Overclock.net › Forums › Cooling › Water Cooling › AIOs Why they should only be considered a short term product.
New Posts  All Forums:Forum Nav:

AIOs Why they should only be considered a short term product. - Page 13

post #121 of 366
Quote:
Originally Posted by Blameless View Post

My Nepton 280L (which was on sale and has quite a strong pump) is my guinea pig; first AIO I've used in a personal build for any length of time.

About 12k hours on the pump now with no noticeable degradation in performance. I'll see how it fares after another 20k hours.
I have some top-end heatsinks and none of them really rival the 280L I'm using in cooling performance, even when I use the same fans. At low noise levels, there isn't a terribly large difference, but as air flow increases, my heatsinks fall further and further behind.

Same pattern with all of my Intel hex core parts going back to the LGA-1366 days, as well as my AMD FX octo-core parts. Past about 200-250w of CPU heat production, getting heat to the fins becomes the limiting factor more than surface area or air flow, and this AIO is proving to be closer to my custom loops than my air coolers.
Nepton 280L pump is a little better, but the radiator is still limiting things.

At a guess your air cooling setups were / are not optimized to supply cool air to the air cooler. That is most often the reason. Reviewers rarely give cooler intake air temp, instead using room temp .. and test in a case saying it's real world. What they are doing is testing different coolers to see how well their system performs, not how well the cooler perform. The only real world is there own because none of us have the same exact system.
post #122 of 366
Quote:
Originally Posted by superkyle1721 View Post

Great data I will add one caveat though. The majority of the noise from the system is not the pump but rather the unnecessary noise the fans produce. The stock fans on the CLCs are garbage which unless you have them around like I did is an added expense that needs to be taken into account. Switching fans did not reduce temps however but it did reduce the noise 10 fold.


Always destroying exergy
How about we slow down your fans to the same noise level as normal air coolers make and compare several coolers. ciarlatano has done this and here is what he found

Feel free to got to HiTech Legion and look at the data in his reviews.
post #123 of 366
Quote:
Originally Posted by doyll View Post

Quote:
Originally Posted by Blameless View Post

My Nepton 280L (which was on sale and has quite a strong pump) is my guinea pig; first AIO I've used in a personal build for any length of time.

About 12k hours on the pump now with no noticeable degradation in performance. I'll see how it fares after another 20k hours.
I have some top-end heatsinks and none of them really rival the 280L I'm using in cooling performance, even when I use the same fans. At low noise levels, there isn't a terribly large difference, but as air flow increases, my heatsinks fall further and further behind.

Same pattern with all of my Intel hex core parts going back to the LGA-1366 days, as well as my AMD FX octo-core parts. Past about 200-250w of CPU heat production, getting heat to the fins becomes the limiting factor more than surface area or air flow, and this AIO is proving to be closer to my custom loops than my air coolers.
Nepton 280L pump is a little better, but the radiator is still limiting things.

At a guess your air cooling setups were / are not optimized to supply cool air to the air cooler. That is most often the reason. Reviewers rarely give cooler intake air temp, instead using room temp .. and test in a case saying it's real world. What they are doing is testing different coolers to see how well their system performs, not how well the cooler perform. The only real world is there own because none of us have the same exact system.
Very true! This is my first CLC build. Previous build were either air or custom loop. My issue with air coolers (not that everyone will have the issue) is when running sli is the amount of heat being dumped into the case by the GPU fans inevitably increase the case temp. The air flow needed to compensate and remove this air fast enough to not make a difference in the air temp entering the air coolers radiator caused a large increase in system noise. Certainly some cases allow for better airflow than others but the location of the CPU in my air 540 made this very difficult to achieve. I actually never tried air cooling this case since visually I couldn't make it work to allow for proper air flow that would provide similar temps as a CLC being used as an intake would provide.


