Let’s now start with the performance tests, starting with the synthetic benchmarks, which are all run 3 times each, with the best result being used. In the case of full LinpackXtreme, a run with 5 loops and the average result output from this is used in each case. The AIDA64 Cache and Memory Benchmark in version 6.60.5900 is well suited as a theoretical indicator for the bandwidth and latency of a configuration.
Of course, the kit already benefits with its high clock rate in the bandwidth-intensive AIDA Read, Write and Copy tests in XMP and even more overclocked at 6800 Mbps. The 100000 MB/s mark is easily cracked by the two manually set configurations, with 1T always just ahead of 2T.
Interestingly, in the latency test, the ratio of frequency and tCL benefits more than the command rate. Thus, 6800c30 2T is just below 50 ns and 6800c32 1T just above. Of course, this test is very synthetic and also not very consistent, so even small background processes in Windows can have an impact on the result. However, it is still remarkable that a DDR5 kit can almost completely close the gap to the DDR4 reference configuration for the first time.
The relatively low latency is confirmed again in the SuperPI 32M benchmark, so both overclocked settings with the ADATA kit are significantly faster than DDR4. It will probably not be long before DDR5 will no longer be a sidegrade of DDR4 in terms of latency.
In LinpackXtreme where latency and throughput are almost equally important, the Hynix M-Die based kit can once again clearly distinguish itself from the competition. Not only the clock rate and the primary timings are better here, but the kit can also run the subtimings, which are so important for bandwidth, much tighter than the competition based on Micron or Samsung DDR5.
I don’t think much needs to be said about Geekbench 3, the results speak for themselves and the advantage of manual OC with Hynix M-Die shows up enormously here as well. And even though you can get close to the 14000 point mark with Samsung’s 16 Gbit B-Die and the 1.6 V crowbar, this is not nearly as efficient and certainly not stable for daily use.
- 1 - Unboxing and design
- 2 - Dimensions and SPD information
- 3 - Teardown and PCB analysis
- 4 - Test systems and methodology
- 5 - Heatsink peformance and overclocking potential
- 6 - Synthetics – AIDA64, Geekbench 3, SPI32M, LinpackXtreme
- 7 - Gaming – SoTR, CS: GO, Cyberpunk 2077
- 8 - RGB lighting and conclusion
42 Antworten
Kommentar
Lade neue Kommentare
Urgestein
1
Mitglied
Urgestein
Urgestein
Veteran
Veteran
Urgestein
1
Urgestein
Urgestein
Urgestein
Mitglied
Mitglied
Mitglied
Mitglied
Mitglied
Mitglied
Mitglied
Alle Kommentare lesen unter igor´sLAB Community →