The Corsair User Forums  

Go Back   The Corsair User Forums > Corsair Product Discussion > Cooling

Notices

Reply
 
Thread Tools Search this Thread Rate Thread Display Modes
  #1  
Old 08-02-2020, 01:47 PM
M3k4 M3k4 is offline
Registered User
M3k4's PC Specs
 
Join Date: Jul 2020
Posts: 3
POST ID # = 1056211
M3k4 Reputation: 10
Default CORSAIR iCUE H150i RGB PRO XT weaker than COOLER MASTER MASTERLIQUID LITE 120 ?

I greet you and apologize if the reading is too long, but I tried to describe the situation by providing as much data as possible, to understand the comparison made by me in the idea of receiving from you, possible tips to lower the CPU temperature Using the AIO CORSAIR system cooling.

The situation is as follows:
1. Until recently I had an AMD Ryzen 5 2600 processor in the system, for the cooling of which I used COOLER MASTER MASTERLIQUID LITE 120 (a 120mm fan and radiator mounted in the Zalman S3 housing on the rear panel-Intake).
It paid off, in the sense that the CPU temperature never exceeded 33 ° C in Idle and during work it never exceeded 43 ° C. It was also extremely quiet, the fan did not exceed 1500-1600rpm and I could only hear the pump if I listened close enough to it 5cm (it was a slight rustle).
2. When I bought an AMD Ryzen7 3700X I decided to buy something more powerful for CPU cooling. So I bought Corsair Hydro Series ™ H100i PRO RGB. I was totally disappointed: From inside the pump he could hear the sound of pickets hitting something. In iCUE software, the coolant temperature was 29-30 ° C, but the temperature of the new processor was in Idle 67-70 ° C! (according to HW info, AMD Ryzen Master or MSI Command Center - all looked the same 67-70 ° C). The processor runs at 2200-2600Mhz. The AIO system from Corsair was mounted in the same Zalman S3 housing on the front panel (without front mask - Intake).
3. I disassembled and sent back the 'Corsair Hydro Series ™ H100i PRO RGB' and mounted back the COOLER MASTER MASTERLIQUID LITE 120 on the back panel. I was quiet again: the temperature (checked with the same programs) was at 40-45 ° C with the new Ryzen 7 3700X processor, running at 2200-2600Mhz in idle.
4. In the meantime I ordered the case Lian Li O11 Dynamic and CORSAIR iCUE H150i RGB PRO XT (I thought it was the best choice, having the set of 3 pieces QL RGB also from CORSAIR but I was disappointed again - from inside the pump you can hear from 1m with all the PC case panel mounted, a broom that increases in intensity depending Performance Profile from iCUE).
a) At the beginning I mounted the radiator and the fans on the side panel-intake, and the 3 QL-RGB at the bottom-intake - CPU temp 59-68 ° C in idle 2200-3000Mhz, in iCUE coolant 30.2 ° C.
b) I mounted the radiator on top with the intake fans: the same result for the CPU at the same operating frequency (but at MOS the temperature has risen over time to 51 ° C only with Google Chrome - no video stream or anything else that consumes resources).
c) Now I have mounted the fan at exhaust (with radiator in the same position: top) and I notice only a slight improvement: temp. CPU 45.18 ° C at 1062Mhz in idle. In iCUE software, coolant 32.9 ° C.
I specify that I do not have Overclocking and I do not run anything special in these moments when I read these temperatures. I am surprised by the low temperatures that other users with more powerful processors have (AMD 3900X / 3950X, Intel I9 10900, etc.

Why a cooling system with a 120mm radiator can keep at the same CPU (AMD 3700X) the temperature at 40-45 ° C / 2200-2600Mhz, and this system with a 360mm radiator fails to lower the temperature at least to the values obtained with the system from COOLER MASTER, on the contrary only at 1062Mhz in idle CPU temp is 45.18 ° C?

Best regards.
Reply With Quote


  #2  
Old 08-02-2020, 05:56 PM
LeDoyen LeDoyen is offline
Registered User
LeDoyen's PC Specs
 
Join Date: Mar 2020
Posts: 300
POST ID # = 1056228
LeDoyen Reputation: 26
Default

When you were using the Coller master you did not have iCUE running in the background.
iCue keeps Ryzen boosting, using higher voltages and heating up a bit more, because it uses too much CPU time.

