QOTW: What do you score in Cinebench Release 23?

[trillo_del_diavolo]

You won't want to add this to any graphic.

Single: 652
Multi: 2069

A10-7860K @4.5GHz
ASRock FM2A88X Extreme4+

EDIT: some intriguing results under Linux.

[ik9000]

I'd be really surprised if the single core bench throttles. I was watching my core temps using HWInfo64 when running the single core and it looks like the process for Cinebench was being moved around the different cores while running, not sure if anyone else notices this behaviour if they monitor the core temps. I only noticed it because I was curious what a light overclock would do to my scores (set all cores to run static @4.9 = 12600/1305).

Yeah Zen should move load around to the best cores IIRC. Not really sure how it does that, but seem to remember people saying it did this. Re throttling I was meaning more on multi core than single thread - due to accumulated heat from many cores. I wondered if it might account for that missing 10%? There's more than enough RAM in Beff'a system.

Edit: when Beffa's manual OC score is applied it hits the right place to correlate with the other CPUs that are doing it natively. This suggests there is something holding back the auto-overclock on multithreaded performance, that doesn't necessarily need to be doing so. Perhaps the PBO settings needs adjusting?

[ik9000]

You won't want to add this to any graphic.

Single: 652
Multi: 2069

A10-7860K @4.5GHz
ASRock FM2A88X Extreme4+

Don't be silly- added (and not the worst score either)

[ik9000]

Latest charts (absolute and normalised) - using thumbnail links now as they're getting large. Thanks all and keep 'em coming.

absolute (ranked by multi cpu score)


normalised (but still ranked by multi cpu score)


normalised and ranked by multi core score/no. threads


normalised and ranked by single core score

[beffa]

I'd be really surprised if the single core bench throttles. I was watching my core temps using HWInfo64 when running the single core and it looks like the process for Cinebench was being moved around the different cores while running, not sure if anyone else notices this behaviour if they monitor the core temps. I only noticed it because I was curious what a light overclock would do to my scores (set all cores to run static @4.9 = 12600/1305).

Had the same experience when running single core test.
All over this proves that AMD had made a great single core score for all their Zen3 processors, also shown in benchmark tests on YT

[ik9000]

Had the same experience when running single core test.
All over this proves that AMD had made a great single core score for all their Zen3 processors, also shown in benchmark tests on YT

As mentioned above: your normalised manual OC score hits the right place to correlate with the other 5000 series CPUs that are doing it natively. This suggests there is something holding back the auto-overclock on multithreaded performance that doesn't necessarily need to be doing so. Perhaps the PBO settings needs adjusting? Did you have PBO on for the first test?

[ZodiacG66]

Asus strix G15 Laptop (2021)
5800H/3070 RTX

12600 Multi . . . not bad for a 45w cpu

[ik9000]

Asus strix G15 Laptop (2021)
5800H/3070 RTX

12600 Multi . . . not bad for a 45w cpu

single score?

[ZodiacG66]

Single core 1363

[beffa]

As mentioned above: your normalised manual OC score hits the right place to correlate with the other 5000 series CPUs that are doing it natively. This suggests there is something holding back the auto-overclock on multithreaded performance that doesn't necessarily need to be doing so. Perhaps the PBO settings needs adjusting? Did you have PBO on for the first test?

When I first looked at your charts with score pr. core I wonder what the **** was wrong, cleared CMOS, flashed bios and still was not able to do better than last night. Then I had to search the big Internet for other that had a 5950x and had made a Cinebench23 test. It looks like my scores was spot on for stock and OC, the problem of a top of the line CPU is almost the same as the GPU. You have to pay 100% more money to get 15 to 30% more power, with some tiny adjustments I might break the 30K. Also had some thoughts of the amount of RAM holding my score back a tiny bit

First test was done with stock settings, re-tested today after clearing CMOS and flashing Bios, got results like +/- 1640/21500, then did some tests with PBO and got +/- 1630/29800

[Ripcord]

MULIT 6496
SINGLE 874

Ryzen 5 1600
Ram 3000Mhz
B350

[Percy1983]

3700x
Single 1263
Multi 12347

[probedb]

Ancient i5-4690K
Multi - 3267
Single - 854

[Gentle Viking]

Okay since you ask so polite, and i can DL from Microsoft where i will get the full gigabit Dl speed.

System spec stock CPU speed on the 1920X and just 48GB or RAM at the moment.

Multi: 12464
Single: 966

[Rob_B]

i3 7100T (2C/4T)
1836 Multi
693 Single

i3 10100F (4C/8T)
5203 Multi
1114 Single

[Bionor]

Single Core: 1631
Multi Core: 21683

CPU: 5900X
Ram: 3600@CL16
MB: Asus X570 ROG Gaming E