Always destroying exergy
post #124 of 366
Quote:
Originally Posted by superkyle1721 View Post

Very true! This is my first CLC build. Previous build were either air or custom loop. My issue with air coolers (not that everyone will have the issue) is when running sli is the amount of heat being dumped into the case by the GPU fans inevitably increase the case temp. The air flow needed to compensate and remove this air fast enough to not make a difference in the air temp entering the air coolers radiator caused a large increase in system noise. Certainly some cases allow for better airflow than others but the location of the CPU in my air 540 made this very difficult to achieve. I actually never tried air cooling this case since visually I couldn't make it work to allow for proper air flow that would provide similar temps as a CLC being used as an intake would provide.


Always destroying exergy
Running multiple GPUs is a cooling nightmare to the 10th power. It's not that air cooling can't do the job, it's that GPU coolers are the worst possible designs to be using. The have nothing in their design that separates their hot exhaust air from the cool air needed to their intakes. Only the reference blower coolers do this, and even they leak lots of hot air out inside the case.

There are few options, and motherboard PCIe socket placement is partly to blame, but if GPU cooler design is the biggest problem.
post #125 of 366
Quote:
Originally Posted by doyll View Post

One of the best is this German test of Sivler Arrow SB-E , Extreme and H100

The H100 is an early gen CLC with a relatively weak pump, narrow tubing, and not a particularly good block for the flow it's given.
Quote:
Originally Posted by doyll View Post

There is all kinds of accurate data showing air doing a as good and better than CLCs. Most of the users who find CLCs give them better cooling than a good air cooler do not have system airflow set up properly and are supplying their cooler with much hotter air then there CLC replacement gets.
Quote:
Originally Posted by doyll View Post

Nepton 280L pump is a little better, but the radiator is still limiting things.

At a guess your air cooling setups were / are not optimized to supply cool air to the air cooler. That is most often the reason.

I have found it easier to get cold intake air to a radiator, that can be positioned anywhere, than with a heatsink, which necessarily must be in the immediate vicinity of the CPU.

That said, a decent flow through a decent block (which some CLCs have) will pull heat from a CPU faster than any collection of heatpipes on an mass produced air cooler to date. With the same higher speed fans, my 280L beats my NH-D15 by about 3-5C, and D14 or my old Silver Arrow (the original with 4x8mm pipes) by another degree or two more, even on an open air bench with identical intake temperatures, when moving ~220w of heat.

My systems are not silent, and air flow/fin area are generally not the limiting factors. Even with the 280L, I run into coolant flow limitations sooner than anything else. I can cut air flow through the rad in half and notice less of an increase in temperatures than if I reduce pump speed by 10%.
Quote:
Originally Posted by doyll View Post

Reviewers rarely give cooler intake air temp, instead using room temp .. and test in a case saying it's real world. What they are doing is testing different coolers to see how well their system performs, not how well the cooler perform. The only real world is there own because none of us have the same exact system.

The problem I have with most cooler reviews is that nine times in ten they are testing on Intel quad cores, which often don't make that much heat.

I don't have any desktop quad cores any more. I have an i7 970 (air), an i7 4930k (air), an i7 5820k (CLC), and an FX 9590 (custom loop)...all of which will easily pass 200w at max load. I don't care how well something cools an Intel quad because I haven't found it worth while to put more than 30-40 dollars of cooling on an Intel quad since my last Bloomfield part.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #126 of 366
Quote:
Originally Posted by Blameless View Post

The H100 is an early gen CLC with a relatively weak pump, narrow tubing, and not a particularly good block for the flow it's given.

I have found it easier to get cold intake air to a radiator, that can be positioned anywhere, than with a heatsink, which necessarily must be in the immediate vicinity of the CPU.

That said, a decent flow through a decent block (which some CLCs have) will pull heat from a CPU faster than any collection of heatpipes on an mass produced air cooler to date. With the same higher speed fans, my 280L beats my NH-D15 by about 3-5C, and D14 or my old Silver Arrow (the original with 4x8mm pipes) by another degree or two more, even on an open air bench with identical intake temperatures, when moving ~220w of heat.

