Very basic question about processor speed

[Typswif2fingers]

Hello,

After the introduction of the new MBA I started feeling a tingling sensation in my banking parts

I have a mid-2009 15" MBP 2.66Ghz C2D 4GB 1067 MHz DDR with a 320GB HD.

My notebook needs are fairly basic and go along the lines of Web browsing, opening word and pdf documents, and playing Rome Total War (ah to be an emperor! )

I also have an late 2010 27" iMac i5 which I use for word processing, downloading files of internet... nothing really more.

Now, I was thinking of purchasing the new MBA 13" 1.8GHz i7, but cannot get my head around the processor speeds.

If someone could help in explaining how the 1.8GHz i7 on MBA compares with my 2.66GHz C2D in terms of what I do on notebooks.

I apologise if this is a silly question. But, to me a meter per second is a meter per second. Hence a cycle per second is a cycle per second and it stands to reason that 2.66GHz processor is faster than 1.8GHz processor.

Having said that, I think that I am incorrect in that view taking into account the flash memory, newer processors, more cores etc. I would really appreciate if someone could write a quick one to provide some light on the issue.

Thanks.

Tw2f

 

[alust2013]

For a few reasons, even the 1.8GHz Air will handily smoke your 2.66GHz Pro. The newer processor architecture is more efficient, it has turbo boost, and it has 2 physical cores and 2 virtual cores. Either way, with light use, you likely wouldn't notice a huge difference other than the faster loading times with the SSD. That said, your MBP will likely be better for gaming due to its beefier graphics card.

 

[paolo-]

http://www.youtube.com/watch?v=I3WnXaWjQYE

I know the video is quite dated, and I'm not sure if pipeline stages are still a relevant factor in modern CPUs. But, @2min31, you get the bulk of the idea. Clock rate is not the only factor. Depending on the number of transistors in the CPU, it might be able to do many operations per cycle. Say if it's doing a big addition, an older CPU might take 4 cycles to do and a new one might only take one. The cache is the memory inside the CPU, normally, the CPU gets instructions from the RAM, but the ram is quite slower than the CPU, so if the CPU can store a lot of instructions in advance, it won't have to wait for the RAM as much.

In modern CPU, we might also want to compare the number of cores, in your case, all the computers you are comparing have two. There is also virtual cores, basically, they add the possibility to run two threads (or commands) at once in the CPU.

This is a dumbed down version of how I understand the CPU, I'm by no means an expert in that area. Bottom line, benchmarks tell you much more about how well a CPU performs against another. Basically, how much time it takes to perform a real world task, or just straight up operations per second.

 

[Typswif2fingers]

Thank you very much guys.






I couldn't resist