Quadcore / Duo Core question

Discussion in 'Hardware' started by Ridu, Apr 12, 2009.

  1. i7 920, at 2.66ghz can be easily overclocked to 3.8ghz, do a search on google to read all the tutorials.




     
    #11     Apr 13, 2009
  2. That's some serious overclocking - won't you need a major cooling solution? I have no experience with overclocking as Dell does not allow for this - but in general it makes me nervous... If you can overclock why doesn't Intel just sell a 3.8ghz CPU for the price of the 2.66ghz?
     
    #12     Apr 13, 2009
  3. JackR

    JackR

    As a guess:

    Intel fabricates a batch of chips. Then tests them all. Probably starts monitoring the processor for error as they increase the clock rate. Probably also monitor temperature rise. Chips are identified as they fail due to increasing clock rate/temp rise. Intel then derates chip to ensure a safety factor and stamps it with the appropriate part number. Might also be true of L2 cache capability and number of processors. Again, just a guess.

    Jack
     
    #13     Apr 13, 2009
  4. gnome

    gnome

    1. At a given price point, there is little performance difference between C2D and Quad... though at that price, the Quad will have the slower clock speed.

    2. It's more than just "Duo vs. Quad"... the mobo factors in, too.

    I recently did a fresh install on my C2D, E8300, 2.83Ghz, Vostro (budget mobo) and my Q9300, quad, 2.5Ghz, T3400 (x38 mobo).

    In spite of the slower clock speed of the quad... doing the fresh install, updating XP downloads, etc... it was significantly quicker.
     
    #14     Apr 13, 2009
  5. My i7 is running it at 3.8ghz, on air cooling. Once in a while, the computer wont POST. When it is running (99.9%) of the time, its very stable.

    Running it at a higher clock raises the temp the CPU can experience significantly. My max temp can reach as high as 90 degree C. This may cut the life of the CPU from 20 years to 5 years. It's a none issue for me.

    During trading day the temp is between 40-50C.



     
    #15     Apr 13, 2009
  6. It depends on the application you're running. If the application takes uses multiple threads to do the work, CPU with more cores may see a performance gain.


    I upgraded my C2D 6420 2.13, overclock to 3.12 to a i7 920, because my avg CPU utilization was near 50% constantly.

    This screen shot shows my trading day CPU utilization avg, on this C2D 3.12
    [​IMG]


    Prime 95, 25.9 v 4

    Intel(R) Core(TM)2 CPU 6420 @ 2.13GHz
    CPU speed: 3120.04 MHz, 2 cores
    CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
    L1 cache size: 32 KB
    L2 cache size: 4 MB
    L1 cache line size: 64 bytes
    L2 cache line size: 64 bytes
    TLBS: 256
    Prime95 32-bit version 25.9, RdtscTiming=1
    Best time for 768K FFT length: 13.110 ms.
    Best time for 896K FFT length: 16.116 ms.
    Best time for 1024K FFT length: 18.276 ms.
    Best time for 1280K FFT length: 23.271 ms.
    Best time for 1536K FFT length: 27.875 ms.
    Best time for 1792K FFT length: 34.073 ms.
    Best time for 2048K FFT length: 37.967 ms.
    Best time for 2560K FFT length: 49.657 ms.
    Best time for 3072K FFT length: 59.432 ms.
    Best time for 3584K FFT length: 72.487 ms.
    Best time for 4096K FFT length: 81.387 ms.
    Best time for 5120K FFT length: 103.814 ms.
    Best time for 6144K FFT length: 122.788 ms.
    Best time for 7168K FFT length: 151.159 ms.
    Best time for 8192K FFT length: 166.858 ms.
    Timing FFTs using 2 threads.
    Best time for 768K FFT length: 7.686 ms.
    Best time for 896K FFT length: 9.243 ms.
    Best time for 1024K FFT length: 13.706 ms.
    Best time for 1280K FFT length: 13.477 ms.
    Best time for 1536K FFT length: 16.220 ms.
    Best time for 1792K FFT length: 19.471 ms.
    Best time for 2048K FFT length: 22.053 ms.
    Best time for 2560K FFT length: 28.903 ms.
    Best time for 3072K FFT length: 34.815 ms.
    Best time for 3584K FFT length: 41.898 ms.
    Best time for 4096K FFT length: 47.305 ms.
    Best time for 5120K FFT length: 59.932 ms.
    Best time for 6144K FFT length: 72.813 ms.
    Best time for 7168K FFT length: 89.278 ms.
    Best time for 8192K FFT length: 100.145 ms.
    Best time for 58 bit trial factors: 3.519 ms.
    Best time for 59 bit trial factors: 3.579 ms.
    Best time for 60 bit trial factors: 3.580 ms.
    Best time for 61 bit trial factors: 3.524 ms.
    Best time for 62 bit trial factors: 3.492 ms.
    Best time for 63 bit trial factors: 5.788 ms.
    Best time for 64 bit trial factors: 5.797 ms.
    Best time for 65 bit trial factors: 5.600 ms.
    Best time for 66 bit trial factors: 5.566 ms.
    Best time for 67 bit trial factors: 5.566 ms.