My systems are not silent, and air flow/fin area are generally not the limiting factors. Even with the 280L, I run into coolant flow limitations sooner than anything else. I can cut air flow through the rad in half and notice less of an increase in temperatures than if I reduce pump speed by 10%.
The problem I have with most cooler reviews is that nine times in ten they are testing on Intel quad cores, which often don't make that much heat.

I don't have any desktop quad cores any more. I have an i7 970 (air), an i7 4930k (air), an i7 5820k (CLC), and an FX 9590 (custom loop)...all of which will easily pass 200w at max load. I don't care how well something cools an Intel quad because I haven't found it worth while to put more than 30-40 dollars of cooling on an Intel quad since my last Bloomfield part.
While I haven't dissected any pumps, I'm told the od H100 pump and new CLC pumps are near identical.

Very few CLCs can out-perform air in a fair test. CLC pumps simply do not flow enough coolant. Air cooling react almost 1:1 to CPU temp verus cooler intake air temp. Water cooling does not.

No idea what high airflow fans you used on air coolers to match CLC fan noise, so I can't comment on that.

Reduction of pump speed reduces coolant flow .. not a good thing when pump cannot really move enough coolant even at full speed if CPU is really making a lot of heat.

Intel quads can make lots of heat, some even more than hex. My 920 @ 4.4GHz makes more heat than 980X does. I have seen AMD being hard to get good contact and therefor good heat transfer to block or cooler.

But are hi-jacking this thread .. even worse than it's already been hi-jacked. blushsmiley.gif
post #127 of 366
Quote:
Originally Posted by Aaron_Henderson View Post

Well...just because you can't personally see a use for them. doesn't mean there isn't one...for example, I'll mention one I can think of right off the top of my head...I was recently asked to build a cheap gaming PC for a < 12 year old kid to play Minecraft on. All he wanted was something that play Minecraft without "lagging", and that it "looked cool" and as much like my most recently sold off water-cooled PC as it could. I was given a budget of $400-600 for the ENTIRE computer...a custom loop would not have fit into the budget without making other compromises, noise is not an issue, they would be doing ZERO maintenance on the machine...the only cost efficient solution to including some type of trouble free water cooling into the build was a CLC...sure, I could have built a crappier PC, used the money saved for a custom loop, and then been on the hook for maintenance from there on out. Anyway...again...stop calling people stupid and such because they don't see things your way. That could go both ways...not sharing an opinion with someone does not indicate a lack of intelligence...I am sorry.

Sorry, next time I will add the stipulation "unless you are a pre-teen playing Minecraft". rolleyes.gif
post #128 of 366
Friends -

There are many, many threads where we can gather to bash on and/or defend CLCs in relationship to other cooling solutions, but this is not one.

This is a thread about the coolant in CLCs and the effects of high heat over extended periods of time on that coolant and the CLCs respective parts.

It has nothing to do with price/performance, noise/performance, air-cooling vs CLCs, CLCs vs custom loops, etc.

(It also has nothing to do with how many CLCs you've owned, how many are alive, how many are dead, etc - unless you specifically measured the heat of the coolant in the CLC and maintained that heat over time and can draw a demonstrable/replicable conclusion from that experience.)

Have a nice day thumb.gif
finally quiet
(20 items)
 
Peggy
(11 items)
 
Betty - WIP
(17 items)
 
