Recent

% | $
Quotes you view appear here for quick access.

Intel Corporation Message Board

semi_equip_junkie 207 posts  |  Last Activity: 11 hours ago Member since: Feb 16, 2012
SortNewest  |  Oldest  |  Highest Rated Expand all messages
  • Reply to

    semi_punk... call your today's bluff

    by getanid61 Mar 10, 2015 12:08 AM
    semi_equip_junkie semi_equip_junkie Mar 10, 2015 12:45 AM Flag

    Intel reinvented the foundry system for x86 servers - a truely custom made (customer specific) solution -
    in the meantime Geee61 turns blue.... and keeps choking

  • Reply to

    semi_punk... call your today's bluff

    by getanid61 Mar 10, 2015 12:08 AM
    semi_equip_junkie semi_equip_junkie Mar 10, 2015 12:31 AM Flag

    have fun - only a guy like you could make stuff like this up

    jamulmike • 7 hours ago

    0 users liked this posts users disliked this posts 0
    Reply

    More on shrink pressure:14nm Xeon microserver SoC

    The race to 10nm will be simply brutal.
    Fewer doing Moore and More Than Moore.
    All all (Cymer)ASML all the time.

    PORTLAND, Ore. — Intel's first Xeon system-on-chip (SoC) is twisting ARM's microserver ambitions, Charles King, principle analyst at Pundit-IT in Hayward, California told EE Times.

    "Its not just Intel's first SoC in the Xeon family," King told us. "Its the beginning of a new era at Intel — expect them to move very fast in SoCs. We are going to see many more SoCs specifically designed to combat ARM microservers plus serve many datacenter functions."

    Today's Xeon SoCs are designed for hosting and cloud services such as web hosting, memory caching, dynamic web serving, and warm storage. But the future Xeon SoC's to which Hayward refers will be optimized for storage and network-optimized products such as storage-area networks (SANs) and network attached storage (NAS), mid-range routers, wireless base stations and embedded IoT devices. About 75 percent of current designs-wins for the Xeon D are for network, storage and IoT designs, whereas microservers are under development at Cisco, HP, NEC, Quanta Cloud Technology, Sugon and Supermicro.

    Intel's previous strategy to best ARM in microservers was to beef-up its Atom line with its second-generation Atom processor C2750, but no more, according to King — "Atom will become a consumer only SoC," he told us. The Xeon D-1500 family delivers 3.4-times faster performance per node and up to 1.7-times better performance per watt, which will also make it useful in high-end IoT devices that ARM cannot match, according to Intel. So far the only ARM 14-nanometer 64-bit core was made by Intel for Altera. Samsung has also shown a 14nm ARM 64-bit prototype core, but release no details or delivery estimates. The closest ARM has come to the Xeon D 14nm 64-bit cores are those made with TSMC's 16nm process, which are not due out until later this year.

    The Xeon D-1500 was actually announced last year, but Intel has taken its t Less

  • semi_equip_junkie semi_equip_junkie Mar 9, 2015 11:20 PM Flag

    BURP
    WASHINGTON—March 2, 2015—The Semiconductor Industry Association (SIA), representing U.S. leadership in semiconductor manufacturing and design, today announced that worldwide sales of semiconductors reached $28.5 billion for the month of January 2015, the industry’s highest-ever January total and an increase of 8.7 percent from January 2014 when sales were $26.3 billion. Global sales from January 2015 were 2 percent lower than the December 2014 total of $29.1 billion, reflecting normal seasonal trends. Regionally, sales in the Americas increased by 16.4 percent compared to last January to lead all regional markets. All monthly sales numbers are compiled by the World Semiconductor Trade Statistics (WSTS) organization and represent a three-month moving average.

    “After a record-setting 2014, the global semiconductor industry is off to a promising start to 2015, posting its highest-ever January sales led by impressive growth in the Americas market,” said John Neuffer, president and CEO, Semiconductor Industry Association. “Global sales have increased on a year-to-year basis for 21 consecutive months and remain strong across most regions and product categories.”

    Regionally, year-to-year sales increased in the Americas (16.4 percent) and Asia Pacific (10.7 percent), but decreased in Europe (-0.2 percent) and Japan (-8 percent). Sales decreased compared to the previous month in Asia Pacific (-0.8 percent), Europe (-2 percent), the Americas (-3.3 percent), and Japan (-6.4 percent).

  • Reply to

    One more time for the ARM dummies

    by semi_equip_junkie Feb 28, 2015 10:54 PM
    semi_equip_junkie semi_equip_junkie Mar 9, 2015 11:12 PM Flag

    "while you emphasize the design aspect the rush to catch up with Intel's Trigate speaks for itself."

