update of the 2015 IC Market Drivers report.
Tablet unit sales have nearly stalled out because incremental improvements in new models have not been enough to convince owners of existing systems to buy replacements. More consumers are opting to buy new large-screen phablets instead using both a smartphone and tablet. Large smartphones have gained traction because more handsets are being used for video applications (including streaming of TV programs and movies) in addition to Internet web browsing, video gaming, GPS navigation, and looking at digital photos.
The market for large-screen smartphones received a boost from Apple’s highly successful iPhone 6 Plus handset, which started shipping in September 2014 and continued to gain momentum in the first half of 2015. Apple joined the phablet movement somewhat belatedly, but its 5.5-inch display iPhone 6 Plus smartphone played a major role in the company shipping 61.2 million iPhone handsets in 1Q15, which was a 40 percent increase over the same quarter in 2014.
SAN JOSE, Calif. - Micron Technology Inc. has begun producing limited NAND wafers at its new 300-mm fab in Singapore, but the company is missing an important piece of the puzzle: Intel Corp. The chip giant is not investing in the Singapore fab, as previously thought.
For years, Intel and Micron have had a successful NAND manufacturing venture called IM Flash Technologies LLC. The venture has one 300-mm fab in Lehi, Utah, which is producing 25-nm NAND devices. Both Intel and Micron share the output.
For some time, IM Flash has been readying a new fab in Singapore, dubbed IM Flash Singapore (IMFS). The fab is producing some wafers, with mass production due in mid-2011.
March 3rd, 2015 at 4:59 am - Author Anton Shilov
Micron Technology on Monday began the expansion of its Singapore NAND flash memory fabrication facility. The company started to build a new fab that will cost $4 billion and will start production in fiscal 2017. The new fab is expected to produce innovative 3D NAND flash (co-developed with Intel) as well as other types of future memory.
Micron already operates four major manufacturing facilities in Singapore and employs around 7 thousand people there. The new fab 10X will be a 255,000 square foot expansion for the current fab 10N. The manufacturing complex will be able to produce around 140 thousand 300mm 3D NAND flash wafers per month, which is a lot. Moreover, the new fab will support 40 – 50 per cent bit growth average per annum over an extended time horizon. In addition, the space will enable production of storage class and other memory technologies.
this might have been true 20 - 30 years ago when the industry depended strictly on DRAM upgrade cycles.
the industry became very diversified - chips are now the nuts and bolts of the global economy.
Currently the transition to SSD is a huge driver -
it's all about "tons of data" that are either stored via SSD or moved via servers.
Robert Maire is not up to date -
prior to Y2K global chip sales were cyclical which was mostly caused by over capacity and dependency on one driver (DRAM) -
nowadays a few players control capacity which makes it also less cyclical
No - I am buying again - beside my core positions- I always keep core and try to avoid shortterm tax rate
Intel placed a volume purchase agreement of a minimum of 50 tools with ASML and placed an order of 6 EUV machines (ASP on the order of $ 120 to 150 million each).
KLAC now 4% dividend and could become a buyout target for LRCX - there is zero overlap between both
I owned ETEC when bought out by AMAT (the price doubled overnight)
I owned Varian and I owned NVLS and a little bit Cymer
all part of early retirement plan - I don't hang out here 24 by 7 like you do.
and I pay attention :several years ago 3 to 5 the old man at TSMC stated that capital intensity is going up
people don't seem to realize a "good shake" could disrupt the complete TSMC eco system for several weeks - it's insane to cluster all the fabs on this tiny island.
Last time it happened about 15 -20 years ago - at this time everything was more diversified.
Using GF or Samsung as a backup is expensive because it requires two designs for the same chip.
Fab 42 could be ready for NAND flash - SSD will take everything over
No need for owning MUth
Intel Micron JV works well
perhaps they will use fab 42
the hole MU buyout is a hoax
Intel is using multiple pattering @ 14 nm and the yields are super excellent - unless there another major change "conventional litho" is most likely not the issue ... to me it points to EUV.
BK is a manufacturing guy and MP versus EUV is a classic manufacturing issue -
EUV is certainly more risky ....and BK is smart enough not to tell the "financial community" -
sometimes it's better to shut up and take the hit...
Pour A.E. ...rather coming up with some constructive stuff as what could be behind the delay he's telling Intel how to run the business - no wonder he can't find a job anywhere else.
Hard to believe Intel loosing its lead overnight