Performance
Performance has been tested on the Intel Rocket Lake-S platform, which contains a Core i7-11700K processor, MSI Z590I Unify motherboard, ASRock RX6800 XT graphics card, Silicon Power US70 1TB NVMe SSD, and Abkoncore 850W 80+ Gold PSU.
All results have been made using Silicon Power X-Power Zenith RGB 32GB DDR4-3600 memory kit. All overclocking results were stable and passed multiple tests, as listed in the comparison. At the same time, the tested memory kit could reach DDR4-4400 CL18 on the AMD Ryzen platform, which could be limited by the memory controller, and that’s why our tests are presented on the Intel platform.
As usual, we will start with the AIDA64 Memory and Cache benchmark, which is probably the best application to check memory bandwidth and latency.
It’s not hard to notice that the bandwidth results are amazing. For sure, the dual-rank design of our memory kit helped. I’m also surprised that the dual-rank memory kit could overclock so high, and it’s not any cherry-picked memory kit. This is exactly what you can find in popular computer hardware stores. After overclocking, we could pass 70GB/s bandwidth in the AIDA64 memory read test. The more impressive is 68GB/s in the memory copy test. Memory copy is close to the internal Windows memory performance test, and it affects daily work the most.
Even though at a higher memory clock, the memory controller runs at Gear 2 mode, then the latency is not losing much. We could register +/- 5ns latency between various settings. The XMP setting was using Gear 1 mode, and it looks the best in this comparison.
In PCMark 10, all results are similar. We can tell that the DDR4-4800 is a bit faster, but the XMP is not far behind.
In rendering benchmarks like Cinebench R23, results at all settings are also similar. Again, we can tell that results are getting slightly better with memory frequency, but it’s nothing that would affect our daily work.
3DMark series benchmarks are also showing a similar level of performance at all settings. There was a little drop at DDR4-4200 in the Fire Strike Extreme test, but except that, all settings are nearly equal.
The least demanding test in VRMark, so the Orange Room gives us up to 7FPS difference between settings. All results at other settings are within’ error margin.
Superposition and Final Fantasy XV benchmarks are one more time showing barely any performance gains from overclocking. We could see a higher difference while using single rank memory kits in previous reviews.
As usual, modern games are showing the highest differences. Depends on the game, our results vary by about 6FPS. What is interesting, The best results are on both the XMP profile and DDR4-4600 CL18. It looks like Gear 1 mode is helping a lot in some games. Without tests, it wouldn’t be so clear. It also proves that not everything that is the most expensive is actually better. Silicon Power RAM is known to deliver good results at a reasonable price. We can count on that also in the case of the Zenith memory series.
On the next page, I will tell you some more about overclocking results.
