Overclock.net banner
Cancel
Create thread New Forums

9800X3D on 360mm AIO temps

3 reading
Jump to Latest
1 - 20 of 27 Posts

CntrlAltDel

· Registered
Joined
·
26 Posts
Discussion starter · #1 · (Edited)
Hi there,
I recently switched from a ID-Cooling FROZN A720 140mm Dual-Tower Air Cooler to the Arctic Freezer III 360mm A-RGB v2 AIO and I'm seeing the same temps that I saw with my air cooler.

PBO is set to CO -25 + Offset 200MHz and Scalar 10x

ASUS ROG STRIX X670E-F Gaming WIFI
Corsair Vengeance DDR5-6000 CL30 32GB Kit

On the ID-Cooling FROZN A720 140mm Dual-Tower Air Cooler I get 92°C after 10 minutes of Cinebench runs.
On the Arctic Freezer III 360mm AIO I get 91°C after 10 minutes of Cinebench runs.

Text Font Screenshot Number Software
Text Number Font Screenshot Software
Text Screenshot Number Font Software
 
#2 ·
That's probably because "beefy" air coolers have more or less same performance of AIOs unless we are talking about top tier (if they exist) AIOs. Hence why when I switched from air to water, I went straight to custom loop (and even that, to get a big difference in between air and water you need some rads). After researching about it, the difference in between a good air cooler, and a AIO was about 2-3C at the time. That might have changed, can't say. What temperatures are you getting?

However, did you applied thermal paste properly? Is the pump of the AIO running at optimal speed? Also, why do you need the Scalar 10x? Try leaving that on AUTO your performance ain't going to change much and the lifespan of your 9800x3d will say thanks.

Also, 9800x3d is already a pretty cool chip, maybe that's the reason why you can't see much difference. Mine basically idles at 35 and hoovers around 50-65C depending on the game. CO -25/200Mhz no Scalar.
 
#3 ·
The 9800X3D shouldn't get that hot, it runs quite cool due to the way the cache is now placed vs the 7800X3D. I was using a D15S with my 7800X3D. Temps would reach close to 90C (PBO -30). That's just the nature of the chip. SWitching to the LFIII dropped temps to around 80C.
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #4 · (Edited)
@thomasck @Kookiemonster75 Sorry guys, I pulled an all nighter trying to wrap my head around this issue. So I completely forgot to add the temp data aside from the screenshots.

thomasck said:
After researching about it, the difference in between a good air cooler, and a AIO was about 2-3C at the time. That might have changed, can't say. What temperatures are you getting?
Click to expand...
On the ID-Cooling FROZN A720 140mm Dual-Tower Air Cooler I get 92°C after 10 minutes of Cinebench runs.
On the Arctic Freezer III 360mm AIO I get 91°C after 10 minutes of Cinebench runs.
Both runs operating at around 1.2v.

thomasck said:
However, did you applied thermal paste properly? Is the pump of the AIO running at optimal speed? Also, why do you need the Scalar 10x? Try leaving that on AUTO your performance ain't going to change much and the lifespan of your 9800x3d will say thanks.
Click to expand...
I'm 100% certain the thermal paste application is spot on correct. I would've also gotten way more aggressive temp spikes much quicker if this were not the case. I will however re-seat the cooler just to make sure.

thomasck said:
Is the pump of the AIO running at optimal speed?
Click to expand...
Pump is using AIO_PUMP header which is running at 100% at all times.

thomasck said:
Also, why do you need the Scalar 10x? Try leaving that on AUTO your performance ain't going to change much and the lifespan of your 9800x3d will say thanks
Click to expand...
It's a theoretical lifespan decrease. The voltage delta is minuscule, I've tested scalar at 1x vs 10x and I saw about 0.015v difference. There was also no noticeable temperature changes, still exactly the same numbers.

thomasck said:
Also, 9800x3d is already a pretty cool chip, maybe that's the reason why you can't see much difference. Mine basically idles at 35 and hoovers around 50-65C depending on the game. CO -25/200Mhz no Scalar.
Click to expand...
The synthethic benches seem to beat this CPU up. I think you will find quite a significant difference in your temps if you run something like Cinebench. I get between 50-65°C exactly as you do in-games as well.

Kookiemonster75 said:
The 9800X3D shouldn't get that hot, it runs quite cool due to the way the cache is now placed vs the 7800X3D. I was using a D15S with my 7800X3D. Temps would reach close to 90C (PBO -30). That's just the nature of the chip. SWitching to the LFIII dropped temps to around 80C.
Click to expand...
What you are saying seems to line up with what alot of others are saying. Just very strange that at 135w on both runs with an air cooler vs 360mm AIO the temps are near identical.
 