CPUMotherboardGraphicsRAM
i5-760 3.8 + turbo MSI P55-GD80 MSI GTX 970 Golden Edition 16GB Tactical LP 1600 c8 
Hard DriveHard DriveHard DriveCooling
Samsung 840 Pro 128GB OCZ Agility 60GB Spinpoint F3 1TB Noctua NH-U12P SE2 
CoolingOSMonitorMonitor
2x NF-P12, CPU OS X 10.10, 8.1 Ultimate Asus VW246H Ergotron Neo-Flex 
KeyboardPowerCaseMouse
CM Quick Fire Rapid EVGA Supernova G2 850W Silverstone FT-02W CM Spawn 
AudioAudioAudio
Sony TA-AX380 Sennheiser HD 280 KLH 911B 
CPUMotherboardGraphicsRAM
i3-2125 GA-H61N USB3 HD3000 8GB Samsung Miracle RAM 
Hard DriveHard DriveHard DriveCooling
Kingston 60GB 3ware/LSI 9750-4i Hitachi 5k3000 2TB x4 RAID 5 Intel for now 
OSPowerCase
OS X 10.8.2 Corsiar CX430M Fractal Design Node 304 
CPUMotherboardGraphicsRAM
i3-2100 ASUS P8H67-M PRO/CSM ASUS Passive GT 440 4GB Dominator 1600 c8 
Hard DriveOptical DriveCoolingCooling
Hitachi 120GB 2.5" LG BR-RW CM GeminII S + NF-P12 2x NF-R8 - Intake 
CoolingOSMonitorKeyboard
NF-B9 Windows 7 Sony KDL-46NX700 MK700 
PowerCaseMouseAudio
Antec EA-650 Zalman HD160XT, Modded MK700 Yamaha RX-V373 
  hide details  
Reply
finally quiet
(20 items)
 
Peggy
(11 items)
 
Betty - WIP
(17 items)
 
CPUMotherboardGraphicsRAM
i5-760 3.8 + turbo MSI P55-GD80 MSI GTX 970 Golden Edition 16GB Tactical LP 1600 c8 
Hard DriveHard DriveHard DriveCooling
Samsung 840 Pro 128GB OCZ Agility 60GB Spinpoint F3 1TB Noctua NH-U12P SE2 
CoolingOSMonitorMonitor
2x NF-P12, CPU OS X 10.10, 8.1 Ultimate Asus VW246H Ergotron Neo-Flex 
KeyboardPowerCaseMouse
CM Quick Fire Rapid EVGA Supernova G2 850W Silverstone FT-02W CM Spawn 
AudioAudioAudio
Sony TA-AX380 Sennheiser HD 280 KLH 911B 
CPUMotherboardGraphicsRAM
i3-2125 GA-H61N USB3 HD3000 8GB Samsung Miracle RAM 
Hard DriveHard DriveHard DriveCooling
Kingston 60GB 3ware/LSI 9750-4i Hitachi 5k3000 2TB x4 RAID 5 Intel for now 
OSPowerCase
OS X 10.8.2 Corsiar CX430M Fractal Design Node 304 
CPUMotherboardGraphicsRAM
i3-2100 ASUS P8H67-M PRO/CSM ASUS Passive GT 440 4GB Dominator 1600 c8 
Hard DriveOptical DriveCoolingCooling
Hitachi 120GB 2.5" LG BR-RW CM GeminII S + NF-P12 2x NF-R8 - Intake 
CoolingOSMonitorKeyboard
NF-B9 Windows 7 Sony KDL-46NX700 MK700 
PowerCaseMouseAudio
Antec EA-650 Zalman HD160XT, Modded MK700 Yamaha RX-V373 
  hide details  
Reply
post #129 of 366
Quote:
Originally Posted by ciarlatano View Post

Sorry, next time I will add the stipulation "unless you are a pre-teen playing Minecraft". rolleyes.gif

Good one rolleyes.gif Or you could be a little more open minded...saying CLC have NO purpose at all is nonsense, and you know it. Just because you don't like something, and can't personally see the need...and what would make you superior in any way to the kid playing Minecraft? I am sure making snarky comments on OCN makes you much better than everyone...anyway, enjoy the OCN news section, have at each other...
post #130 of 366
Quote:
Originally Posted by ciarlatano View Post

