Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

My Macbook has a 2.4 GHz dual core processor. Because it has two separate cores, that means it can execute two functions at once, right?

So does that mean each core is approx. 1.2GHz each (for a combined 2.4 GHz) or are they actually 2.4 GHz each, and therefore is effectively running at 4.8 GHz?

Dual (and quad) core processors confuse me.

share|improve this question
1  
It's worth pointing out as well that core frequency is not really an important factor in determining the number of operations/instructions a second a processor can make (these days). –  Callum Rogers Jun 1 '12 at 21:04
    
Do two cars going down the highway at 50 miles per hour "effectively run at 100 miles per hour"? You can't add speeds that way, it makes no sense. –  David Schwartz Jul 26 '12 at 10:46

4 Answers 4

up vote 15 down vote accepted

Each core runs at 2.4 GHz.

Though that doesn't mean that your system is twice as fast as a 2.4 GHz single-core system. Parallelism has (in principle) at most a linear speedup but in reality it's often much slower (though still faster than a single core).

share|improve this answer
3  
faster than single core except in the bizarre case of python threading. –  cobbal Jul 18 '09 at 15:03
    
Really? Python threading actually slows the program down? Or is that just an exageration? –  stalepretzel Jul 18 '09 at 15:28
    
+1 - You might want a short blurb on parallelism, but it's not essential. Maybe with a pretty graph? :) –  romandas Jul 18 '09 at 16:53
1  
@stalepretzel: Many notable python implementations have something called a Global Interlock Cache which means only one thread can modify global (ie shared state at a time). This really slows things down when trying to multithread, in many cases actually making the program slower than the single-threaded variety. –  Callum Rogers Jun 1 '12 at 21:07

To be perfectly clear, having multiple cores does NOT speed up any one program unless it was specifically designed for multi-threading.

The effect of having multiple cores is:

Each thread runs at most as fast as your core clock speed. And, parallel processing allows an application to run more than one thread at once.

This means that at most, your processing power will be double, but not on any one thread.

share|improve this answer
6  
This is mostly, but not completely, true. Modern OSes are multithreaded, so even single threaded apps can see performance gains on multi-core hardware, due to OS calls and miscelaneous overhead being capable of being parallelized. –  Wedge Sep 18 '09 at 1:49

Johannes above is correct when talking about the max frequency for each core in a multi-core procesor. The frequency rating relates to each core. So, a 2.4GHz multi-core processor typically runs each processor at the 2.4 GHz max frequency. There are two exceptions (one very common and one new):

  • A processor's frequency can be dynamically scaled based upon the required performance and power draw to save power when power is not needed or to lower temperature (P-State).
  • Also, recent processors have a "Turbo" feature where one core runs faster than rated when other cores are not being used (Turbo Boost).
share|improve this answer

Your question is about Core 2 CPUs, not Core i7. Note, though, that if you buy a Core i7 920, you will see that the CPU is a quad core processor running at 2.66 GHz. That means that each core runs at 2.66 GHz. However, due to the Core i7's Turbo Boost, the CPU may actually run at 2.8 GHz, particularly if you are running single-threaded applications.

As others have pointed out, you can't multiply the CPU's speed with the number of cores to determine how 'fast' it is, I'm just pointing out that things are actually even more confusing than that with the current generation of Intel CPUs.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.