Watercooling reality vs expectations

antok86

[H]F Junkie
Joined
Feb 26, 2006
Messages
8,252
Would i see any drastic improvment cooling a cpu with an AIO vs a custom loop? (240mm rads)
 

NoOther

Supreme [H]ardness
Joined
May 14, 2008
Messages
6,468
Would i see any drastic improvment cooling a cpu with an AIO vs a custom loop? (240mm rads)

Confused, are you asking if an AIO would be better than a custom loop? That would kind of defeat some of the purpose of going the custom route... You can definitely get better cooling using a custom loop, but it would largely depend on how you make the loop and what parts you are using.
 

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,710
Would i see any drastic improvment cooling a cpu with an AIO vs a custom loop? (240mm rads)

Cooling- you could. But the purpose behind building a custom loop should probably be explored. No consumer CPU, barring an overclocked Threadripper, would make use of more cooling capacity than a decent 240mm AIO would provide. Noise might be another factor, both in terms of fans (280mm might be quieter for the same cooling) and pumps.
 
Joined
Apr 5, 2016
Messages
2,286
If you're going custom loop, why restrict yourself to a 240mm? A good chunk of the advantage to a custom loop is the ability to cram the biggest possible radiator in every available hole and run your fans at inaudible levels - and STILL outperform an AIO in terms of cooling.
 

antok86

[H]F Junkie
Joined
Feb 26, 2006
Messages
8,252
If you're going custom loop, why restrict yourself to a 240mm? A good chunk of the advantage to a custom loop is the ability to cram the biggest possible radiator in every available hole and run your fans at inaudible levels - and STILL outperform an AIO in terms of cooling.
my case at max i think takes 240mm i would have to double check.

Cooling- you could. But the purpose behind building a custom loop should probably be explored. No consumer CPU, barring an overclocked Threadripper, would make use of more cooling capacity than a decent 240mm AIO would provide. Noise might be another factor, both in terms of fans (280mm might be quieter for the same cooling) and pumps.
thanks

Confused, are you asking if an AIO would be better than a custom loop? That would kind of defeat some of the purpose of going the custom route... You can definitely get better cooling using a custom loop, but it would largely depend on how you make the loop and what parts you are using.
im asking if i were to do a custom loop to cool a cpu would i see any benefits over the AIO equivalent. i am new to watercooling to it would be soft tubing and res/pump combo. again to start it would just be cpu cooling
 

jmilcher

Supreme [H]ardness
Joined
Feb 3, 2008
Messages
4,971
In the end once the loop saturates temp wise and reaches equilibrium, you are going to have a constant temp. A custom loop would yield a lower average temp.
 

Spartacus09

[H]ard|Gawd
Joined
Apr 21, 2018
Messages
1,809
In the end once the loop saturates temp wise and reaches equilibrium, you are going to have a constant temp. A custom loop would yield a lower average temp.
Not just average temp, but often peak temp as well.
Depending on the AIO model/pump custom loop pumps are closer or above the optimal 1GPM flow rate.
Although the AIOs have gotten alot better in recent years, I couldn't find any specifics other than the H100i which was estimated to be about 0.5 GPM.
 

techguymaxc

Limp Gawd
Joined
Jul 6, 2016
Messages
139
Cooling- you could. But the purpose behind building a custom loop should probably be explored. No consumer CPU, barring an overclocked Threadripper, would make use of more cooling capacity than a decent 240mm AIO would provide. Noise might be another factor, both in terms of fans (280mm might be quieter for the same cooling) and pumps.

My de-lidded 7960x at a mere 4GHz and just over 1.0V can reach temps of almost 80C on the hottest core on a custom loop with a 360 rad when running Handbrake batch encodes. At that clockspeed and with that voltage some folks wouldn't even call that an overclock.

Even my de-lidded 7700k @ 5.2GHz can hit similar temps during stress testing, also on a custom loop with a 360 rad.

No one's saying a custom loop is a requirement for normal operation, but I can guarantee you an AIO will throttle in the above situations. I used an H100 with a 3770k back in the day, could only hit 4.7GHz and temps hit TjMax during stress testing. After de-lidding and swappintg to a custom loop, temps dropped by 30C+ and I could also hit 4.9GHz. There is a dramatic difference in the cooling capacity of AIO coolers and a custom loop which has been properly spec'd for the workload.
 

