-2

My system processor is Intel Pentium(R) Dual-Core CPU E5300 @ 2.60GHz.

My processor speed is 2.6 GHz as mentioned in process name. The processor name tells this is dual core.

My doubt is:

Processor speed is 2.6 GHz and CPU cores 2.

What is the meaning of 2 cores. I think 2 cores means (2.6 GHz + 2.6 GHz = 5.2 GHz).

Is it right?

2

3 Answers 3

3

A core is basically a CPU, just physically a multicore cpu is in one package. Unless you are in the habit of touching them you can thing of a core as a cpu.

You may be told a lot of hog wash about cores, such as with multiple cores you can run multiple tasks. Well you can run multiple tasks on one core/cpu. Multiple cores mean that your operating system has to split up the work, this is not always possible. Therefore 2.6GHz + 2.6GHz < 5.2GHz, and 2.6GHz + 2.6GHz ≥ 2.6GHz. Yes multi core can sometimes be only as good as one core. It all depends on OS, and usage.

Readers note: The + sign above is not normal addition, it is core addition. That is why 2.6GHz + 2.6GHz = 2.6GHz is possible.


Edit: see Paul A. Clayton's comment, it points out that multicore can at times be better than single, but only when all cores can be utilised (multi-thread/multi-process).

3
  • 2
    Real world performance is much more complex than clock frequency. A higher performance single-threaded core can be limited by memory latency (ignoring diminishing performance returns for power and chip area budgets). Hardware multithreading increases the amount of memory-level parallelism, allowing more memory latency to be hidden (it also helps with branch resolution and execution latency). Multicore provides similar memory latency hiding benefits without the L1 cache contention. These techniques can turn a memory latency bottleneck into a memory bandwidth bottleneck.
    – user180742
    Jul 27, 2014 at 19:03
  • I am sorry if the above comment was a bit harsh. The original poster was asking a simple question and not looking for a treastise on the tradeoffs in processor design. Your use of "hog wash" triggered a reflex that reacts to the belittling of the complex tradeoffs in processor design. Yes, some marketing has replaced the megahertz myth with the core-count myth, but both are at best coarse measures of performance which as you noted "all depends" (workload, non-core CPU and non-CPU system components).
    – user180742
    Jul 27, 2014 at 19:04
  • @PaulA.Clayton thanks for pointing out some benefits of multi-core, what you say is true (if the work load allows, more that one thread). Thanks for also pointing out the Hz myth. Jul 27, 2014 at 19:17
2

"core" is hard stuff, so: two CPU's that can be handed separate tasks.

You may also have e.g. hyperthreading which then makes one/each CPU appear as more than one LOGICALLY.

Example: http://ark.intel.com/products/75122/Intel-Core-i7-4770-Processor-8M-Cache-up-to-3_90-GHz - 4 cores, 8 threads. 8 tasks can be run, but note that it will not provide / be equal to "8x efficiency/power" - there are bottlenecks and limitations.

1
  • A tip related to this: look up GNU parallel and see if it is available for your OS (GNU/Linux; likely) - it is a tool that allows playing with this, and in certain situations it may well provide quite a boost for a task at hand - it is command line only as far as I know.
    – Hannu
    Jul 27, 2014 at 12:21
1

No, it definitely doesn't not mean you can clock it twice as fast.

And, most of the time, you will get a speed up less than 2x. (And, no you don't get that from doubling the clock rate, you keep the clock rate the same.) Still, you won't have enough work to use both cores, or worse all 4 hyperthreads if it is a core that supports hyperthreading. So, it is generally faster than a single core chip at the same clock rate, but how much depends upon your workload.

And, incredibly enough, under the right workloads and the right configuration, you can sometimes make the dual core more than 2 times faster than a single core version. However, most of this is covered by the phrase: lies, damned lies, statistics, and benchmarking. I know because I used to do benchmarks for Intel's (network processing group), so we created benchmarks for various chips under development doing various networking tasks, e.g. acting like routers or web servers, etc. Depending upon what we wanted to show, we would often pick different chips to do the demonstration. It wasn't a lie, but it was very carefully tuned. Not the kind of workload you as a home user are likely to get, but maybe appropriate for a use-case Google, AT&T, or Cisco might have in mind.

And, by the way, there are sadly cases where the other side of that equation is true, where a dual core chip is actually slower than a single core chip. Fortunately those cases are also rare.

But whichever case applies, if you are a home user, you can assume that generally, you are getting some speed up, probably not quite 2x, more like 1.2-1.8x, but some. And, more importantly, it probably gets better when the system is under load. The more you push your computer to do many things at once, the more likely you get better performance out of a dual core machine.

And the number of cores that exhibit a performance boost is also generally limited. You don't get another 2x boost by going to 4 cores (except in particularly limited cases), much less a third 2x by going to 8 or a fourth by going to 16. Yes, it is possible to contrive such cases, but they are definitely not the norm.

So, my 4 core Atom netbook is still noticeably slower than this Xeon laptop and it isn't due to core count or even clock frequency. The speed at which your computer runs is much more complicated than that. But, if I have only one website open, the netbook seems to run just as fast at this laptop because the real limiting factor is network bandwidth, not computer speed at all.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .