F5F Stay Refreshed Power Users Overclocking Are these overclocking outcomes typical for your Ryzen 5 5600X setup?

Are these overclocking outcomes typical for your Ryzen 5 5600X setup?

Are these overclocking outcomes typical for your Ryzen 5 5600X setup?

C
211
03-26-2020, 06:52 AM
#1
I have reached these findings after extensive testing. Is this sufficient for reliable performance?
C
coolbencool890
03-26-2020, 06:52 AM #1

I have reached these findings after extensive testing. Is this sufficient for reliable performance?

B
BougGroug
Member
118
03-26-2020, 07:02 AM
#2
Run CPU benches for long periods of time (at least 1 hour) to validate CPU and OC stability. Cinebench 2024, Prime95, AIDA64. Take your pick. (E.g PRIME95 has solid torture test in it.)
If CPU is stable, then OC is stable.
But i wouldn't call 100 Mhz over boost clocks an "OC". There are 0 real life differences that you can notice.
Modern CPUs are so fine tuned that they have little, if any OC headroom (frequencies over max turbo ratio). That much you can also see from your chip.
Back in the day, with older CPUs, CPU OC was worthwhile.
Now, if i'd take old chip, e.g like my i5-6600K which has 3.5 GHz base and 3.9 Ghz boost clocks. With CPU OC, i could get it 4.5 Ghz all core (increase of 600 Mhz over boost), or with delid, ~4.7 Ghz all core (800 Mhz over boost). And there have been some delidded i5-6600K CPUs, that can hold 5 Ghz all core.
500+Mhz over boost clocks is what i'd call "CPU OC".
B
BougGroug
03-26-2020, 07:02 AM #2

Run CPU benches for long periods of time (at least 1 hour) to validate CPU and OC stability. Cinebench 2024, Prime95, AIDA64. Take your pick. (E.g PRIME95 has solid torture test in it.)
If CPU is stable, then OC is stable.
But i wouldn't call 100 Mhz over boost clocks an "OC". There are 0 real life differences that you can notice.
Modern CPUs are so fine tuned that they have little, if any OC headroom (frequencies over max turbo ratio). That much you can also see from your chip.
Back in the day, with older CPUs, CPU OC was worthwhile.
Now, if i'd take old chip, e.g like my i5-6600K which has 3.5 GHz base and 3.9 Ghz boost clocks. With CPU OC, i could get it 4.5 Ghz all core (increase of 600 Mhz over boost), or with delid, ~4.7 Ghz all core (800 Mhz over boost). And there have been some delidded i5-6600K CPUs, that can hold 5 Ghz all core.
500+Mhz over boost clocks is what i'd call "CPU OC".

M
Mozeus
Junior Member
14
03-26-2020, 08:04 AM
#3
Hey there,
I'm running a moderate OC on my 5600x with solid results—4.7/4.75ghz all-core and 4.85ghz single/dual-core. Your numbers are almost the same, temperatures are similar too. I managed 73c with CB and Prime pushes get much higher temps, but my stable run is around 83c in Prime95. I have a 240 AIO.
Why are you targeting 73c? You're missing out on some performance by setting it that low. But your scores confirm it's a solid upgrade. Maybe try 80c and you could score a bit higher on synth benches I'd expect.
M
Mozeus
03-26-2020, 08:04 AM #3

Hey there,
I'm running a moderate OC on my 5600x with solid results—4.7/4.75ghz all-core and 4.85ghz single/dual-core. Your numbers are almost the same, temperatures are similar too. I managed 73c with CB and Prime pushes get much higher temps, but my stable run is around 83c in Prime95. I have a 240 AIO.
Why are you targeting 73c? You're missing out on some performance by setting it that low. But your scores confirm it's a solid upgrade. Maybe try 80c and you could score a bit higher on synth benches I'd expect.

J
jjsoini
Posting Freak
809
04-07-2020, 03:31 PM
#4
The picture I included displays the 17-hour Prime95-based Corecycler outcome. Adjusting overclocking on older chips proved more successful, which aligns with your comments.
J
jjsoini
04-07-2020, 03:31 PM #4

The picture I included displays the 17-hour Prime95-based Corecycler outcome. Adjusting overclocking on older chips proved more successful, which aligns with your comments.

M
McJoelPlayz
Member
65
04-11-2020, 06:42 PM
#5
Ensuring all cores operate at the same frequency is not ideal for my silicon. I require a setting around 1.45v at 4.6 GHz. I believe 73 degrees will provide the best results. After removing power and temperature limits, Cinebench R23 Multi Core scored 12002 points, but the temperature reached 82-83 degrees. I aim to keep it below 85 degrees during extended runs. The performance drop is minor, so it's not a major issue. Appreciate your feedback.
M
McJoelPlayz
04-11-2020, 06:42 PM #5

Ensuring all cores operate at the same frequency is not ideal for my silicon. I require a setting around 1.45v at 4.6 GHz. I believe 73 degrees will provide the best results. After removing power and temperature limits, Cinebench R23 Multi Core scored 12002 points, but the temperature reached 82-83 degrees. I aim to keep it below 85 degrees during extended runs. The performance drop is minor, so it's not a major issue. Appreciate your feedback.

K
Klitaurus
Member
69
04-11-2020, 08:14 PM
#6
Ouch! That's not ideal. Sorry about that. Have you considered moving the LLC up to level two or medium, depending on your board's capabilities? This might improve vdroop and stability.
Still, my rig has only been above 12k a few times. I'm consistently around 75c on CB multi, mostly near the 11900 mark for points. That means you're getting a solid output. It's working well.
Upgrading to a better cooler would also provide more thermal margin.
K
Klitaurus
04-11-2020, 08:14 PM #6

Ouch! That's not ideal. Sorry about that. Have you considered moving the LLC up to level two or medium, depending on your board's capabilities? This might improve vdroop and stability.
Still, my rig has only been above 12k a few times. I'm consistently around 75c on CB multi, mostly near the 11900 mark for points. That means you're getting a solid output. It's working well.
Upgrading to a better cooler would also provide more thermal margin.

_
_pawol_
Member
102
04-13-2020, 03:25 PM
#7
Regardless of the Load Line Calibration setting I choose, it produces fewer results compared to the Auto option. My DGI+VRM configuration is at 130%, and I'm using it in Extreme mode. It seems this is the only option without improved cooling.
_
_pawol_
04-13-2020, 03:25 PM #7

Regardless of the Load Line Calibration setting I choose, it produces fewer results compared to the Auto option. My DGI+VRM configuration is at 130%, and I'm using it in Extreme mode. It seems this is the only option without improved cooling.

C
210
04-21-2020, 11:22 AM
#8
Good luck with your project. Please share your progress!
C
CreeperRocket1
04-21-2020, 11:22 AM #8

Good luck with your project. Please share your progress!