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,710
That sounds like a golden 9900k, what stress test? What voltage? How long was it run for? What AIO?

Intel XTU CPU and Memory, quick 5-minute sanity check, MCE on in ASRock BIOS (Z370 board), Corsair H115i with one default fan running, auto voltage (I'd have to check). I have a Z390 board to put in there and I'll test it further, but it already looks to be a much better sample than the 8700K it's replacing. That one is going into the Linux fileserver for KVM duties.
 

Spartacus09

[H]ard|Gawd
Joined
Apr 21, 2018
Messages
1,809
Intel XTU CPU and Memory, quick 5-minute sanity check, MCE on in ASRock BIOS (Z370 board), Corsair H115i with one default fan running, auto voltage (I'd have to check). I have a Z390 board to put in there and I'll test it further, but it already looks to be a much better sample than the 8700K it's replacing. That one is going into the Linux fileserver for KVM duties.
XTU is a good software overclocking tool but not great stress tester my 9900k tests at about 15c lower with XTU than intel burn test, AIDA64 Extreme is a much better for testing real world usage.

Highest stability and top end heat generation will be revealed by Intel burn test (very high/extreme) or prime95 (2 hour blend).
 

SvenBent

2[H]4U
Joined
Sep 13, 2008
Messages
3,271
Intel XTU CPU and Memory, quick 5-minute sanity check, MCE on in ASRock BIOS (Z370 board), Corsair H115i with one default fan running, auto voltage (I'd have to check). I have a Z390 board to put in there and I'll test it further, but it already looks to be a much better sample than the 8700K it's replacing. That one is going into the Linux fileserver for KVM duties.

using weak stress etst not really prove there is no neede for more cooling
Intel XTU is pretty weak as a CPU stress test


--- one of the stress test tool testing i did way back ---

Code:
Core I7 2700K           4.4GHz +0.030v                  4.4GHz +0.20v                   4.5GHz +0.20v                   4.6GHz +0.20v                   4.7 +0.20v
                        Status  temp    Power           Status  temp    Power           Status  temp    Power           Status  temp    Power
Asus RealBench          OK      79-84   318w*           OK      80-85   318w*           OK      80-87   303-319/326*    C  0:22 76-83   xxx/325*        C 0:00  XX-XX   xxx/315*
Aida64 Extreme          OK      74-81   181w            OK      73-79   176-180w        OK 7:40 74-81   180-185/187     OK 4:50 76-82   182-186/190     C 0:15  73-82   xxx/193

Core Damage 0.8h                                        OK      76-82   180w            OK15:28 79-85   184/188         OK 9:50 79-85   186/190         C 1:16  79-85   xxx/192
Cpu Burn-in 1.01                                                                        OK 4:05 73-77   181/184         OK 8:15 81-77   185/187         C 0:06  73-78   xxx/190

Folding@Home            OK      73-78   182w            OK      73-79   179w            OK 9:30 73-79   180/182         C  2:49 74-81   xxx/186         C 0:00  31-31   xxx/185

HeavyLoad               OK      72-76   184w            OK      72-76   182w            OK 4:20 73-78   187/189         C  2:59 74-80   xxx/193         C 0:00  xx-xx   xxx/193
Handbrake 1.0.2         OK      71-76   165-177w        OK      72-76   96-180          OK 8:08 73-77   95-180/185      OK 6:30 73-80   160-182/186     C 0:29  72-76   xxx/186

IBT 2.54 Standard                                       OK      78-88   167-211w        OK 4:00 80-90   171-215/216     C  0:02 71-79   xxx/218         C 0:00  xx-xx   xxx/215
IBT 2.54 Maximum                                        OK      80-91   106-212w        C  1:19 83-94   xxx/220         C  0:02 72-82   xxx/217         C 0:01  63-68   xxx/206

