Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Processor speed

  1. May 11, 2008 #1
    How come processors run faster under cold conditions and slower under hot conditions?
    I always assumed resitance would play an important part in this and if a processor is made up of some type of semiconductor then resistance would decrease with temperature thus allowing more current to flow and therefore the processor would be faster. But this doesn't seem to be the case.
     
  2. jcsd
  3. May 11, 2008 #2

    russ_watters

    User Avatar

    Staff: Mentor

    They don't, where did you hear that?

    Processors can be run faster if they are kept colder because they are able to dissipate heat better. But they don't just start running faster when cooled, you need to adjust the settings of the motherboard to make them run faster.
     
  4. May 11, 2008 #3
    Oh ok, a teacher in my school.

    That makes sense, thanks very much.
     
  5. May 11, 2008 #4
    Russ is dead on. If a cooling solution for even a stock-speed processor is inadequate, heat can build up which causes an increase in internal resistance and therefore, errors to potentially form.

    When people start overclocking processors, an increase in voltage is required to maintain stability. That increase in supplied voltage causes an increase in temperature... which then requires a further increase in cooling. As you overclock more and more, it builds into a cumulative effect... where say, five times the cooling is required for just a 30 to 35% overclock. That's when people start playing with solutions like refrigeration and liquid nitrogen.
     
  6. May 11, 2008 #5
    Russ is dead on. If a cooling solution for even a stock-speed processor is inadequate, heat can build up which causes an increase in internal resistance and therefore, errors to potentially form.

    When people start overclocking processors, an increase in voltage is required to maintain stability. That increase in supplied voltage causes an increase in temperature... which then requires a further increase in cooling. As you overclock more and more, it builds into a cumulative effect... where say, five times the cooling is required for just a 30 to 35% overclock. Push it even more, and that's when people start playing with solutions like refrigeration and liquid nitrogen.
     
  7. May 11, 2008 #6
    B. Elliott is right.

    Processors will not "become faster" by lowering the temperature, but depending on the processor you have, you COULD modify the parameters of the CPU (front-side bus, multiplier, voltage...) so it performs faster than it should (if anyone remembers the old celeron FCPGA processors, Intel used to just sell underclocked CPUs).

    The side-effect is the extra heat, which is why the more you overclock the processor, the better the cooling solution must be in order to provide stability.

    Cooling down a stock CPU won't do anything though. Heating it up however does do something, and that something is pretty much 'instability.' Some CPUs have the ability to underclock itself to remain stable in these situations, such as Pentium 4s. However, if the CPU gets way past its threshold, then the die will eventually turn to poop.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?