Slashdot Mirror

← Back to Stories (view on slashdot.org)

Examining Benchmarking

Posted by michael on from the no-benches-involved dept.

VL writes "Benchmarks exist to determine how a particular piece of hardware performs in relation to itself, and to others. Question is, are readers getting the information they really need?"

2 of 95 comments (clear)

  1. Re:Goedel says benchmarks are inherently flawed. by Jeremy+Erwin · · Score: 3, Informative

    The application of Goedel's Incompleteness theorem to benchmarks borders on 100% organic bullshit. On the other hand, the statement

    Software is an external reference point, its somthing outside of the system. is itself iffy.

    We know that the video cards are designed, in part, to benchmark well. Some manufacturers have even gone so far as to write drivers that inflate framerate at the expense of accuracy, under certain benchmark like conditions. (Quake.exe v. Quack.exe, anyone?). Apple inflated its spec results by using a unrealistic single threaded malloc library. Intel's icc is rumoured to detect, and optimize for SPEC.

    The Dynamic application of intellect is what defines real intellegence..not theorys..thats just memorization. :-D

    Theories? Theories are meant to be proven as an exercise for the student, not just memorized.

  2. Re:Not all of us by JamesP · · Score: 3, Informative

    These are NOT redered using Video Card power, but the processor...

    Video card power is only used in DESIGNING and PREVIEWING the scenes...

    Video card is good but not as good as raw math over some minutes (or even hours)

    --
    how long until /. fixes commenting on Chrome?