Performance Memory

At all times and measurements the system baseclock and multiplier will remain the same, meaning each and every difference you spot on performance is a direct results of changed memory frequency and timings.

Memory Read test

Okay the first round of tests, synthetic testing. Above you can see read memory performance of the quad-channel memory configurations at 1333/1866/2133 MHz on the X79 platform with a Core i7 3960X processor.

Now you need to continuously compare memory frequency performance as that is what we'll be looking at today. For all synthetic tests I also included dual-channel performance on a Z77 system. In the real world tests we'll swap back to X79 though. This is just to demonstrate dual-channel performance on Z77 compared to quad channel on X79.

Intel's entire Sandy bridge architecture and specifically here the memory controller harbored inside these processors just hauls ass, with a quad-channel memory setup with 2133MHz read perf closing in at 20GB/sec.

Memory Write test

Write performance then -- To give you an idea, an AMD Phenom II series processor would perform roughly in-between 7000 MB/sec -- 8000 MB/sec here with really nice timings.

Corsair in dual-channel mode reaches a astonishing 20000 MB/sec at 2133 on Z77.

But going back to quad-channel on the X79 we notice the write perf drop back to 15GB sec. And that's the trivial aspect of today's tested memory on an Intel series 7 chipset/ Ivy Bridge processor platform against quad-channel Z79. This primarily shows with AIDA as a test though. In SiSoft Sandra it's actually vice versa ...

Sandra Memory Index

Sandra is again a synthetic memory test, here we make use of the Sandra Memory performance test, it gives us an index score of overall memory performance. Here again the numbers show amazing performance differences. As you can see quad-channel literally doubles up memory bandwidth. 52 GB/sec for the Corsair DIMMs, now that doesn't suck.