    Larrabee failed on the design aspect...
    Mobile lost $4B last year alone due to the design aspect...
    buy hey... they were all on superior fabs...

    you are the one bragging how TSMC et al is catching up with Intel in regards to process technology - what is it now little troll? process technology does not matter - make up your mind
    comes from a guy who stated Trigate would only apply to CPUs and FPGA's but not to SoC-
    why would I make stuff like this up - must come from somebody who is really "smart"

    how many billions did MS write off on RT - the best decision ever made by MS......

  • Reply to

    Chief

    by lost_my_shorts Mar 9, 2015 11:24 AM
    semi_equip_junkie semi_equip_junkie Mar 9, 2015 1:44 PM Flag

    it could be a non public start up - there might be some (non public) low hanging fruits - I don't believe it's MTSN

  • Reply to

    One more time for the ARM dummies

    by semi_equip_junkie Feb 28, 2015 10:54 PM
    semi_equip_junkie semi_equip_junkie Mar 9, 2015 1:41 PM Flag

    while you emphasize the design aspect the rush to catch up with Intel's Trigate speaks for itself.
    Also stacked memory via TSV is strictly a manufacturing / cost issue IMO

  • Reply to

    One more time for the ARM dummies

    by semi_equip_junkie Feb 28, 2015 10:54 PM
    semi_equip_junkie semi_equip_junkie Mar 8, 2015 12:53 AM Flag

    "design is the most important characteristic of a chip and always primarily determines its market and asp."

    let me rephrase a productionworthy design is the most important characteristic of a chip and always primarily determines its market and asp."
    for many years NVDA could not come up with a productionworthy design because just like you they did not comprehend the manufacturing aspect and they complained bitterly about TSMC - do you want to argue about this?

    why would anybody spend billions on an off the shelf ARM design - you tell me
    I think Mars was refering to Apple

    "I believe the backend is far more challenging although the front end "
    Really now... what have you done in the front end without being an engineer ??
    or is it another buzzword you have NO CLUE... NONE what it means..

    you don't know which side is up or down -
    you have ZERO understanding of the manufacturing aspect - that's your problem

    you claimed windows RT was the best decision ever made by MS and you come here and gloat how smart you are and everybody else is stupid - I think you completely lost your mind -

    compared to you nenni is smart - alot smarter than you - hate to admitt

  • Reply to

    One more time for the ARM dummies

    by semi_equip_junkie Feb 28, 2015 10:54 PM
    semi_equip_junkie semi_equip_junkie Mar 7, 2015 11:37 PM Flag

    The most important part about Intel's process is not its lead but the fact that PC/Server chips already pay for it meaning Intel's mobile chips can be sold very close to cost rather than having to give 40% to an external foundry like the ARM licensees.

    agree - the PC//server chips provide a critical mass (trailblazzing technically and financially the path for SoC) - PC (CPU) chips have already integrated graphics and the "integrated voltage regulator" has replaced discrete analog chips, memory is connected via 2.5 D and in the future we will see stacked memory connected via TSV (already implemented in server chips - correct?) . As Bohr stated the borderline between PC chips and SoC is becoming more fluent
    It's nice to have a "rational" conversations about those matters.
    I worked for many years in metal etch and I am familiar with many integration issues - like planarization (etch back , copper CMP).
    Of course copper is being used for many years but using 15 plus metal layers is no cake walk since those layers need to be defined via dual damascene and finally connected - the aspect ratios of the vias are "huge"
    and they need to be etched maintaining a precise angle and they need to filled back with diffusion barriers and copper and other stuff - I believe the backend is far more challenging although the front end gets tricky as well (Ion implants!!!!).
    If Intel can come up with a true 10nm (and Bohr hinted that 10nm will require quadruple pattering) Intel will have another ace - beside having processed a megaton of Trigate silicon
    everybody knows the more wafer processed the further ahead of the learning curve

  • Reply to

    Getanid 61 - do you get it? I doubt it

    by semi_equip_junkie Mar 7, 2015 12:30 AM
    semi_equip_junkie semi_equip_junkie Mar 7, 2015 12:11 PM Flag

    International Technology Roadmap for Semiconductors

    google it - feature size used to be expressed in microns

    the roadmap also includes defect density (defects per cm2) and number of metal layers (aka backend)

    for example in 1992 there were 3 layers common (Al)
    I believe IMB's server chip is North of 15 layers and of course copper.
    logic converted to Cu @ 130nm

    and now take your medication - I'll be out of the "office" for the rest of the weekend

  • Reply to

    Getanid 61 - do you get it? I doubt it

    by semi_equip_junkie Mar 7, 2015 12:30 AM
    semi_equip_junkie semi_equip_junkie Mar 7, 2015 2:35 AM Flag

    You have NO CLUE... NONE !!!! to 20nm design rules or 14nm design rules !!!

    so why did ALTR selectted Intel
    because they were bribed by Intel?
    ALTR has NO clue - you're dumber than I thought

  • semi_equip_junkie by semi_equip_junkie Mar 7, 2015 12:30 AM Flag

    Daane cited reports that other foundries are grafting a first-generation of FinFETs on to existing 20-nm design rules to create what they are calling a 14-nm node
    with best regards from semi punk

    Altera surveyed foundries for a year before striking the deal with Intel. It will continue to make chips at TSMC and conduct ongoing evaluations of other processes as they develop.

