Question G.SKILL Trident Z Neo DDR4 3600 C14 2x16GB underperforming

Anton Hunter

Distinguished
Apr 24, 2013
57
0
18,530
My DRAM (G.Skill : F4-3600C14D-32GTZNA) in DOCP : 3600 @ 14-14-14-14-34-85-630 & 1T, are operating at 108% performance level according to Userbenchmark; but, the average rating is 125%. I have them slotted in A2B2 on an ASUS TUF gaming B550-Plus Motherboard... they boot fine; but, I'm wondering if there isn't something I'm missing that would raise the performance level if I knew to implemented it. As it is, I haven't found any info from ASUS, G.SKILL, Reddit, or here (yet) to satify my curiosity... so, any thoughts on the matter would be welcome. Thanks.

My Rig: ASUS TUF gaming B550-Plus mb, Ryzen 7 5800X, G.SKILL Trident Z Neo 2x16GB DDR4 3600 CL14 DRAM, Sapphire Pulse RX 6800, Corsair HX1050 PSU, Samsung 970 EVO 500GB NVME system SSD, and Teamgroup 2TB NVME SSD for storage, (Custom water-loop on CPU... underload temps range from 55C to 62C with occasional spikes to upper 60s C); (GPU temps rarely reach the 70C mark), I have 4 Noctua industrial 140mm fans for intake and 2 in exhaust... keeps the NZXT case nice and cool while running at 1500-ish rpm.
 
Userbenchmark is virtually worthless and doesn't actually provide meaningful information. The only way you're going to be able to tell differences is if you have context with regards to another setup which is actually identical to yours.

If you're looking to hyper optimize your system check out what Buildzoid does (Actually Hardcore Overclocking is his YT channel) and read through information/guides like this: https://github.com/integralfx/MemTestHelper/blob/oc-guide/DDR4 OC Guide.md

I wish I had some of Buildzoid's older cheat sheets regarding DDR4 timings, but it may be possible to go back and find those.
 
  • Like
Reactions: Anton Hunter

Zerk2012

Titan
Ambassador
My DRAM (G.Skill : F4-3600C14D-32GTZNA) in DOCP : 3600 @ 14-14-14-14-34-85-630 & 1T, are operating at 108% performance level according to Userbenchmark; but, the average rating is 125%. I have them slotted in A2B2 on an ASUS TUF gaming B550-Plus Motherboard... they boot fine; but, I'm wondering if there isn't something I'm missing that would raise the performance level if I knew to implemented it. As it is, I haven't found any info from ASUS, G.SKILL, Reddit, or here (yet) to satify my curiosity... so, any thoughts on the matter would be welcome. Thanks.

My Rig: ASUS TUF gaming B550-Plus mb, Ryzen 7 5800X, G.SKILL Trident Z Neo 2x16GB DDR4 3600 CL14 DRAM, Sapphire Pulse RX 6800, Corsair HX1050 PSU, Samsung 970 EVO 500GB NVME system SSD, and Teamgroup 2TB NVME SSD for storage, (Custom water-loop on CPU... underload temps range from 55C to 62C with occasional spikes to upper 60s C); (GPU temps rarely reach the 70C mark), I have 4 Noctua industrial 140mm fans for intake and 2 in exhaust... keeps the NZXT case nice and cool while running at 1500-ish rpm.
Junk benchmark site just use your PC and enjoy!!!!!!!!!!
 
  • Like
Reactions: Anton Hunter

Anton Hunter

Distinguished
Apr 24, 2013
57
0
18,530
Userbenchmark is virtually worthless and doesn't actually provide meaningful information. The only way you're going to be able to tell differences is if you have context with regards to another setup which is actually identical to yours.

If you're looking to hyper optimize your system check out what Buildzoid does (Actually Hardcore Overclocking is his YT channel) and read through information/guides like this: https://github.com/integralfx/MemTestHelper/blob/oc-guide/DDR4 OC Guide.md

I wish I had some of Buildzoid's older cheat sheets regarding DDR4 timings, but it may be possible to go back and find those.
Thanks for the Buildzoid reference. I'm familiar with him, so will search it out. I'm in agreement about Userbench; but a 17% performance drop got my attention none-the-less. LOL As far as the DRAM timings go, I haven't been able to find anything better than the settings I have now. Buildzoid might have an answer for that too. I'm checking out the Mem-test Helper, too... its impressive... whether it helps, or not, I'm certainly going to research it. Thanks for the reference.
 
Last edited:

Anton Hunter

Distinguished
Apr 24, 2013
57
0
18,530
I was just thinking, having the link to a ubm run where system has been rebooted and let sit a few with ALL Unnecessary programs shut down would help everyone alot.
Thanks for the input. I've removed all the "Bloatware" I could from Windows10 Pro, which dropped my system ram idle load from 8GB to just under 4GB. There are still quite a few background apps running, mostly for my Security software, gaming apps, and diagnostics. It HAS made a difference, too. I'm even considering going the Tiny10 OS route; but, just haven't pulled-the-trigger on that move , yet.
 

Misgar

Respectable
Mar 2, 2023
1,610
420
2,090
Don't install TinyOS without studying what it throws out. You might find it's a step too far.

RAM is relatively inexpensive and you've already got 32GB. What difference did reducing base RAM usage from 8GB to 4GB make?

If 4GB extra free RAM makes a big difference to your programs in a 32GB system, perhaps you should upgrade to 64GB.

I disable all the Microsoft Windows background apps and leave it at that.
 
Yeah I wouldn't trust user benchmark at all. And Debloated OS's I always found to be useless, I've ran them, and even have a system that currently running Tiny OS, i2 2100 system, and it makes like no difference if you have enough memory. In fact It makes things worse as something you may need wont work anymore, or updates wont install.

Good Luck!