The Corsair User Forums  

Go Back   The Corsair User Forums > Corsair Product Discussion > iCUE Software

Reply
 
Thread Tools Search this Thread Rate Thread Display Modes
  #1  
Old 11-09-2018, 12:59 AM
LBCdirtyD LBCdirtyD is offline
Registered User
LBCdirtyD's PC Specs
 
Join Date: Nov 2018
Posts: 4
POST ID # = 979582
LBCdirtyD Reputation: 10
Default ICUE showing different CPU temp than other monitoring apps

Hello All,

As per title recently got Corsair H115i Pro RGB and per ICUE when playing BF1 or BFV full load is about 15C degrees below what my other two temp monitoring apps Speccy and Driver Fusion. I have link to screenshot as you can see below. In ICUE the temp states 35C but in Speccy it is 49C and Driver Fusion is close to Speccy temps. Anyone else have same issues as why they so different?

Link to image

https://imgur.com/a/DYDlSx3
Reply With Quote
  #2  
Old 11-09-2018, 01:52 AM
DevBiker's Avatar
DevBiker DevBiker is offline
iCue βeta Team Green
DevBiker's PC Specs
 
Join Date: Feb 2017
Location: Republic of Texas
Posts: 5,150
POST ID # = 979587
DevBiker Reputation: 76
Default

You're comparing the temperature of the coolant to the CPU temperature. Different beasties. Go to the Dashboard and add the system to see the CPU temp that iCue "sees".
__________________
Please click "Edit System Specs" and fill out your system info.

This comment is provided "as-is", without warranty of any kind, express or implied, including, but not limited to, the sanity or mental fitness of the author. The author is not a Corsair employee, does not represent Corsair, and no comment should be construed as an official statement from Corsair.

Reply With Quote
  #3  
Old 11-10-2018, 01:07 AM
LBCdirtyD LBCdirtyD is offline
Registered User
LBCdirtyD's PC Specs
 
Join Date: Nov 2018
Posts: 4
POST ID # = 979721
LBCdirtyD Reputation: 10
Default

Quote:
Originally Posted by DevBiker View Post
You're comparing the temperature of the coolant to the CPU temperature. Different beasties. Go to the Dashboard and add the system to see the CPU temp that iCue "sees".
Oh great thanks for feedback I will try this. I didn’t know the temp was for the coolant in ICUE.
Reply With Quote
  #4  
Old 11-10-2018, 09:41 AM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 6,807
POST ID # = 979765
c-attack Reputation: 96
Default

Yes, two different bits of information. Your CPU temperature is largely influenced by voltage, less the amount of heat the cold plate (on any type of cooler) can physically conduct away. That heat is transmitted into the H115i PRO, which for you serves as the waste heat removal system. Think of it as a storage tank for heat. It can hold a lot, but if you let it fill up, then there are negative consequences to the CPU since heat flows both ways across the cold plate. The coolant temp is effectively the lowest possible CPU temperature. The actual CPU temp will be coolant temp + whatever voltage based heat you add through CPU activity/voltage load, etc. +1C rise in coolant temp results in a +1C rise in CPU temp and of course, the same is true for reducing temperatures. Since the fans can only move heat of the coolant stream and not cool the CPU directly, it becomes the most efficient and logical control variable for fan speed.

If you turn on Prime95 you can watch your CPU temps bump +30-50C in the next 1 second. That is your voltage based heat. That is all about CPU design, conductivity, and your Vcore, with the last one being the only thing you can change. In that moment, it doesn't matter if you fans are at 0 rpm or 4000 rpm, there will be no change to that instant temperature change. It's all conductive. However, if you let the test run for a few minutes, you will see the CPU package temp slowly climb up at of rate of maybe +1C ever minute or so. That, is the CPU coolant rise. Your conductivity and Vcore did not change, but more heat is being added to the cooler than it can release in a single pass. That is the number and value fan speed can effect. However, since your coolant temp rise is likely only to be +4-6C, there isn't a lot of reduction available. Larger surface area coolers like the 280mm H115i Pro don't need a lot of fan speed. You can keep things moderate to relaxed with a minimal impact on end CPU temperature.

The one monkey wrench in all of this is the effect of additional GPU heat on the rest of the case, including the radiator and cooling system. Since the baseline coolant temp is always going to be interior case temperature, a +5C rise in internal ambient temp will cause the coolant to go up +5C as well. You cannot reduce this with radiator fan speed alone, only by moving more GPU waste heat out of the case -- something much easier said than done. This is true with all cooling systems, but with the software based coolers you can actually see the change and it does impact your fan curve. Some people are surprised to see a higher coolant temp when gaming than a straight 100% CPU test. That is the additional case heat in play. So rather than set your curves based on a CPU stress tester, be sure and align them with whatever things you actually do. If that's gaming, figure out your typical peak coolant temperature when playing and set the fans to a speed you can live with. You can make them as quiet as you like at the desktop idle temp. On a 280mm, you should never need anything beyond 1000 rpm for most loads. I benchmark at 1200 rpm. I am still looking at wattage figures for the 9900K, but that was plenty of fan speed for my 215W draw X99 system. You can experiment and see, just remember that 1C reduction in coolant only yield -1C on the CPU.
Reply With Quote
  #5  
Old 11-12-2018, 10:45 PM
LBCdirtyD LBCdirtyD is offline
Registered User
LBCdirtyD's PC Specs
 
