Performance
Performance has been tested on the ASUS TUF A15 laptop equipped with AMD Ryzen 7 4800H, 8-core processor, and GTX1660Ti graphics card. All tests were performed in Windows 10 x64 environment with the latest updates.
To make our comparison more interesting, I’ve added results with an edited SPD profile what I will describe in the overclocking section of this review. As an addition, there is also a Corsair memory kit, which runs at DDR4-2666 to show the difference between more standard settings and DDR4-3200, which is optimal for the latest processors.
As usual, we will start with the AIDA64 Memory and Cache benchmark, which is probably the best application to check memory bandwidth and latency.
AIDA64 bandwidth results are higher than expected and simply exceptional. It looks like used memory chips make the motherboard to set tighter sub-timings. Our results are about as high as that of 2x8GB Hynix based DDR4 SODIMM memory. We have to remember that higher memory capacity almost always means more relaxed sub-timings, so everything that we can’t set on laptops.
In PCMark 10, so something closer to the “real-life” workload, results are not so much better than other memory kits, but still high.
The same situation we can see in Cinebench R20. Our results are slightly higher than on most other memory kits but nothing that would be really visible in daily work.
3DMark series benchmarks are showing similar results in all settings. We can tell which one is higher or lower, but most are closer to the error margin. Of course, it’s good for the Silicon Power memory as it shows high performance at its default settings.
In VRMark, we can see quite an improvement in the Orange test. Depends on the used memory kit, we can gain up to 5FPS. The Silicon Power memory performs well and only 1FPS lower than its tuned version. More demanding tests in the VRMark benchmark are not really reacting to memory settings.
In benchmarks based on popular games, results are again not far from each other. HyperX Impact looks the best in these tests, but Silicon Power isn’t far behind. All results look good.
Newer games at 1080p react well to memory settings and clearly like lower latency. In this test, the best results are on tuned Silicon Power memory. The default settings are 2-3FPS slower what wouldn’t be really visible considering that the average is above 80-90FPS.
Silicon Power DDR4-3200 SODIMM shows high scores in all tests and games. I’m sure that everyone will be satisfied with its performance. Considering it’s a 32GB kit, it should be perfect for both games and more demanding work.
Overclocking
Disclaimer: Overclocking is never guaranteed, so that the results may vary depending on certain conditions and various hardware configurations. I am not recommending overclocking if you do not know what you are doing. High voltages may damage hardware, and the warranty will not cover it.
SODIMM memory is quite limited if we talk about overclocking. Not many laptops have an option to adjust memory timings or frequency, so all who want to set something more have to use additional tools. One of the only ways to improve SODIMM memory performance is to edit the SPD profile. For that, we can use Thaiphoon Burner software, which will work as long as the memory doesn’t have write protection on the SPD. Lucky for us, Silicon Power doesn’t have write protection, so we could play some more to check how low we can set memory timings.
The SPD table isn’t really long, but lowering main timings from CL22-22-22-52 to CL18-18-18-38 and, additionally, lower tRC and tRFC gave us quite visible results, which were added to the comparison on the previous page of this review. I’m sure we could go even lower with timings, but we were limited to 1.20V VDIMM. There were no problems at modified settings, but we can’t count that every memory module will run the same.
It’s of course, not recommended to edit SPD profiles, and for most users, it’s not required. The Silicon Power DDR4-3200 memory runs great even without further tuning.
