Wed, Jul 30, 2014, 6:07 PM EDT - U.S. Markets closed

Recent

% | $
Click the to save as a favorite.

Intel Corporation Message Board

  • dnenni dnenni Jul 10, 2011 3:52 PM Flag

    Intel Twisting ARM?

    http://www.semiwiki.com/forum/content/620-intel-twisting-arm.html

    My briefing with Intel is on Wednesday morning.

    D.A.N.

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • gregory.lynn@rocketmail.com gregory.lynn Jul 13, 2011 4:39 PM Flag

      So, what about your briefing today at Intel?

    • ltisteve@verizon.net ltisteve Jul 11, 2011 7:48 AM Flag

      I am following along with what D.A.N. is saying. I do think that for most part Intel won't answer a lot of his questions because most of what he wants to know is proprietary information that most of the Intel staff isn't privy to .

      The thing that Intel knows, and what he kinda fears is that Intel doesn't always get it right. He may be 100% dead on right. Everything about Tri Gate may be wrong. TSMC may be closer to the right direction with their version of 3D. It doesn't matter. Intel can afford to get it wrong. Look at their history with AMD. AMD was known to offer better processors in the middle of last decade. The problem that TSMC probably realizes are:

      1. Intel can afford to battle for years in R&D. The battle is at R&D, not at the foundry. Even D.A.N. seem to recognize that ARM may run out of tricks sooner or later.

      2. Let's say that his research is right and his 3D design is better. Intel can reverse engineer it, improve it, and pay up their lawyers to remain in court for the next decade. Or write a check with all the smartphone revenue kinda like they did with AMD last year.

      3. Intel has channel relationships. TSMC does as well, but Intel's has been around for decades.

      4. Here's the one point that D.A.N. and the guys at TSMC realize as well. People like x86. Oh sure, Windows 8 will run on ARM, in emulation mode. They are not rewriting Windows OS to work on ARM. What I am suspecting is that when it comes to tablets people may spend a little extra, give up some battery life, and take an Ivy Bridge version of the i3 over any ARM processor in any configuration. Some people will need 20 hours of battery life, other can accept 3 to five hours if it means that all of Windows and maybe even OS/X runs on their tablet.


      I did like this line however

      "TSMC has chosen to wait until 14nm to bring 3D transistor technology to the foundry business." When is that? 2014 at the earliest?

      I don't know, I won't buy a future that X86 programs are left behind. Nor do I believe that companies like Microsoft and Adobe will rewrite code for the ARM processor. The ARM based laptops are coming out this year, let's see how well they sell. I think that ARM will die there, and then at the tablet level. Phones are a different story, for now.

      • 2 Replies to ltisteve
      • 4. Here's the one point that D.A.N. and the guys at TSMC realize as well. People like x86. Oh sure, Windows 8 will run on ARM, in emulation mode. They are not rewriting Windows OS to work on ARM. What I am suspecting is that when it comes to tablets people may spend a little extra, give up some battery life, and take an Ivy Bridge version of the i3 over any ARM processor in any configuration. Some people will need 20 hours of battery life, other can accept 3 to five hours if it means that all of Windows and maybe even OS/X runs on their tablet.
        -----

        Windows on ARM is native not an emulation. It's also the "full" OS not some "lite" version. MS even have the office suite running.

        At this stage we don't know how MS are going to handle the binary issue and how they are going to communicate this to to the consumer...

        I agree with you. If you can buy a x86 tablet/laptop that runs Win8 and legacy code, ARM is dead (in windows that is).

      • I agree on all points, well stated. As for TSMCs version of TriGate (FinFet), here is the post on it that started this whole thing:

        http://www.semiwiki.com/forum/content/609-tsmc-versus-intel-race-semiconductors-3d.html

        2014 sounds about right. Most semiconductor crystal ball predictions like this slip. Intel was smart in not talking about Tri-Gate until they had working silicon.

        Thank you for taking the time to reply.

        D.A.N.
        http://www.semiwiki.com/


        **************
        I am following along with what D.A.N. is saying. I do think that for most part Intel won't answer a lot of his questions because most of what he wants to know is proprietary information that most of the Intel staff isn't privy to .