LinPack V11.2.2.010     OK      78-90   191-210w        OK      78-89   162-207w        OK 4:15 80-91   161-211/213     c  0:05 78-88   xxx/218         C 0:00  xx-xx   xxx/210
LinPack V11.2.2.010 x2  OK      84-96   220w            OK      81-92   169-216w        C  0:55 82-94   xxx/223         c  0:01 73-76   xxx/223         C 0:00  xx-xx   xxx/182

OCCT 4.4.3 (Own/large)  OK      78-88   211w            OK      76-84   185-204w        OK 4:00 77-87   198-205/211     C  0:06 73-81   XXX/212         C 0:00  31-31   xxx/193
OCCT 4.4.3 (Linpack90%) OK      82-93   207-214                                         OK 4:05 74-83   191-196/199     C  0:48 76-85   xxx/204         C 0:02  37-53   xxx/199

Prime95 26.6 (small)                                                                    OK 4:05 77-85   195/198         C  0:12 76-85   xxx/201         C 0:00  xx-xx   xxx/195

PRime95 28.5 (8-8-30)   OK      84-93   217w                                            C 0:40  82-93   xxx/218         C  0:00 68-76   xxx/215         C 0:00  xx-xx   xxx/213
Prime95 28.5 (small)    OK      79-89   212w            OK      80-90   211w            C 1:52          xxx/219         C  0:00 63-72   xxx/214         C 0:00  xx-xx   xxx/215
Prime95 28.5 (Blend     OK      82-92   215w            OK      84-94   201w            C 2:26  83-93   xxx/226         C  0:03 67-71   XXX/196         C 0:00  36-36   xxx/194

Passmark BurnInTest 8.1 OK      67-72   104-180         OK      67-72   103-178w        OK      69-73   105-179w/183    OK 6:35 69-74   108-175/182     OK 4:25 70-74   109-183/187
XTU (CPU)               OK      73-79   176-190w                                        OK 7:25 73-79   173-186/188     C  2:17 73-81   xxx/210         C  0:04 67-73   xxx/191

Winrar                  OK      60-67   151w            OK      60-67   150w            OK      61-67   15-/155/177     OK11:00 62-69   155/156         C  1:20 61-67   xxx/165
7-zip                   OK      68-71   124-173         OK      68-72   126-171w        OK 9:20 70-74   126-175/178     OK 9:40 71-74   128-177/179     OK 6:40 71-76   134-179/181


* This uses OpenCL to put load on GPU as well.
  Not direct comparable against other candidates

Status key
OK = Did not find any errors
C  = Crash/BSOD
R  = Reboot/Reset
D  = Error detected


Really wish more forums would support a fixed with font thar support tabs


bottom line is that intel xtu will happily run over 7 hours without failure on a system that crashes at around 1hours makr in prim95

or even several hours and then crash on a system that crashe winthin a few mins with prime95
but it does seem to appear to beat out aida 64

-- edit --
converted tab adjusted text to space adjusted so it was readable on this forum
To received an OK mark the system had to be running for 4 hours with the tested CPU stress software

PWR draw is indicationg what it was showing at the laste few mins / peak draw during the entire test time.
for system crashing only peak pwr draw is shown

Temps is the measured peak temps over the cores during the stretch of the test
So 76-82 maens the lowets peak temp for a core was 76 and the highst 82
measure with realtemp

I've run these kind of testing a couple of times over the years. and all the times prime95 and linpack based stresstesting came out on top
 
Last edited:

IdiotInCharge

NVIDIA SHILL
Joined
Jun 13, 2003
Messages
14,710
Agreed on XTU- that's a quick 'gut check' and helps verify other readouts. I plan to find the ceiling for the 9900K including any needed AVX offset.
 

jmilcher

Supreme [H]ardness
Joined
Feb 3, 2008
Messages
4,971
Not just average temp, but often peak temp as well.
Depending on the AIO model/pump custom loop pumps are closer or above the optimal 1GPM flow rate.
Although the AIOs have gotten alot better in recent years, I couldn't find any specifics other than the H100i which was estimated to be about 0.5 GPM.
Agreed.

But a lot also depends on the amount of heat you’re putting into the loop.

Lots of variables, but it was a generalization.
 
Top