Also, the Cooler master fan is made for static pressure and is more efficient at pushing air through dense fins of a radiator.
QL fans have weak static pressure and need more speed to give the same result (they are more of a case fan than a radiator fan really).

Since you do not overclock, a 120mm rad is totally adequate for the task, so your results do not surprise me too much for idle.

Now, do you have other fans in the O11? and what is the typical ambient temperature where your PC is?
Reply With Quote


  #3  
Old 08-03-2020, 04:27 AM
pontageek pontageek is offline
Registered User
pontageek's PC Specs
 
Join Date: Jun 2020
Posts: 20
POST ID # = 1056288
pontageek Reputation: 10
Default

Quote:
Originally Posted by LeDoyen View Post
When you were using the Coller master you did not have iCUE running in the background.
iCue keeps Ryzen boosting, using higher voltages and heating up a bit more, because it uses too much CPU time.
?
Does iCue really consume that much CPU processing power? I notice running nothing except start up apps CPU usage is around 5 % for me.

Now 1 piece of information that is missing is ambient temperature. Coolant temp at 30 degC is normal so I'm not sure why the CPU temp is rather high at idle. Maybe the core voltage is too high? Can you do CPU stress test at least 10 minutes and report back the CPU and coolant temps? I believe many long Corsair product users here can help you.
Reply With Quote


  #4  
Old 08-03-2020, 05:31 AM
LeDoyen LeDoyen is offline
Registered User
LeDoyen's PC Specs
 
Join Date: Mar 2020
Posts: 300
POST ID # = 1056300
LeDoyen Reputation: 26
Default

Yes it does. Just the corsair services do use a few % and iCue adds on top of it. It'ss enough to have lots of Ryzen CPUs always keep voltages up, leading to increased idle temps.
For me i was just like you, around 5 - 6% with nothing running. Uninstalling iCue makes it drop to 1%. That alone can explain idle temperatures.

The rest though, that quick rise in temp looks odd.
Reply With Quote


  #5  
Old 08-03-2020, 01:09 PM
M3k4 M3k4 is offline
Registered User
M3k4's PC Specs
 
Join Date: Jul 2020
Posts: 3
POST ID # = 1056352
M3k4 Reputation: 10
Thumbs up

@ LeDoyen Thanks for the reply, but ...

I apologize, but I don't understand how QL fans have to do with CPU cooling. Both Corsair kits have ML series fans: static pressure 4.2mm H2O... Are you saying that a single fan from Cooler Master has a static pressure of more than 3 pieces from the ML series from Corsair? If you say that a single 120mm radiator is enough, then why can't a 360mm radiator get at least the same results or better?
In idle in Task manager it shows me the processor load 5% of which iCUE uses 0.2-0,3%, Corsair Service 1,1%, WMI Provider Host 2,4%, … everything at a frequency of 1Ghz. Is this a lot and justifies a temperature over 60 ° C for the Corsair with a cooling surface 3 times larger? Then the same load of 5% but at a higher frequency 2.2-2.5Ghz ( I don´t now why) justifies a temperature of 40-45 ° C for the Cooler Master which has a cooling surface 3 times smaller? Let's not forget the difference between the volume of coolant in the 2 systems. Unfortunately, the numbers do not match! I honestly don't understand your explanation (maybe because I'm not a professional in this field).
If it doesn't surprise you, it surprises me in a rather unpleasant way because it is the second system from Corsair that offers the same results ... Even if there is no OC at the moment, maybe in the future I will use OC (for this reason 'I wanted and paid for better cooling or for the name'). But with these results I dare not even try ...
As I wrote in the first post, at the bottom of the case I have 3 QL fans - mounted for intake to cool the case / system. If you think that to run the part of the program responsible for the RGB effects of these 3 fans in the QL series contributes to raising the temperature of the processor, I specify that before buying them and now after being mounted ( or when RGB it is disconected) the results were the same: no differences!
The ambient temperature is constant between 25,1-26,3 ° C / System temp. 33,3°C.

