The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Question about different KINDS of processors

    Discussion in 'Hardware Components and Aftermarket Upgrades' started by iNoob.x, Oct 23, 2009.

  1. iNoob.x

    iNoob.x Notebook Evangelist

    Reputations:
    42
    Messages:
    395
    Likes Received:
    0
    Trophy Points:
    30
    I don't know if there is an answer to this question, but I'm going to ask anyways. So just out of curiosity, how do cell phone processors and their rated speeds (553Mhz, 1Ghz snapdragon, etc) compare to say a computer processor of the same frequency? For example, 1Ghz snapdragon vs. a 1Ghz Pentium III CPU, etc?

    I ask because I'm wondering how far we've come in terms of putting powerful processors in very small things... :rolleyes:
     
    iNoob.x, Oct 23, 2009
    #1
  2. sgogeta4

    sgogeta4 Notebook Nobel Laureate

    Reputations:
    2,389
    Messages:
    10,552
    Likes Received:
    7
    Trophy Points:
    456
    It's difficult to compare since they're designed for different tasks. Even comparing speeds of two architectures of Intel processors, isn't absolute. Frequency is a poor measure of general performance because of the vastly different tasks that processors are designed for.
     
    sgogeta4, Oct 23, 2009
    #2
  3. sirmetman

    sirmetman Notebook Virtuoso

    Reputations:
    679
    Messages:
    3,291
    Likes Received:
    0
    Trophy Points:
    105
    At least Intel CPUs all use roughly the same instruction set and are all multi-thread/multi-task capable these days. Lots of cellphones use single threaded or even unitasking processors and are RISC, not CISC. Frequency alone is largely meaningless when comparing even disperate cellphone processors to eachother, let alone comparing them to full computer procs.
     
    sirmetman, Oct 23, 2009
    #3
  4. iNoob.x

    iNoob.x Notebook Evangelist

    Reputations:
    42
    Messages:
    395
    Likes Received:
    0
    Trophy Points:
    30
    Is there anything like a number crunching ability test for small processors? Maybe a floating-point-operations (FLOP) benchmark? I remember personal computer CPUs can do many GFLOPs right?
     
    iNoob.x, Oct 23, 2009
    #4
  5. LoneWolf15

    LoneWolf15 The Chairman

    Reputations:
    976
    Messages:
    1,537
    Likes Received:
    1
    Trophy Points:
    55
    Exactly.

    Instructions-per-clock cycle isn't even a good measurement in this case, unless you're comparing processors that use the same instruction set.

    Example:

    Intel/AMD desktop/laptop/netbook processors -x86 (or x86-64) instruction set (usually a form of CISC)

    Mobile phone processors (e.g., ARM, Intel XScale) -Usually a RISC-based instruction set of some sort, dissimilar to desktop processors

    Graphics processing units (i.e., nVidia, AMD, etc.) --other instruction sets, dissimilar to the previous two

    Video encode/decode processors (e.g., Sigma Designs) -another instruction set.

    All of these mean you can't benchmark CPUs across instruction sets without inventing a platform-agnostic API to do it. Even then, results wouldn't be very meaningful, because one processor might perform a certain task every day, and another might not perform the same task ever, due to differences in their intended design.
     
    LoneWolf15, Oct 23, 2009
    #5
  6. iNoob.x

    iNoob.x Notebook Evangelist

    Reputations:
    42
    Messages:
    395
    Likes Received:
    0
    Trophy Points:
    30
    lol, ok. My question is teh stupid. Next!
     
    iNoob.x, Oct 25, 2009
    #6
  7. yotano211

    yotano211 Notebook Evangelist

    Reputations:
    63
    Messages:
    370
    Likes Received:
    0
    Trophy Points:
    30
    there is no stupid question in the world, only stupid people who dont ask the questions that should be stupid.
     
    yotano211, Oct 26, 2009
    #7