        The thing that Intel knows, and what he kinda fears is that Intel doesn't always get it right. He may be 100% dead on right. Everything about Tri Gate may be wrong. TSMC may be closer to the right direction with their version of 3D. It doesn't matter. Intel can afford to get it wrong. Look at their history with AMD. AMD was known to offer better processors in the middle of last decade. The problem that TSMC probably realizes are:

        1. Intel can afford to battle for years in R&D. The battle is at R&D, not at the foundry. Even D.A.N. seem to recognize that ARM may run out of tricks sooner or later.

        2. Let's say that his research is right and his 3D design is better. Intel can reverse engineer it, improve it, and pay up their lawyers to remain in court for the next decade. Or write a check with all the smartphone revenue kinda like they did with AMD last year.

        3. Intel has channel relationships. TSMC does as well, but Intel's has been around for decades.

        4. Here's the one point that D.A.N. and the guys at TSMC realize as well. People like x86. Oh sure, Windows 8 will run on ARM, in emulation mode. They are not rewriting Windows OS to work on ARM. What I am suspecting is that when it comes to tablets people may spend a little extra, give up some battery life, and take an Ivy Bridge version of the i3 over any ARM processor in any configuration. Some people will need 20 hours of battery life, other can accept 3 to five hours if it means that all of Windows and maybe even OS/X runs on their tablet.


        I did like this line however

        "TSMC has chosen to wait until 14nm to bring 3D transistor technology to the foundry business." When is that? 2014 at the earliest?

        I don't know, I won't buy a future that X86 programs are left behind. Nor do I believe that companies like Microsoft and Adobe will rewrite code for the ARM processor. The ARM based laptops are coming out this year, let's see how well they sell. I think that ARM will die there, and then at the tablet level. Phones are a different story, for now.

    • Here a video http://www.sematech.org/videos/3D/sitaram_final.html

      As for there claim of faster speed cost...... its really not for high performace part. 3D stacking does not improve single chip performance nor lower is TDP nor change anything.


      Its main advantage is to allow different process for each part, like intel could use is standart process for is CPU and the SOC process for the rest of the chip. Make interconnection easier so there you could see a performance improvement, but like in a sandy bridge the ring bus is nowhere near a bottleneck. Its could allow you to reach higher frenquency. What its does not say is that from a termal disipation point of view this is a disaster. As the total area that are again the package HS is much lower per given transistor.

    • http://www.tezzaron.com/technology/3D_IC_Summary.html


      That should end the debate right here you are a very low skilled journalist that bring FUD and on top of everything you cant even get the difference bettween a transistor and packing technologie.

    • I would be interested in how much TriGate effectively reduces the size of a die for the performance segment of their products.

      It seems to me that there are a couple things happening now but not necessarily tied together:
      1. 22nm
      2. TriGate
      3. Ivy Bridge
      4. 450mm wafers

      I think all Ivy Bridge are supposed to be 22nm.
      I think all 22nm will be TriGate, so all Ivy Bridge will therefore be 22nm TriGate.

      Going to 22nm provides a reduction in die size and therefore an increase in die-sites per wafer. There should be an additional increase in die-sites per wafer from TriGate.

      • 1 Reply to alexander.dumbass
      • Great question, thank you. It is on my list.

        450mm is a very interesting topic. I'm researching it right now and will write about it soon.


        ***********
        I would be interested in how much TriGate effectively reduces the size of a die for the performance segment of their products.

        It seems to me that there are a couple things happening now but not necessarily tied together:
        1. 22nm
        2. TriGate
        3. Ivy Bridge
        4. 450mm wafers

        I think all Ivy Bridge are supposed to be 22nm.
        I think all 22nm will be TriGate, so all Ivy Bridge will therefore be 22nm TriGate.

        Going to 22nm provides a reduction in die size and therefore an increase in die-sites per wafer. There should be an additional increase in die-sites per wafer from TriGate.

    • yawn.

      no real content there fudd boy.

    • Go away, self-promotion boy...

 
INTC
34.35+0.16(+0.47%)Jul 30 4:00 PMEDT

Trending Tickers

i
Trending Tickers features significant U.S. stocks showing the most dramatic increase in user interest in Yahoo Finance in the previous hour over historic norms. The list is limited to those equities which trade at least 100,000 shares on an average day and have a market cap of more than $300 million.