Performance
Performance has been tested on the Intel Comet Lake-S platform, which contains the i9-10900K, 10-core processor, ASRock Z490 PG Velocita motherboard, XFX 5600XT THICC III Ultra graphics card, and Patriot VPN100 500GB SSD. Used OS is Windows 10 x64 and the latest updates.
Our comparison includes overclocking results and settings at which the Viper Steel 64GB DDR4-3600 was stable. The maximum settings on our motherboard were DDR4-4100 CL18-24-24 at still quite low, 1.45V. The memory could also run at DDR4-3600 CL16-19-19 1.42V while CL17-19-19 offers about the same results at a lower, 1.35V.
As usual, we will start with AIDA64 Cache and Memory benchmark, which is probably the best software for synthetic memory speed tests.
Results at the XMP profile are quite good, and we can compare them to lower capacity memory kits at similar primary timings. If we used AMD Ryzen, then we could say that the results are optimal, and nothing more will significantly improve the performance. However, this is new Intel, and we can see performance scaling above DDR4-3600, and at least synthetic benchmarks are showing better results at overclocked DDR4-4000+.
PCMark 10 is showing that DDR4-3600 is an optimal clock for memory nowadays. The difference is maybe not really high, but the only significant improvement is between DDR4-3200 and DDR4-3600. Everything above the DDR4-3600 is barely visible in this benchmark.
In Cinebench series benchmarks, there is almost no difference between various settings. We can still say that the XMP profile at DDR4-3600 offers high performance.
It’s time on some 3D benchmarks from UL(previously Futuremark).
3DMark and VRMark series benchmarks are showing similar results at all settings. These benchmarks are showing higher differences in CPU and physics tests, but it’s a low percentage of the total score.
We are used to seeing higher performance gain at higher memory clock, but it just proves that all our settings provide optimal performance.
More demanding 3D tests at the display resolution up to 8K are not much different. In Final Fantasy XV and Superposition benchmarks, results are slightly better at higher memory frequency but nothing that would profoundly affect our gaming experience. At DDR4-3200 and higher memory clock, the performance gain is not as high as expected, and it’s clearly better to save some money on a faster SSD or graphics card than spend it on the fastest memory kit. Lucky for us, DDR4-3600 memory kits are already so popular that their price became affordable.
Results in new games like Assassin’s Creed and Metro Exodus at 1080p are showing up to 3FPS difference between settings where DDR4-3600 again provides optimal settings. We can still overclock the memory for an additional 1FPS, but I guess it’s better to spend this time playing the favorite game rather than run stability tests.
The Viper Steel 2x32GB DDR4-3600 memory kit performs well at all settings, but optimal in most tests seems DDR4-3600, so either XMP profile or its slightly tweaked timings. Considering how new AMD and Intel platforms are acting, the DDR4-3600 will be the most recommended memory option, which we can still get at a reasonable price. It also performs well and doesn’t require a high series and expensive motherboard.
