Performance
The memory kits have been tested on our usual Intel X99 platform which contains an Intel i7 5820K processor and an ASUS Rampage V Extreme motherboard. The CPU clock has been overclocked to 4.2GHz while the cache is clocked to 3.5GHz which should help to show the differences between the memory kits.
The Kingston Predator DDR4-3000 memory has been tested on both XMP profiles which you can see in the tables with results.
I am always starting memory reviews from AIDA64 memory and cache benchmark which is one of my favorite benchmarks. It is quick but also gives a lot of information about the memory and cache speeds.
Our Kingston Predator memory is performing the best at both XMP profiles. The G.Skill memory is close to the DDR4-2666 Kingston profile but CL14 gives better results.
Next we will take a closer look at the HyperPi 32M benchmark which really loves fast memory. Also here we see that a higher clock and tighter timings give the Kingston memory better results and it is actually scaling pretty good with each tighter timing or higher memory clock.
Rendering benchmarks such as Cinebench R11.5 and R15 are showing similar differences. Not so big but we can see that the Kingston memory performs the best. I hope that we will add some more memory kits at DDR4-3000 or higher clocks in not-so-distant future to see how the Kingston Predator DDR4-3000 looks like against faster kits.
We also tested our memory kits with the PCMark 8 software from Futuremark which is the latest version of whole-system benchmark. Once we pass the DDR4-2666 clock, the differences are not so big but again the Kingston memory performs the best in our comparison. We should not see any difference in daily usage but if the price is similar then we know which memory to choose. The price is actually good but that I will tell you more about that later.
At the end 3DMark is quite demanding in the Fire Strike Extreme test. There is also a new Fire Strike Ultra test but I think Extreme is good enough for our needs.
In this test the biggest difference can be noticed in the physics tests and we actually see that the higher clocked Kingston is once again the best. However, the DDR4-2666 kits at tighter timings will not be much slower and I think it is an optimal clock for daily work unless we find DDR4-3000 or higher memory in a reasonable price. In this case we do as the Kingston DDR4-3000 costs not much more than most available DDR4-2666 kits including the G.Skill and Corsair memory used in our review.
Let us move forward and take a look at overclocking potential of the Kingston HyperX Predator DDR4-3000 CL15.
Overclocking
Overclocking is never guaranteed so the presented results may vary from results on other memory kits. I am not recommending overclocking if you do not know what are you doing. High voltages may damage hardware and it will not be covered by warranty.
At the beginning I wanted to check how high I could set the memory clock without changing the default voltage. I was able to reach a fully stable DDR4-3200 CL16-16-17-16 1N 1.35V which passed a couple of hours of diagnostic tests like AIDA64 and memtest86+. It is quite a good result and I bet I could make even more if not for the platform limitations.
Later I wanted to check how the memory would act at a higher voltage which I set to 1.45V. I could set it even higher but 1.45V seems reasonable for daily work. At this voltage I was able to set the DDR4-3000 14-14-15-14 1N which was also stable for all my tests.
As you see the differences between both settings are low and we can clearly see the memory controller’s or perhaps the cache bandwidth limit; even though for this comparison the cache clock has been raised to 4200-4250MHz. We can clearly see that DDR4 is a fast memory even without overclocking but other factors are limiting the final effect.
This does not change the fact that the Kingston HyperX Predator DDR4 memory has a high overclocking potential.
If you have one of these memory kits then do not hesitate to share your overclocking results on the Funky Kit forums or in the comments to this review.
