Analysis of results
As for the two games examined, in both games and in all resolutions no significant differences were found. Minor fluctuations in favor of one memory or another are not significant in a way that sets the scales to one side or the other. You can clearly see that games of this type are affected exclusively by the power of the processor and graphics card and that the memories are not a limitation.
In WinRAR, on the other hand, we have definitely seen better results with increasing the frequency of memories. The software introduced an improvement of nearly 9 percent at a speed of 1820MHz relative to the result achieved at 1600MHz speed. This should be attributed, of course, to the fact that such software reads and writes large amounts of data to memory and therefore gains almost directly from increasing bandwidth to memory derived from its working speed. The software also improves its performance by downloading the memory timings from 9-9-9-24 to 7-7-7-21 by approximately 5 percent. The improvement is explained by the fact that downloading the timings also contributes to increasing the bandwidth to memory.
The SuperPI results show once again that this software relies almost exclusively on processor power and most processing based on its cache memory, so memory speed has no effect on the results of this test. The small differences are due to the slightly different processor speed between test modes.
Sandra's bandwidth testing also shows that the two memories are very similar in performance at 1600MHz, a logical result if we remember that they are based on the same memory chips. Increasing the frequency causes an almost linear increase in memory bandwidth.
Memories were also tested at their specified intervals to ensure that they did meet their characteristics and to understand the differences between timings released relative to more tight times. The Super Talent ProjectX memories were tested at an 900MHz frequency and in 7-7-7-21 timings. While the Team Xtreem memories were tested at the 800MHz frequency and in the 7-7-7-21 timings.
It can be seen that in tests that use memory bandwidth such as Sandra, PCMark and WinRAR, there is an improvement with the timeliness of the same frequency with free timings. On the other hand, all the other programs that rely mainly on the processing capabilities of the graphics card or processor do not differ in performance.
You can also see that The fast Significant memory - as in the case of theTeam Group Xtreem, which has increased by about 10 percent - can also be an advantage in the price of raising the schedules. Memory bandwidth increased by about 5 percent and latency was shortened slightly. In practice, it is possible to increase quickly even without raising all the timings to their maximum value, with the best course of action being to find the balance point between raising the frequency and raising the timings, where the best performance is obtained.