Recent

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

Intel Corporation Message Board

  • SortNewest  |  Oldest  |  Most Replied Expand all replies
    • 'Intel has over 80 percent market share in both servers and supercomputers.'

      http://www.idc.com/getdoc.jsp?containerId=prUS22998411

      'The market for non-x86 servers, including servers based on RISC, EPIC (Itanium-based), and CISC processors, increased 23.3% year over year to $4.8 billion in 2Q11. This is the third consecutive quarter in which non-x86 servers have exhibited positive growth and the second consecutive quarter that non-x86 based system revenue has grown faster than the market overall. Growth in non-x86 server revenue was driven by improved demand for Unix servers and IBM System z platforms.

      Demand for x86 servers continued to improve in 2Q11, with revenues growing 15.1% in the quarter to $8.4 billion worldwide as unit shipments increased 5.4% to 1.9 million servers. '

      x86 only has 64% market share, once again you are making things up.


      'If you want to have a serious discussion, that's fine. But I've grown tired of your inability to conduct a serious discussion. You have an alarming tendency to resort to character assassination instead of documenting your views. You fabricate information whenever you deem it to be convenient. You state your opinions as if there were a body of facts in evidence confirming them. There never is.'

      LOL, THIS DESCRIBES YOU TO A T !!!!

    • "Maybe PCs but not servers or supercomputers. IBM mainframes still sell well, Unix servers still well, gpus are in supercomputers etc. Volume is what helped Intel then and what will help ARM now."

      ***

      More of you making things up. Why? I guess because you simply don't know how to check the facts. Intel has over 80 percent market share in both servers and supercomputers.

      I'll give you one more chance to stop with the dishonest responses. After that I'll simply put you on the ignore list.

      If you want to have a serious discussion, that's fine. But I've grown tired of your inability to conduct a serious discussion. You have an alarming tendency to resort to character assassination instead of documenting your views. You fabricate information whenever you deem it to be convenient. You state your opinions as if there were a body of facts in evidence confirming them. There never is.

      I've had enough of it. It's your call what's next...

    • 'Yeah, because you can change the laws of physics with micro-architecture. You've gone from fuzzy logic to the ridiculous.'

      If you have a more efficient architecture you are doing less electrical work therefore burning less power. Despite all the Otellini boasting revolutions in actual delivered x86 processor power were not unleashed by going from 45m to 32nm to 22nm. There is still no sub 10W top line x86 processor (eg Sandy/Ivy Bridge) and Atom won't be 90mw at 22nm unlike a MIPS Xburst1 processor at 65nm which is 3 times the transistor size !

      Micro-architecture dominates processor characteristics, the facts unlike all your arm waving are undeniable ! Any ridiculousness is only coming from you who has not once discussed actual processor characteristics instead retreating to generalities and public pronouncements from your leader which is poor arguing and poor DD.


      'Otellini says it's going to be all about the physics. I think I'll go with him on this one.'

      He also touted Itanium strongly under Barrett as the future server leader and look how that turned out with the processor only being kept alive by HP payments. Larrabee was touted under him by Gelsinger, again a total failure as a gpu.


      'After all, Intel has crushed competitors in PCs, data center/servers and supercomputers.'

      Maybe PCs but not servers or supercomputers. IBM mainframes still sell well, Unix servers still well, gpus are in supercomputers etc. Volume is what helped Intel then and what will help ARM now.


      'ARM and mobility are up next'

      Not at 32nm and probably not at 22nm either considering how far behind Atom is in performance/power/die size.

    • Yeah, because you can change the laws of physics with micro-architecture.

      You've gone from fuzzy logic to the ridiculous.

      Otellini says it's going to be all about the physics. I think I'll go with him on this one. After all, Intel has crushed competitors in PCs, data center/servers and supercomputers.

      ARM and mobility are up next...

    • Micro-architecture has always been more important which I clearly demonstrated using actual data by a 65nm MIPS cpu being 1/7 the power of a 45nm Atom cpu of similar performance. You on the other hand are just flapping and waving your arms and hands about shouting 'fabrication ! fabrication !' and hoping you will convince similar technical illiterates.

    • "No it isn't and that fantasy just ain't going to happen no matter how many times you click your heels. 2012 is going to look ugly for single-core 32nm Medfield when it faces faster dual-core A15s/Kraits and quad-core A9s."

      ***

      My fantasy? Your fantasy is that fabrication doesn't matter. The major dynamic in the processor business you say is a fantasy. Companies out their spending billions of dollars on what you say is a fantasy. The entire ordering principle of the computer business you say is a fantasy.

      You might as well say up is down or black is white. I don't care how you package it, fanboi - no one and I mean no one is going to buy the idea that fabrication doesn't matter.

      I'm thinking if you keep saying that fabrication doesn't matter that your brain will turn to mush and run out your nose. Or perhaps it already has...

    • "Intel is hugely focused on the power issue and they are going to use their fabrication advantage to reel ARM in. It won't happen overnight but it doesn't need to."

      Atom is a prime example of Intel being 'hugely focused on the power issue' and it still fell short. There are technical reasons why x86 will always struggle to match RISC in power consumptions :- x86 decoding, less logical registers, more physical registers, more aggressive instruction re-ordering, CISC instructions doing more work etc etc.


      "The bottom line is that the fabrication dynamic is in play and it's going to grind ARM to powder."

      No it isn't and that fantasy just ain't going to happen no matter how many times you click your heels. 2012 is going to look ugly for single-core 32nm Medfield when it faces faster dual-core A15s/Kraits and quad-core A9s.

    • Yeah, yeah, yeah. Your constant mantra. ARM doesn't need state-of-the-art fabrication.

      Well they do. Intel is hugely focused on the power issue and they are going to use their fabrication advantage to reel ARM in.

      It won't happen overnight but it doesn't need to. The bottom line is that the fabrication dynamic is in play and it's going to grind ARM to powder.

      So, go ahead and cling to your "the water isn't at my doorstep yet" philosophy. At some point in the future you'll be standing on your roof wondering why you didn't invest in a boat...

    • Do you think a 14nm Atom will make a 20nm ARM processor obsolete ? A 1 GHz 65nm MIPS chips has a TDP of 90mw, a 0.8 GHz HKMG 45nm Atom of similar performance has a TDP of 650mW on a much bigger die size, so why exactly does any RISC chip need to be on the same process size ? You are touting a process advantage that is quite irrelevant considering the much different characteristics of x86 and Risc cpus.

      http://ark.intel.com/products/35472/Intel-Atom-Processor-Z500-%28512K-Cache-800-MHz-400-MHz-FSB%29
      http://en.ingenic.cn/product.aspx?CID=9

    • He's the little brother that has to be heard (under his getanid61 and marisvan aliases).

      Poor little brother, always wants to be seen and as a result must cry the loudest.

    • View More Messages
 
INTC
32.08-0.270(-0.83%)Apr 24 4:00 PMEDT