Question Cinebench 2024 Low Score (665 points) ?

[dann252]

Hi Guys Good day

I tested my i5-12400F CPU using Cinebench 2024 using Multicore Test.. After the test is complete I noticed my score is around 665 - 666 points.
But i saw this article and the score for an i5-12400 should be around 701 points:
https://www.cgdirector.com/cinebench-2024-scores/

These are the settings i used when i runned the test.

BIOS:
Power Limits: changed from Tower Cooling to Water Cooling
Enabled XMP: Profile 1 - 3200 Mhz

OS: Windows 11 Pro 24H2
Power Settings: changed from Balanced to High Performance

Specs
Mobo: MSI MAG B760 Mortar WiFi DDR4
CPU: Intel Core i5-12400 F
GPU: Palit RTX 3060 12GB
RAM: Kingston Fury Beast RGB 2x16GB 3200 Mhz
PSU: MSI MAG A850GL PCIe5 850w
Storage: 1 x SATA 500GB SSD, 1 x SATA 4TB HDD, 1 SATA 1TB HDD

Here are some screenshots when i ran the test:

Cinbebench 2024 score
https://drive.google.com/file/d/1KrGKNJ1iaKMU_vJv3EXHfED8aTcsFrTy/view?usp=drive_link
HW Temps and Voltages etc.
https://drive.google.com/file/d/1NUnLG7F2JF3jiVjyqOeN2AhTpLqvowRA/view?usp=drive_link

IMG_20250209_175113.jpg

drive.google.com

https://drive.google.com/file/d/1I07hy6ISiJJYvJ5OGz1OP_wk7N8GpZvg/view?usp=sharing

What should i do to increase my score to at least 701 pts? as seen on these article https://www.cgdirector.com/cinebench-2024-scores/
others suggest on reddit to adjust the CPU V-Core but he is using a b660 Asrock Mobo
Im not sure whats the naming scheme of CPU Vcore in MSI motherboard
i still haven't touched anything on the advanced cpu page of the bios yet. but here's the screenshot of my bios. please let me know what to adjust for my CPU for me to atleast squeeze a little bit of performance. and reach a higher Cinebench 2024 score.

BIOS Advanced CPU setup
https://drive.google.com/file/d/1d8JSvTxhM_EeqLrWdAqik0F4HLFBPVoi/view?usp=sharing


Thanks

 

[Ralston18]

What specific problem are you trying to solve?

If there is nothing wrong and you are satisfied with your system's performance then do not go "chasing numbers".

That said:

Accuracy: the first question to ask is: "what is the + or - margin of error in the points determined by the testing tool?

Your 666 point score, for example, is only 35 points away from the cited 701.

5% or so.

Very likely that your build is different any number of ways if compared to the build with 701 points.

However, I would probably be curious about how the points would change if I disconnected the three storage drives and ran only the boot drive. Especially the storage HDD's. Changing nothing else.

Then reconnect the storage drives one by one (again changing nothing else) and running the test again each time.

How does each drive's presence or absence affect your build's CPU score?

Then maybe try some other things to learn what system changes on your build affect the CPU point score.

Be methodical, keep notes. Then decide if you wish to push it all further with overclocking and voltages.

Will leave it to the overclocking community to provide other ideas and suggestions.

 

[dann252]

What specific problem are you trying to solve?

If there is nothing wrong and you are satisfied with your system's performance then do not go "chasing numbers".

That said:

Accuracy: the first question to ask is: "what is the + or - margin of error in the points determined by the testing tool?

Your 666 point score, for example, is only 35 points away from the cited 701.

5% or so.

Very likely that your build is different any number of ways if compared to the build with 701 points.

However, I would probably be curious about how the points would change if I disconnected the three storage drives and ran only the boot drive. Especially the storage HDD's. Changing nothing else.

Then reconnect the storage drives one by one (again changing nothing else) and running the test again each time.

How does each drive's presence or absence affect your build's CPU score?

Then maybe try some other things to learn what system changes on your build affect the CPU point score.

Be methodical, keep notes. Then decide if you wish to push it all further with overclocking and voltages.

Will leave it to the overclocking community to provide other ideas and suggestions.

hi tbh im just chasing the 701 pts. generally im already satisfied with my machines performance. i just want to squeeze it a bit if possible XD. right now im testing the CPU lite Load on the Advanced Bios settings so far what i have learned the more i reduce the Lite mode ex. from 12 to 1 the Wattage and temps are reduced at the cost of little performance tho... im trying to find sweet spot maybe mode 9 or 10 will give me more perf. with low power draw and temps.

 

[TerryLaze]

hi tbh im just chasing the 701 pts. generally im already satisfied with my machines performance. i just want to squeeze it a bit if possible XD. right now im testing the CPU lite Load on the Advanced Bios settings so far what i have learned the more i reduce the Lite mode ex. from 12 to 1 the Wattage and temps are reduced at the cost of little performance tho... im trying to find sweet spot maybe mode 9 or 10 will give me more perf. with low power draw and temps.

Stop everything that is running in the background and don't even move the mouse, also run the benchmark at least three continuous times and take the average of these runs, that's how most benchmarkers do it.

The benchmarks you see on the net are on a clean install and under controlled conditions.

 

[Ralston18]

Seems reasonable overall.

No harm in experimenting or testing provided that backups are in place to recovery when necessary.

Likewise, per @TerryLaze, the benchmarking process must be controlled and methodical.

Good way to learn more about your system and computers overall.

Versus chasing some small MOE point benchmark difference and then randomly installing or doing "X" from somewhere in the internet world where X (free or otherwise) claims you will get some far greater number of points. That is likely to end badly.