Included in AMD’s AM5 review kit, G.Skill’s F5-6000J3038F16GX2-TZ5N DDR5-6000 C30 is the first AMD EXPO-compatible kit we’ve tested. That just means that it has AMD’s version of Intel XMP technology, which is nothing more than extended configuration data within the little flash IC that each module contains for automatic configuration. Like XMP, AMD EXPO instructs the motherboard on which overclock settings will be required to got an overclocked memory kit to run at its overclocked rating. In this case, the memory chips are very nicely overclockable DDR5-4800, as seen in our Sabrent DDR5-4800 Review.
Rated at DDR5-6000 C30, those primary timings extend to 30-38-38-96. This, as opposed to the 40-40-40-76 timings of our reviewed Patriot DDR5-6200 C40.
We were told to expect some memory to ship with both so-called technologies, yet G.Skill’s EXPO kit does not include an Intel XMP profile, and Patriot’s XMP kit was manufactured before AMD EXPO existed. We wanted our first dive into AMD’s AM5 to be completely accurate, and were happy to find it completely stable after manually configuring its EXPO settings in the XMP-only firmware of our Intel platform.
We later retested the Intel platform with the XMP kit and found that the performance numbers were similar, but never revealed how similar as we were hoping to do a deep dive at a later date. Thanks to a little prompt by a naysayer, today is that date!
|CPU||Intel Core i9-12900KF: 16C/24T, 3.2-5.2 GHz, 30 MB L3 Cache, LGA 1700|
|CPU Cooler||Alphacool Eisblock XPX CPU, Eisbecher D5 150mm, NexXxoS UT60 X-Flow 240mm|
|Motherboards||ASRock Z690 Taichi: LGA 1700, BIOS 13.05|
|Graphics Card||Powercolor Red Devil Radeon 6750 XT: 2324-2623MHz GPU, 12GB GDDR6|
|DRAM||Patriot PVVR532G620C40K 2x16GB (32GB) DDR5-6200 CL40-40-40-76 1.35V
G.Skill F5-6000J3038F16GX2-TZ5N 2x16GB (32GB) DDR5-6000 CL30-38-38-96 1.35V
|Power||be quiet! Dark Power Pro 10 850W: ATX12V v2.3, EPS12V, 80 PLUS Platinum|
|Hard Drive||Sabrent Rocket 4 Plus 2TB PCIe 4.0 M.2 SSD|
|Sound||Integrated HD Audio|
|Graphics Driver||AMD Adrenalin Edition 2022.10.1|
Sandra and AIDA64 both show the DDR5-6200’s superior bandwidth, but AIDA64 also thinks that it has lower latency…but how? Simple math regarding cycle time would show that the numbers are roughly equal apart from the first (c30 vs c40) and the last (96 vs 76), and we can’t imagine the faster memory’s lower tRAS making up the difference between c30 and c40. Perhaps apps will be more fruitful?
The 12900KF CPU performs better across every bench of Time Spy when using the DDR5-6200 c40 and doesn’t really start to fall back until we get to 3DMark’s storage benchmark. And that slight loss is probably due to the drive aging.
The DDR5-6200 c40 leads in PCMark Essentials but falls back slightly as the tests progress into productivity and content creation.
DDR5-6200 C40 wins more games than it loses in F1 2021, but the difference is too slight to matter.
The DDR5-6200 C40 continues to dominate as we move into application benchmarks, even allowing timed test to finish more quickly when compared to DDR5-6000 C30.
And now for the overall score: The Creativity and Productivity suites of PCMark really kicked the DDR5-6200 down, but timed apps and games helped to lift it back up.
Overall scoring differences are down to ZERO POINT THREE PERCENT, which makes us extremely comfortable with our decision to use the XMP kit in future Intel platform tests and the EXPO kit in future AMD platform tests. Thanks to the C30 kit’s loose tRCD, tRP and tRAS these are, for most intents and purposes, equivalent.