Recent

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

Intel Corporation Message Board

you are viewing a single comment's thread.

view the rest of the posts
  • ltisteve@verizon.net ltisteve Nov 6, 2011 11:10 PM Flag

    Some interesting observations about ARM

    >>>So Steve, are you ready to get it through that thick skull of yours that Intel is going to have major problems at the hands of ARMH?


    Well, no, I don't see that at all. 32nm means lithography, a new step for the ARM fabs to perfect, though Intel got it down fairly quickly. At 28nm there is gate leakage that ARM has yet to address. ARM's 28nm is going to come out next year though they say it won't be a major part of the market for a year or two, and yet they need to sell 100 million units to break even.

    The ARM fanboys will always be quick point out how their technology is superior. What I am seeing, in a real world example is that the multi core SoC's don't add a lot to the equation in real world terms to the cell phone. The apps are light and the graphics processor seems to do more

    I am going to use some ARM lingo here. "Good enough" as in the x86 for smartphones will be good enough, especially if there is a good GPU associated with the Intel processors.

    $24 a share will be history, that won't be a historic high by any stretch of the imagination. I am fairly confident you'll see Intel in the 30's next year.

    Next year there is going to be a whole lot of capacity coming online. Intel seems to feel like they can make a go of it with x86 processors.

    But, just for argument sake lets imagine that Intel's processor for phones are horrible. Let's say they are twice as bad the ARM fanboys would lead us to believe. What would Intel do? What if they cranked up their 32nm plants to make ARM processors, wouldn't that be a hoot? Yes, I've seen other ARM fans suggest it. Why not price them at $1 more than the lowest cost in the market? They can't be pegged for dumping products in other markets, and think about it, Intel could be making the Cortex A15 right away. Imagine that? And then with Trigate they could run 22nm and get a fairly good price for it because nobody else would have 22nm and won't for a long time. They could get their 60% margins.

    What would that do to the other ARM fabs? Global Foundries would be gone, and TSMC would be on it's knees. Give it a few years and ARM would be at the mercy of Intel, who could dispose of them whenever they felt like it.

    Still think $24 is a high?

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • You are DESPERATE !!!!

      "Well, no, I don't see that at all. 32nm means lithography, a new step for the ARM fabs to perfect, though Intel got it down fairly quickly. At 28nm there is gate leakage that ARM has yet to address. "

      OK, so your background isn't in semiconductors... SHouldn't put you down for it, but quit acting like you have a clue...

      "What I am seeing, in a real world example is that the multi core SoC's don't add a lot to the equation in real world terms to the cell phone."

      Keep looking backwards...

      "I am going to use some ARM lingo here. "Good enough" as in the x86 for smartphones will be good enough, especially if there is a good GPU associated with the Intel processors."

      all there is to it... huh

      "Intel seems to feel like they can make a go of it with x86 processors. "

      Gee... thanks for your insight

      "just for argument sake lets imagine that Intel's processor for phones are horrible. "

      HOw many phones have them now... just for argument sake

      "Let's say they are twice as bad the ARM fanboys would lead us to believe. What would Intel do? "

      INtel will claim they are in it for the long haul, and something is coming next year...

      "Intel could be making the Cortex A15 right away"

      Yes, how long would that take...
      Or... you don't have a clue !!!

      "What would that do to the other ARM fabs? Global Foundries would be gone, and TSMC would be on it's knees. Give it a few years and ARM would be at the mercy of Intel, who could dispose of them whenever they felt like it."

      Sure... just like INtel killed NVidia with Larabee...

      keep the faith...

      • 1 Reply to getanid61
      • ltisteve@verizon.net ltisteve Nov 7, 2011 2:15 AM Flag

        Oh yeah, Larabee. Andrew Cuomo, does that name ring a bell? Governor of New York? Remember when he was Attorney General of New York bringing up lawsuts against Intel, to keep Intel out of the graphics card business? How did Intel fight back? Sandy Bridge. They put the GPU in the die of the processor and cut NVidia out. But wait, oh yeah that's right Intel makes sucky graphics, right? AMD has much better graphics, right? So they threw together their version of a Sandy Bridge with superior graphics and a five year old processor design and called it Llano. Except, wait for it, Global Foundries couldn't get the 40nm production down right and botched their 3rd quarter.

        Oh, by the way, Apple went with Sandy Bridge and it's sucky graphics in the Macbook Air because, as it turns out, it's good enough. How much discrete graphics card business do you really expect Nvidia to pull off in the future? How much longer before they give up and sell off their graphics card division? I say two years max.

        Intel is expanding it's fabs, next year just one of them will produce 90,000 wafers a month at 22nm in full ramp up by next year.

        "http://www.conceivablytech.com/513/science-research/intel-confirms-production-of-22nm-processors-for-late-2011"

        So, since you are so bright, what should Intel do with all this capacity? As far as we know it's all going to be x86 right now. But, yes, Intel can license ARM processors and make them in their fabs. Isn't this what you want? You would win, and you could tell everybody that ARMH has the superior design.

        But, realistically, what would all this excess ARM capacity do to the competition? Do you really expect Nividia to survive? Global Foundries? TSMC would struggle as well. Didn't they say that the break even was 100 million 28nm processors for TSMC? Yikes!

        The worst part of all, it would all be legal. Unless of course, ARMH threw some kinda of injunction on Intel to not license their Cortex A15 design.

        The thing that you can't explain away is that a mind boggling amount of capacity is coming online from Intel next year. Tons and tons of processors coming into the market from Intel. Don't pretend for a moment that Intel will be locked out of the phone market.

        The worst part about Intel from a competition point of view is that they are secretive. How is Intel going to find a home for all this excess capacity? Clearly the management at Intel knows something that you and I don't. They tend to spring the announcements on the industry at CES in January.

        Intel made this huge investment in FABS with the intent to go after the phone market. Do you really think that they will be locked out? The word on the street is that Intel is having a problem keeping up with demand. Sure some of this capacity will go to other things. But, just the same, I don't know of any FABS at Intel that sit by idle.

        Intel, 22nm next year. Go ahead tell us how it's not going to work, and why the stock is never going to see more than $24. Please explain to all of us with thick skulls and no EE degrees how this plan is going to fail.

 
INTC
30.23+0.20(+0.66%)May 23 4:00 PMEDT