Join Date: Nov 2018
Posts: 4
POST ID # = 980160
LBCdirtyD Reputation: 10
Default

Quote:
Originally Posted by c-attack View Post
Yes, two different bits of information. Your CPU temperature is largely influenced by voltage, less the amount of heat the cold plate (on any type of cooler) can physically conduct away. That heat is transmitted into the H115i PRO, which for you serves as the waste heat removal system. Think of it as a storage tank for heat. It can hold a lot, but if you let it fill up, then there are negative consequences to the CPU since heat flows both ways across the cold plate. The coolant temp is effectively the lowest possible CPU temperature. The actual CPU temp will be coolant temp + whatever voltage based heat you add through CPU activity/voltage load, etc. +1C rise in coolant temp results in a +1C rise in CPU temp and of course, the same is true for reducing temperatures. Since the fans can only move heat of the coolant stream and not cool the CPU directly, it becomes the most efficient and logical control variable for fan speed.

If you turn on Prime95 you can watch your CPU temps bump +30-50C in the next 1 second. That is your voltage based heat. That is all about CPU design, conductivity, and your Vcore, with the last one being the only thing you can change. In that moment, it doesn't matter if you fans are at 0 rpm or 4000 rpm, there will be no change to that instant temperature change. It's all conductive. However, if you let the test run for a few minutes, you will see the CPU package temp slowly climb up at of rate of maybe +1C ever minute or so. That, is the CPU coolant rise. Your conductivity and Vcore did not change, but more heat is being added to the cooler than it can release in a single pass. That is the number and value fan speed can effect. However, since your coolant temp rise is likely only to be +4-6C, there isn't a lot of reduction available. Larger surface area coolers like the 280mm H115i Pro don't need a lot of fan speed. You can keep things moderate to relaxed with a minimal impact on end CPU temperature.

The one monkey wrench in all of this is the effect of additional GPU heat on the rest of the case, including the radiator and cooling system. Since the baseline coolant temp is always going to be interior case temperature, a +5C rise in internal ambient temp will cause the coolant to go up +5C as well. You cannot reduce this with radiator fan speed alone, only by moving more GPU waste heat out of the case -- something much easier said than done. This is true with all cooling systems, but with the software based coolers you can actually see the change and it does impact your fan curve. Some people are surprised to see a higher coolant temp when gaming than a straight 100% CPU test. That is the additional case heat in play. So rather than set your curves based on a CPU stress tester, be sure and align them with whatever things you actually do. If that's gaming, figure out your typical peak coolant temperature when playing and set the fans to a speed you can live with. You can make them as quiet as you like at the desktop idle temp. On a 280mm, you should never need anything beyond 1000 rpm for most loads. I benchmark at 1200 rpm. I am still looking at wattage figures for the 9900K, but that was plenty of fan speed for my 215W draw X99 system. You can experiment and see, just remember that 1C reduction in coolant only yield -1C on the CPU.
That is great information. Would you say its best to play with Custom Fan Curve in ICUE versus using the Balanced or Extreme Preset? I have been messing around with the basic ones while I play BFV now and really cannot tell a big difference on temps being much different. It could be the case I have is not the greatest at releasing the GPU temps out of the case maybe too though.
Reply With Quote
  #6  
Old 11-12-2018, 11:08 PM
DevBiker's Avatar
DevBiker DevBiker is offline
iCue βeta Team Green
DevBiker's PC Specs
 
Join Date: Feb 2017
Location: Republic of Texas
Posts: 5,150
POST ID # = 980161
DevBiker Reputation: 76
Default

Take a look at the Liquid Cooler FAQ. I took one of c-attacks explanations of the process to create a fan curve and just pasted it in there. It's pretty detailed and pretty awesome. It's at the end of section A.
__________________
Please click "Edit System Specs" and fill out your system info.

This comment is provided "as-is", without warranty of any kind, express or implied, including, but not limited to, the sanity or mental fitness of the author. The author is not a Corsair employee, does not represent Corsair, and no comment should be construed as an official statement from Corsair.

Reply With Quote
  #7  
Old 11-13-2018, 08:03 PM
LBCdirtyD LBCdirtyD is offline
Registered User
LBCdirtyD's PC Specs
 
Join Date: Nov 2018
Posts: 4
POST ID # = 980344
LBCdirtyD Reputation: 10
Default

I will look at that info on the fan curve. Thanks for your replies regarding all this.
Reply With Quote
  #8  
Old 11-13-2018, 08:27 PM
c-attack c-attack is offline
Registered User
c-attack's PC Specs
 
Join Date: Jun 2014
Posts: 6,807
POST ID # = 980347
c-attack Reputation: 96
Default

Yes, always make your own custom curve. The 3 included presets all have a standard room temp base value and are not unique for your specific environment. You may be above or below. They might all be fine in Winter, but intrusive in Summer with no additional cooling effect.
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 12:56 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2018, vBulletin Solutions, Inc.