Best way to determine CPU speed?

Discussion in 'Mac Pro' started by Demigod Mac, May 3, 2013.

  1. Demigod Mac macrumors 6502a

    Apr 25, 2008
    Back in the day it was easy: clock speed was king.

    Nowadays we've got things to consider like family/generation, core count, etc. CPU model numbers and prices seem all over the place, too.

    And of course there's also the difference in speed in multicore/threaded apps vs limited or single threaded apps... a 3.33 GHz 6 core would be faster in those but slower in efficiently programmed multithreaded apps than a 12 core at 2.8 GHz.

    So, how do you determine what's a good CPU speed and price for your needs? Are the days of looking at a simple number gone and it requires a load of research per CPU?
  2. simsaladimbamba

    Nov 28, 2010
    You research the applications you gonna be using and then find out, if those are better threaded and can only take advantage of the faster clockspeed.
    As you probably use applications that take advantage of either, you have to compromise.

    I for instance chose an i7 which I can overclock and have applications, that are not yet threaded that good, take advantage of that, while threaded applications can take advantage of the many threads available.
  3. Macsonic macrumors 65816


    Sep 6, 2009
    Not sure of what's the best way to check cpu clock speed I just based my choices for cpu speed from past experiences in relation to my work. Pricing also plays a role in my choices. For light to medium general tasks there is not much difference or not noticeable. Ram frequency speed and GPU may also play a role. Some users would check out Geekbench scores as reference. For heavy design tasks, the 2.4ghz 8 core was faster than the higher clock speed 2008 2.8ghz 8 core Mac Pro. This may be due to the difference architecture used, ram specs.

    I think mac users should look at both the specs on paper, their hands on experiences in the past and their current needs and pricing. Like the 2.4ghz and the 2.66ghz, the clock speed difference is .26 mhz but pricing for 2.66ghz is way high. And is there a noticeable difference in real time?
  4. Tesselator macrumors 601


    Jan 9, 2008
    Yup, as CPU features diversify so must our testing of them.

    There are suites which perform many kinds of tests in a scripted fashion - I suppose those are the most interesting artificial benchmark applications. Of course real-world is just as Mr. Sim Salad presents it here - just try it and time it. :)

Share This Page