As usual during memory tests we will use our standard set of benchmarks. Popular benchmarks like AIDA64, Maxxmem Preview, HyperPi 32M or Cinebench can show us clearly the performance differences we can expect between memory kits.
Test setup base on Intel Haswell platform which is perfect to compare memory kits and memory overclocking in general.
The latest AIDA64 version provides multithreaded tests so we can check the overall performance of our memory in a multitasking environment.
The HyperX 2800 memory kit achieved the highest results in memory write and copy tests. To be honest I wasn’t expecting such a high memory copy result; as most single-sided Hynix kits ( like Team Group memory in our comparison ) are performing worse than double-sided modules. Kingston clearly improved something in their memory that let it run so fast.
MaxxMem Preview provides similar tests to AIDA64 but in this case only on one thread. A lot of software is still using single threads so this benchmark can also show us some important info about performance.
The Maxxmem results are also good. The HyperX 2800 memory was the fastest in the memory read test and also close to the top in all other tests.
HyperPi is a software based on the very popular SuperPi benchmark; which simply loves fast memory so we can easily compare different memory kits and settings.
The HyperPi 32M test clearly shows the slower speed of single-sided modules. Both memory kits based on Hynix MFR achieved a similar time, which is much lower than other memory kits in the comparison.
Cinebench is a rendering benchmark and more of a “real-world” test, which tests overall computer performance.
As you can see, Cinebench is not really affected by memory clock and all results are close to each other. The HyperX 2800 memory is performing quite well while the lower clocked 2400 HyperX memory is the fastest memory kit on the list.
Overclocking is never guaranteed so presented results may vary from results on the other memory kits. We do not recommend overclocking if you do not know what are you doing. High voltages may damage hardware and in most cases will not be covered by warranty.
For the overclocking tests, I have used a slightly different setup with a Core i5 4670K processor and ASRock Z87M OC Formula motherboard.
The maximum clock achieved on this particular testing rig was 1681MHz ( DDR3-3362 ) and relaxed timings of 13-16-16-37. This is really a great result considering that the memory has been cooled by nothing else than the Kingston Predator heatsinks. There were also no overheating issues even though memory voltage was above 1.8V.
The maximum clock that passed HyperPi 32M was 1600MHz ( DDR3-3200 ) 13-16-16-35 1.76V. It’s hard to guarantee that these settings will be 100% stable but everything that passes HyperPi 32M is always near stable settings.
These are clearly awesome results for DDR3 memory and I don’t think that overclockers will be disappointed with the Kingston HyperX Predator 2800 memory series.