    Daane cited reports that other foundries are grafting a first-generation of FinFETs on to existing 20-nm design rules to create what they are calling a 14-nm node. “Intel’s 14-nm is a second generation FinFET process, while others are just starting to implement their first,” he said.

  • Reply to

    Nenni

    by semi_equip_junkie Mar 6, 2015 4:45 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 11:49 PM Flag

    are you ready to blow Geee61 - you used the term semi punk at least a hundred times ....
    take some Asprin - I don't want you to get a cardiac arrest

    than again ggogle ALTR 14nm FinFet -
    John Daane of ALTR is a lot better to explain it to morons like you than me
    enough fun G61
    Leck mich am #$%$

  • semi_equip_junkie semi_equip_junkie Mar 6, 2015 11:31 PM Flag

    Explain what is a 16nm backend... what does 16nm stand for ??

    Geee61 - I am NOT your mother little troll
    why don't you google ALTR 14nm FinFet
    John Daane explains why they selected Intel rather TSMC - only 2 years old news
    are you ready to blow or do you need some more egg in your face -
    BTW you keep repeating yourself.....

  • Reply to

    LOL getanid61 melts down to end the week.

    by wallisweaver Mar 6, 2015 7:45 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 10:26 PM Flag

    waldblow
    do you have anything to contribute to the board that would be of interest for Intel shareholders?
    Nothing I suppose - surprise, surprise
    and here is the difference between blowguy, poop, Jean , apology et al and Waldo

    waldo posts a lot - true- but at least there is some content

    and BTW
    Romney lost - get over it

  • semi_equip_junkie semi_equip_junkie Mar 6, 2015 9:33 PM Flag

    bump
    read TSMC roadmap about "10nm" and than shut up

  • Reply to

    Nenni

    by semi_equip_junkie Mar 6, 2015 4:45 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 9:27 PM Flag

    Explain what is a 16nm backend..
    "Explain what is a 16nm backend".
    that's your rear end - #$%$ mit Ohren" (google it)
    there is a BIG difference between you and Nenni:
    Nenni systematically tries to create the perception TSMC is matching Intel's performance
    you Gee61 are only a benchmark monkey - pretty clueless about the manufacturing aspect
    go and read what ALTR said about selecting Intel for 14 nm rather TSMC -
    you are so dumb - dumb and dumber
    ALTR spelled it out two years ago but you still don't get -
    why am I wasting time on a moron like you?

  • Reply to

    Nenni

    by semi_equip_junkie Mar 6, 2015 4:45 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 5:14 PM Flag

    10nm does not require quad patterning thankfully because it is expensive. 10nm has already been solved so don't bet against it.

    One has to "appreciate" that statement -
    TSMC 's "10nm FinFet" uses 16nm backend - the problem has been "solved" by not shrinking the backend (diesize).
    Bohr stated that 10nm will require some quadruple patterning because EUV won't be availble -
    truely amazing how this guy is trying to brainwash.
    Of course there is an optimization going on with regards to cost - cost advantage due to shrinking die size versus addiition capex

  • semi_equip_junkie by semi_equip_junkie Mar 6, 2015 4:45 PM Flag

    time to float that ALTR rumor again?
    You know what you are called in Germany?
    Schleimscheisser (Yahoo zensor too stupid to follow :-) !!!!

  • Reply to

    II love the "mood" on this board

    by semi_equip_junkie Mar 6, 2015 1:22 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 4:23 PM Flag

    10nm does not require quad patterning thankfully because it is expensive. 10nm has already been solved so don't bet against it.
    OH yeah - you're correct .....TSMC backend is lagging again (it's really 16nm @ 10nm)
    you're a PUTZ twisting everything as much as you can.

  • Reply to

    II love the "mood" on this board

    by semi_equip_junkie Mar 6, 2015 1:22 PM
    semi_equip_junkie semi_equip_junkie Mar 6, 2015 4:05 PM Flag

    Nenni - the usual BS
    I own 3500 shares of ASML and listen to LRCX....EUV won't be productionworthy ...anyhow ASML is a Win Win no matter what -
    only idiots like you and Geee61 are pimping 10nm
    TSMC can not even shrink the backend (only marginally) compared to 20nm planar....you think people are stupid?
    and if they I turn it into my advantage

INTC
32.41-0.32(-0.98%)11:29 AMEDT