Question ~621 ST R20 score 5800X?

[BTRY B 529th FA BN]

I'm assuming my 5800X should be scoring a little higher than 621 in R20 ST.

The things I assume effect the performance:
BIOS
-AMD Cool&Quiet enabled - increases score VS disabled
-CPPC - enabled helps steer thread to fastest core in the system
-CPPC Preferred cores - enabled helps steer thread to fastest core in the system

OS
-Processor idle disable - disabled - lowers score
-Processor Performance autonomous mode - enabled helps steer thread to fastest core in the system


I don't use any CPU tuning, pbo, etc. etc. with my chip. I just have my RAM tuned and 1:1 1866/3733 DR (dual rank) 1900 and I get WHEA errors


I was thinking I should be closer to 630. Am I doing something wrong?

 

[Shmee]

What is your cooling and ambient temps? This could have an effect on performance.

 

[BTRY B 529th FA BN]

Ambient 74*F
Noctua NH-D15S Chromax. Black
4x140mm Cougar fans

When I disable all those things that help, and run Ultimate Performance with Processor Idle disable disabled my score drops to 605

EDIT:
my temps hover around 60*c according to Ryzen Master and my single core sustains 4750MHz with the bios settings that hinder performance in R20 ST.

EDIT2:
With AMD Cool&Quiet, CPPC, CPPC preferred cores, and Processor Performance autonomous mode enabled, the single core MHz speed increases to a sustaining avg of 4799 and temps around 65*c

 

[Kenmitch]

Am I doing something wrong?

Yep....Your worrying about a few points!

 

[BTRY B 529th FA BN]

@Kenmitch There are a few things that influence ST outcome. Two or three points is fair to say but more that imo is worth looking into. One setup example brings 605 while a few tweaks brings 616. This is partly from Processor idle disable in Processor Power settings. So sometimes tweaks can cause for a negative influence.

 

[Kenmitch]

@Kenmitch There are a few things that influence ST outcome. Two or three points is fair to say but more that imo is worth looking into. One setup example brings 605 while a few tweaks brings 616. This is partly from Processor idle disable in Processor Power settings. So sometimes tweaks can cause for a negative influence.

What MB are you using? It could be nothing more then the current uEFI version it's running. Did you test with a previous version?

I never usually run the single threaded one, but I'm about to fire up my 5900x rig in a couple of moments and I'll see what she gets.

 

[Makaveli]

Your score looks fine to me I wouldn't worry about it.

 

[BTRY B 529th FA BN]

I remember of the first runs I had it was 627 and inline with most of the 5800X review/bench results. I think that was the first bios for Ryzen 5 series. This 5800X is on a X570 Xtreme. Currently using F33f bios, but it's been this way for more than a couple of the previous bios'

 

[inf64]

It is 1% deviation versus the AT results, that is less than margin of error. A slight secondary memory timing difference can easily net more than 1% difference.

 

[Kocicak]

Are you using any monitoring software while you run benchmarks?

 

[thigobr]

Your score looks normal for PBO disabled.

One thing to have in mind is that Cinebench is not super sensitive to memory overclock and increasing the Infinity Fabric/Memory controller clocks along with vSOC will eat into the CPU power budget. That's not the case for Single thread benchmark but it can certainly reduce scores for benchs that are less memory sensitive as the CPU will have lower power budget to work with when memory/SOC is overclocked.

On my 5950X for example I see SOC power going up from 9W (2666MHz memory) to 18W (3800MHz memory)...

 

[Kenmitch]

Are you using any monitoring software while you run benchmarks?

This makes a 3-4 point difference with my 5900x having HWiFO64 running in the background.

 

[Makaveli]

This makes a 3-4 point difference with my 5900x having HWiFO64 running in the background.

I normally find its best to run this after a reboot and closing all unnecessary apps that are running in the system tray. Even having steam running can affect your score.

 

[BTRY B 529th FA BN]

I'm not running anything else besides R20 ST. I don't even like to move my mouse when it's running.

Keeping AMD Cool&Quiet helps keep the thread on the best cores. Doesn't seem to matter if C-States is disabled or enabled.

Processor idle disable - disable idle, makes a big difference so I'm temporarily keeping it enabled. Mouse movement feels better with it disabled however there's suppose to be a fix for AMD mouse issues with Ryzen 5k, 3k, and X470 & X570 configurations with AGESA 1.2.0.2.

I can't seem to hit 622 again. I'm hovering around 614-616 now.

 

[Kocicak]

Moth hitting a lightbulb.