All the best!
Reply With Quote


  #6  
Old 08-03-2020, 01:36 PM
LeDoyen LeDoyen is offline
Registered User
LeDoyen's PC Specs
 
Join Date: Mar 2020
Posts: 300
POST ID # = 1056354
LeDoyen Reputation: 26
Default

aaah i thought the AIO had QL fans, which i found odd.. nevermind then! i'll take a glass of milk and go to bed :p

Regarding CPU usage, remember you have 16 threads, i have 8. your percentages will be smaller but the absolute load will be roughly the same.

But again, it doesn't explain the massive difference.
at this stage i only see two possibilities : a deffective AIO, or bad contact between the cooling plate and the CPU.
Also make sure the pump is not running in "Quiet" mode, but since you said the noise changed depending on the performance mode, i guess you make it run faster.
Reply With Quote


  #7  
Old 08-03-2020, 02:35 PM
M3k4 M3k4 is offline
Registered User
M3k4's PC Specs
 
Join Date: Jul 2020
Posts: 3
POST ID # = 1056363
M3k4 Reputation: 10
Sad

Hi,
The cooling pump is set to Balanced, as are the fans. I heard that noise that changes in intensity, when I wanted to test each fan separately and the others were on ZERO RPM each. When I tested the pump to see if it has the same noise of pebbles hitting the inside walls as in the case of the previous Kit (240mm), I noticed that the brumm coming from inside changes depending on the speed setting in iCUE.

When I disassembled the first AIO kit from CORSAIR, the thermally conductive paste was spread evenly over the entire surface of the processor, and on the flat contact of the pump was a beautiful square with slightly rounded corners (exactly as it should be).

On the second processor (the current one), after the first 2 attempts with the radiator mounted on the side and on top (both on intake) I disassembled the pump from the processor again and I noticed again that the thermal paste was the same as In the first case: uniformly over the entire surface. I cleaned and put the thermal paste from Noctua, after which I mounted the pump again, taking care to be exactly in the middle referring to the AM4 socket and the screws to be equally tensioned. In total with the 2 processors and the 3 AIO cooling I mounted the water pump on the processor 5 times in total and each time it was applied and tensioned very well.

Defective AIO? Again? It's true that I've read quite a lot of negative feedback on Amazon, but really? Two one after the another? And both we received sealed were not returned by other customers.
Reply With Quote


  #8  
Old 08-03-2020, 03:52 PM
LeDoyen LeDoyen is offline
Registered User
LeDoyen's PC Specs
 
Join Date: Mar 2020
Posts: 300
POST ID # = 1056369
LeDoyen Reputation: 26
Default

I'll leave that to someone who knows AIOs more than i do. I only have suppositions at this stage since i mostly have experience on custom loop ^^'
Two in a row would really be surprising
Reply With Quote


  #9  
Old 08-03-2020, 08:47 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 11,715
POST ID # = 1056414
c-attack c-attack Reputation: 122
Default

When there is a problem with the cooler, either poor flow, a stuttering pump, fan speed to low, etc., it will result in a marked coolant temperature rise. From what you are describing, that does not seem to be the case.

When you get large differences between the coolant temp and core temps at idle, there are a few possible reasons why. Poor conductivity or contact between CPU and cold plate is one. You've stated the TIM application looked good and when contact is poor, it's not just idle temps that bad. Any type of load or moderate activity would be unsustainable with core temps hitting the thermal limit. You haven't mentioned that and it seems unlikely this is the issue.

The other and increasingly common obstacle is default CPU behavior. Recent CPUs are very reactive and quick to devote resources to trivial tasks. Other Ryzen 3000 owners have had difficulty with elevated idle power consumption and/or CPU activity. The commit percentage you reported is low, but I am not sure that tells the entire story. Certainly another way to assess that is to quit iCUE and it's associated services and monitor the effect in Ryzen Master or whatever else you are using. There could also be legitimate OS or BIOS features keeping the voltage and core temps up, but I am not a x570 owner and can't detail specific settings to be mindful of. However, since that does not seem to be in play on the CM model, the global settings are not likely to be the central issue.

The noise you will have be make your own evaluation of and whether it is sustainable for you. Don't be afraid to make your own fan curves and most people will not want to be on the three presets in Summer weather with elevated room temperatures. Strong pump noise is more difficult to live with.
Reply With Quote


Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 12:25 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.