In the past I ran the full battery of benchmarks when testing memory. I decided that all tests excepting memory specific tests are pretty pointless, because memory is only a small part of the entire system and changing the clock speed or timings don’t make a huge amount of difference by themselves. Also, raising memory speed by overclocking the CPU really skews the results, making it appear that changing the memory speed made more difference that it actually did. And the real danger is the results being evaluated by someone that doesn’t understand these things…i.e., thinking a memory kit sucks because overclocking it 40% only gave a .15 FPS increase in his favorite game.
I ran the tests at the advertised settings of DDR3-1600 7-7-7-24, and again at the max clock at CAS 7, DDR3-1705 7-7-7-24. Finally I ran the tests at my max overclock of DDR3-1948 9-8-8-24. I also ran the tests at CAS 6, and there was a performance gain, but not significant enough to add to the charts.