     
    #16     Apr 13, 2009
  7. gnome

    gnome

    A C2D has two cores to handle 2 threads... how many apps even use 2 cores, let alone 4 or more?
     
    #17     Apr 13, 2009
  8. Once I upgraded to the i7, 920, overclocked to 3.8. My avg CPU utilization dropped quite a bit.


    This picture shows my max CPU utilization during my trading day. Applications running here are the same as the applications in the screen shot above.
    [​IMG]




    Your mileage my vary. Overclock at your own risk.




    [Mon Mar 23 19:05:41 2009]
    Compare your results to other computers at
    http://www.mersenne.org/report_benchmarks
    Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz
    CPU speed: 3800.00 MHz, 4 hyperthreaded cores
    CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2, SSE4
    L1 cache size: 32 KB
    L2 cache size: 256 KB, L3 cache size: 8 MB
    L1 cache line size: 64 bytes
    L2 cache line size: 64 bytes
    TLBS: 64
    Prime95 32-bit version 25.9, RdtscTiming=1
    Best time for 768K FFT length: 10.189 ms.
    Best time for 896K FFT length: 12.321 ms.
    Best time for 1024K FFT length: 14.052 ms.
    Best time for 1280K FFT length: 17.585 ms.
    Best time for 1536K FFT length: 21.073 ms.
    Best time for 1792K FFT length: 25.605 ms.
    Best time for 2048K FFT length: 28.987 ms.
    Best time for 2560K FFT length: 38.324 ms.
    Best time for 3072K FFT length: 46.469 ms.
    Best time for 3584K FFT length: 56.590 ms.
    Best time for 4096K FFT length: 64.036 ms.
    Best time for 5120K FFT length: 81.902 ms.
    Best time for 6144K FFT length: 97.383 ms.
    Best time for 7168K FFT length: 120.904 ms.
    Best time for 8192K FFT length: 133.190 ms.
    Timing FFTs using 2 threads on 1 physical CPUs.
    Best time for 768K FFT length: 5.568 ms.
    Best time for 896K FFT length: 6.558 ms.
    Best time for 1024K FFT length: 7.658 ms.
    Best time for 1280K FFT length: 9.215 ms.
    Best time for 1536K FFT length: 11.143 ms.
    Best time for 1792K FFT length: 13.515 ms.
    Best time for 2048K FFT length: 15.217 ms.
    Best time for 2560K FFT length: 19.865 ms.
    Best time for 3072K FFT length: 23.988 ms.
    Best time for 3584K FFT length: 29.017 ms.
    Best time for 4096K FFT length: 32.775 ms.
    Best time for 5120K FFT length: 42.058 ms.
    Best time for 6144K FFT length: 50.276 ms.
    Best time for 7168K FFT length: 62.124 ms.
    Best time for 8192K FFT length: 68.781 ms.
    Timing FFTs using 4 threads on 2 physical CPUs.
    Best time for 768K FFT length: 4.337 ms.
    Best time for 896K FFT length: 5.331 ms.
    Best time for 1024K FFT length: 6.859 ms.
    Best time for 1280K FFT length: 6.028 ms.
    Best time for 1536K FFT length: 7.281 ms.
    Best time for 1792K FFT length: 7.475 ms.
    Best time for 2048K FFT length: 8.513 ms.
    Best time for 2560K FFT length: 10.999 ms.
    Best time for 3072K FFT length: 12.366 ms.
    Best time for 3584K FFT length: 15.883 ms.
    Best time for 4096K FFT length: 17.441 ms.
    Best time for 5120K FFT length: 22.125 ms.
    Best time for 6144K FFT length: 28.390 ms.
    Best time for 7168K FFT length: 34.055 ms.
    Best time for 8192K FFT length: 39.914 ms.
    Timing FFTs using 6 threads on 3 physical CPUs.
    Best time for 768K FFT length: 4.179 ms.
    Best time for 896K FFT length: 4.742 ms.
    Best time for 1024K FFT length: 6.465 ms.
    Best time for 1280K FFT length: 5.980 ms.
    Best time for 1536K FFT length: 7.465 ms.
    Best time for 1792K FFT length: 12.605 ms.
    Best time for 2048K FFT length: 13.716 ms.
    Best time for 2560K FFT length: 16.966 ms.
    Best time for 3072K FFT length: 16.800 ms.
    Best time for 3584K FFT length: 17.660 ms.
    Best time for 4096K FFT length: 19.836 ms.
    Best time for 5120K FFT length: 25.423 ms.
    Best time for 6144K FFT length: 33.190 ms.
    Best time for 7168K FFT length: 40.220 ms.
    Best time for 8192K FFT length: 45.401 ms.
    Timing FFTs using 8 threads on 4 physical CPUs.
    Best time for 768K FFT length: 4.075 ms.
     
    #18     Apr 13, 2009
  9. Right. At any given time, there are at least 200-300 threads running on your computer. Not all of them are doing something. If they do, more cores will allow them to run a bit more efficiently.

    I am running 4 apps just for trading. and they spawn 54 threads total.



     
    #19     Apr 13, 2009