F5F Stay Refreshed Hardware Desktop Comparison of 3600mhzCL18 and 3600mhzCL14 on Cinebench20 performance

Comparison of 3600mhzCL18 and 3600mhzCL14 on Cinebench20 performance

Comparison of 3600mhzCL18 and 3600mhzCL14 on Cinebench20 performance

Pages (2): 1 2 Next
R
RockyTFL
Member
69
04-07-2023, 12:31 AM
#1
I increased my RAM from 3600mhzCL18 to 3600mhzCL14 (both 2*8GB bundles). My Cinebench score got a slight boost, though not as much as expected. Previously, raising the frequency led to a significant jump in Cinebench. This round I kept the same speed but achieved noticeably faster times. Could the timing gains really impact the 1% lows more than the overall ceiling? For more details, see the full system spec in the signature.
R
RockyTFL
04-07-2023, 12:31 AM #1

I increased my RAM from 3600mhzCL18 to 3600mhzCL14 (both 2*8GB bundles). My Cinebench score got a slight boost, though not as much as expected. Previously, raising the frequency led to a significant jump in Cinebench. This round I kept the same speed but achieved noticeably faster times. Could the timing gains really impact the 1% lows more than the overall ceiling? For more details, see the full system spec in the signature.

X
xboxfldig
Junior Member
42
04-07-2023, 12:39 AM
#2
Cinebench, both old and recent releases, showed minimal sensitivity to RAM speed. The slight variation in the screenshots is under 1%, which might mean other factors play a larger role unless you ran several tests to improve it.
X
xboxfldig
04-07-2023, 12:39 AM #2

Cinebench, both old and recent releases, showed minimal sensitivity to RAM speed. The slight variation in the screenshots is under 1%, which might mean other factors play a larger role unless you ran several tests to improve it.

H
horrie
Member
71
04-07-2023, 07:46 AM
#3
Kinda looks like a little bit of run-to-run variance but hey, that's something at least, now that you mention it I'm gonna try that with my own ram
H
horrie
04-07-2023, 07:46 AM #3

Kinda looks like a little bit of run-to-run variance but hey, that's something at least, now that you mention it I'm gonna try that with my own ram

N
natsu40
Member
239
04-08-2023, 12:24 PM
#4
My recent efforts have achieved an average score of: CL18: 4450 CL14: 4550. That should be around 100 points with the faster RAM speeds. I was hoping for a much larger improvement.
N
natsu40
04-08-2023, 12:24 PM #4

My recent efforts have achieved an average score of: CL18: 4450 CL14: 4550. That should be around 100 points with the faster RAM speeds. I was hoping for a much larger improvement.

L
lantern_might
Junior Member
2
04-08-2023, 03:37 PM
#5
Looking into your outcomes would be fascinating! I won’t introduce any new twists right now since my bottom position is blocking the CMOS reset. I plan to install an old reset switch so there’s a dedicated ‘clear CMOS’ button during the loop rebuild in summer.
L
lantern_might
04-08-2023, 03:37 PM #5

Looking into your outcomes would be fascinating! I won’t introduce any new twists right now since my bottom position is blocking the CMOS reset. I plan to install an old reset switch so there’s a dedicated ‘clear CMOS’ button during the loop rebuild in summer.

_
_Gifirax_PvP_
Member
114
04-16-2023, 12:05 AM
#6
It seems my setup is running smoother and more efficiently. I suspect my low points will get better soon.
_
_Gifirax_PvP_
04-16-2023, 12:05 AM #6

It seems my setup is running smoother and more efficiently. I suspect my low points will get better soon.

K
KewlDerp
Member
54
04-16-2023, 02:06 AM
#7
The images display 7526 and 7481, indicating a difference in the results. If you have the detailed scores, I can compare the highest values from each set. Competitive evaluators typically focus on these top numbers to demonstrate the strongest performance. Other system factors might affect speed, but anything beyond intentional shortcuts won<|pad|> can't exceed natural limits.
K
KewlDerp
04-16-2023, 02:06 AM #7

The images display 7526 and 7481, indicating a difference in the results. If you have the detailed scores, I can compare the highest values from each set. Competitive evaluators typically focus on these top numbers to demonstrate the strongest performance. Other system factors might affect speed, but anything beyond intentional shortcuts won<|pad|> can't exceed natural limits.

D
djpumuslink01
Senior Member
577
04-16-2023, 09:10 AM
#8
For better performance, using 4 sticks is recommended if your imc supports it. Consider testing 4x8 cl18 and a 2% setting. RAM upgrades mainly improve capacity and minmaxing.
D
djpumuslink01
04-16-2023, 09:10 AM #8

For better performance, using 4 sticks is recommended if your imc supports it. Consider testing 4x8 cl18 and a 2% setting. RAM upgrades mainly improve capacity and minmaxing.

B
BlackWard
Junior Member
2
04-22-2023, 11:43 AM
#9
Test your performance with Run Geekbench 3 to discover your potential scores. The results will likely show a noticeable improvement.
B
BlackWard
04-22-2023, 11:43 AM #9

Test your performance with Run Geekbench 3 to discover your potential scores. The results will likely show a noticeable improvement.

S
SiphonicVirus
Member
201
04-22-2023, 01:23 PM
#10
I also sense that a more optimized model might produce better results.
S
SiphonicVirus
04-22-2023, 01:23 PM #10

I also sense that a more optimized model might produce better results.

Pages (2): 1 2 Next