#5 ·
The chip will often boost to it's thermal limits - in that case temps will be the same when using different coolers so you have to go by average clock or benchmark score in something like cinebench to reveal cooling performance.
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #6 ·
cssorkinman said:
The chip will often boost to it's thermal limits - in that case temps will be the same when using different coolers so you have to go by average clock or benchmark score in something like cinebench to reveal cooling performance.
Click to expand...
I've read quite a bit of people saying this. I do see the thermal limiter in the BIOS as well. Even though set to auto, I see a 10-20 score difference in Cinebench R23 runs which is negligible. Anecdotally speaking, I don't feel this is the case.
 
#7 ·
CntrlAltDel said:
I've read quite a bit of people saying this. I do see the thermal limiter in the BIOS as well. Even though set to auto, I see a 10-20 score difference in Cinebench R23 runs which is negligible. Anecdotally speaking, I don't feel this is the case.
Click to expand...
Do you use HWINFO while doing the comparisons?
 
Save
Reply Quote Rep+ Rep+
#9 ·
CntrlAltDel said:
Yes, I do. I use both mobo and CPU sensors and compare them based on that data. You can see an example in the first post under the spoiler.
Click to expand...
It's possible that the coolers are just that close in capability as others have mentioned. The max temp you reported on the AIO indicates that it's very very close to handling the heat load at the maximum boost clocks under current settings. ( 90c is pretty common on similar machines I've built as the value where boost clocks are limited).

Dual 140 = 280 surface area vs 360mm surface area but there are three dead spots under the fan centers vs at most 2 on the tower cooler. Not sure how the thickness's compare though.

The newest thermalright coolers are very impressive in their abilities. I've used the Frost commander 140's and peerless assassin 120's -will have the 120 phantom spirit EVO on a 9800x3d as soon as the cpu shows up ( Feb 10? its on backorder). The case that will go in has a cheap 360 mm AIO in it at the moment, could do a comparison fairly easily.

Might lower fan speeds on the 360mm and see if the scores drop on a longer cinebench run
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #11 · (Edited)
Alright, so I just dismounted the pump. Cleaned off the thermal paste with isopropyl alcohol. Reseated the Thermalright AM5 v2 secure frame bracket. Reapplied thermal paste, this time much thicker.
Now my temps appear to be behaving a bit better.

This is with PBO set to -30 CO; 200MHz Boost Clock Override; 10x Scalar

Idle is now 43°C; Max is no longer at 91°C, Average is nicely at 81°C

I feel I did everything the same 1:1 in the remounting except for the additional thermal paste. However I did see the spread of the thermal paste on the previous mounting was what I would consider to be a really good spread. So I honestly don't know what was the culprit here.

Text Number Font Screenshot Technology

Text Number Font Screenshot Technology

Text Font Screenshot Number Software

Text Screenshot Font Number Software
 
#14 ·
Now try pbo - 20 and see if your score increases.
 
Save
Reply Quote Rep+ Rep+
#15 ·
CntrlAltDel said:
What temps are you getting?
Click to expand...
I have a 360/60 cooler and a block so there is no comparison. At 24 degrees in the room, after 10 minutes CB R23 is 84 degrees (Phanteks T30 - 850 rpm) and 24145 points (with HwInfo open)
 
Save
Reply Quote Rep+ Rep+
Discussion starter · #16 ·
ju-rek said:
I have a 360/60 cooler and a block so there is no comparison. At 24 degrees in the room, after 10 minutes CB R23 is 84 degrees (Phanteks T30 - 850 rpm) and 24145 points (with HwInfo open)
Click to expand...
That's very similar in comparison though. Room temp on my end is 28°C. And my last post with a 10 minute Cinebench R23 run and mustered roughly 1°C difference from yours. Although not sure what voltages you're running at, regardless with that thick of a radiator and custom waterblock I'd expect your temps to be quite a bit lower.
 
#17 ·
CntrlAltDel said:
That's very similar in comparison though. Room temp on my end is 28°C. And my last post with a 10 minute Cinebench R23 run and mustered roughly 1°C difference from yours. Although not sure what voltages you're running at, regardless with that thick of a radiator and custom waterblock I'd expect your temps to be quite a bit lower.
Click to expand...
Just be happy, you are better off than me on CLC :)
 
Save
Reply Quote Rep+ Rep+
#20 ·
Just adding some data points - Arctic LF3 360 with contact frame, 60% pump speed, fans set to ramp up to 80% max, seeing 5357MHz clock average, 85.8C average over 10 minutes of R23 at 22C room temp. Per core CO with +200MHz clock, 6x scalar, 1.194V Core VID average after 10 minutes.
 
  • Rep+
Reactions: PROBN4LYFE
Save
Reply Quote Rep+ Rep+
1 - 20 of 27 Posts
About this Discussion
26 Replies
12 Participants
Last post:
PROBN4LYFE
Overclock.net
posts
28M
members
562K
Since
2004
The premier forum for overclocking experts and enthusiasts. Discuss hardware optimization, custom builds, benchmarking, cooling solutions, and pushing the boundaries of computing performance. From beginner guides to extreme overclocking, join our technical community to master system tuning.
Full Forum Listing
Explore Our Forums
Intel CPUs Hardware News NVIDIA PC Gaming Water Cooling