Bests & Worsts Reviews from Amazon

Some buyers have expressed concern after buying this RAM and installing it in their iMacs (27" Ivy Bridge model, late 2012). That iMac is specced for DDR3 1600 RAM, and ships with RAM like that, but after you install this Corsair RAM your iMac will show that it is running aRAM at 1333 MHz, not 1600. So: (a) The reason this is happening is that DRAM is (to simplify things tremendously) described by two parameters. One is bandwidth (essentially how much data it can transfer per second), the other is latency (how long does it take to set things up before that data starts being transferred). (b) Your APPLE RAM in the iMac has a bandwidth of 1600MHz and a latency (CAS) of 11 cycles. The Corsair RAM has a bandwidth of 1600MHz and a latency (CAS) of 10 cycles. (c) Given that the two sets of RAM do not have identical specs, how should the memory controller configure things for optimal performance? The timings have to be identical for both sets of RAM, meaning they have to run with identical bandwidth and identical CAS. (d) The CORRECT solution in this case would be to run both at 1600MHz and CAS of 11. However the solution chosen by the memory controller (I am guessing this is handled automatically by Ivy Bridge, not by Apple firmware, but I honestly have no idea) is to retain a CAS of 10, and drop the bandwidth to 1333, which is slow enough that the Apple RAM can run at a CAS of 10 rather than 11. This is not a completely crazy solution. For most purposes, latency is more relevant to performance than bandwidth, so if you are faced with having to choose one over the other, you should choose to drop bandwidth while preserving latency. It's just unfortunate than in this case the algorithm (which, remember, has to work for the possibility of people plugging in four RAM DIMMs, possibly all from different manufacturers, with different bandwidths and latencies) chooses a solution that is slightly sub-optimal compared to the best possible solution. (e) Does it actually matter? Short answer: no. Neither Apple nor Corsair have lied to you about their RAM, and as a practical matter the difference in performance between CAS 10 and CAS 11 ram, or between running at 1600 vs 1333 is negligible (like 1%). Yes, you can undoubtedly craft some benchmark which is sensitive to the issue and which magnifies the problem, but for any real app it just doesn't matter. If you are so obsessive that you cannot stand this, if you buy a second batch of Corsair 16GB RAM and replace the Apple RAM, you will get back to 1600MHz, now at CAS 10 rather than CAS 11. But you honestly won't notice any performance difference. Bottom lines: - The extra RAM is nice --- it means paging on my iMac moves from being a rare occurrence to a non-existent occurrence, and allows that much more file system data to be stored in RAM; - iMac will show the speed of your combined system (if you bought only two sticks, not four) as 1333 MHz, not 1600; - There is no reason to be upset or worried about this. It does not mean anyone cheated you or that your system is running slower than it otherwise would.