Quote:
Originally Posted by Aaron_Henderson View Post

Well...just because you can't personally see a use for them. doesn't mean there isn't one...for example, I'll mention one I can think of right off the top of my head...I was recently asked to build a cheap gaming PC for a < 12 year old kid to play Minecraft on. All he wanted was something that play Minecraft without "lagging", and that it "looked cool" and as much like my most recently sold off water-cooled PC as it could. I was given a budget of $400-600 for the ENTIRE computer...a custom loop would not have fit into the budget without making other compromises, noise is not an issue, they would be doing ZERO maintenance on the machine...the only cost efficient solution to including some type of trouble free water cooling into the build was a CLC...sure, I could have built a crappier PC, used the money saved for a custom loop, and then been on the hook for maintenance from there on out. Anyway...again...stop calling people stupid and such because they don't see things your way. That could go both ways...not sharing an opinion with someone does not indicate a lack of intelligence...I am sorry.

Sorry, next time I will add the stipulation "unless you are a pre-teen playing Minecraft". rolleyes.gif

This is what I was referring to. Simply no need for this kind of reply. All it does is incite venom and vitriol in this community. eh-smiley.gif

And tbqh CLCs aren't just for kids who play Minecraft.

~Ceadder smil3dbd4e4c2e742.gif
 
CPUMotherboardGraphicsGraphics
AMD Phenom II X6 1100T Crosshair IV Formula Sapphire Radeon HD 6870 Sapphire Radeon HD 6870 
RAMRAMRAMHard Drive
Corsair Dominator G. Skill Ripjaws X G. Skill x2 HITACHI 1TB Deskstars 1TB RAID0 
Optical DriveCoolingCoolingCooling
ASUS DRW-24B1ST  Swiftech 6950/6870 GPU Heatsink Swiftech 6950/6870 GPU Heatsink Monsoon 1/2"x5/8" PETG White Hardline tubing. 
CoolingCoolingCoolingCooling
Monsoon Economy Hardline fittings EK Dual DDC V2 top Swiftech Pump DDC-1T (2) Liquid Fusion V Double helix Reservoir 
CoolingCoolingCoolingCooling
Yate Loon 120x20 (D12SM-12C) Medium Speed Silen... EK Supreme HF Classified EK Thermosphere GPU Block EK Thermosphere GPU Block 
OSMonitorKeyboardPower
Win7 64 bit Ultimate OEM Asus VH222H Black HDMI Widescreen Logitech the OCZ went buhbye adios. Don't know ... Corsair CMPSU HX 850w Modular 
CaseMouseMouse PadAudio
HAF 932 Logitech g9x COOLER MASTER Weapon of Choice: M4 Duramesh Pad Logitech X-240 2.1 speakers 
OtherOther
Sunbeam Rheosmart 3 Creative X-Fi 
  hide details  
Reply
 
CPUMotherboardGraphicsGraphics
AMD Phenom II X6 1100T Crosshair IV Formula Sapphire Radeon HD 6870 Sapphire Radeon HD 6870 
RAMRAMRAMHard Drive
Corsair Dominator G. Skill Ripjaws X G. Skill x2 HITACHI 1TB Deskstars 1TB RAID0 
Optical DriveCoolingCoolingCooling
ASUS DRW-24B1ST  Swiftech 6950/6870 GPU Heatsink Swiftech 6950/6870 GPU Heatsink Monsoon 1/2"x5/8" PETG White Hardline tubing. 
CoolingCoolingCoolingCooling
Monsoon Economy Hardline fittings EK Dual DDC V2 top Swiftech Pump DDC-1T (2) Liquid Fusion V Double helix Reservoir 
CoolingCoolingCoolingCooling
Yate Loon 120x20 (D12SM-12C) Medium Speed Silen... EK Supreme HF Classified EK Thermosphere GPU Block EK Thermosphere GPU Block 
OSMonitorKeyboardPower
Win7 64 bit Ultimate OEM Asus VH222H Black HDMI Widescreen Logitech the OCZ went buhbye adios. Don't know ... Corsair CMPSU HX 850w Modular 
CaseMouseMouse PadAudio
HAF 932 Logitech g9x COOLER MASTER Weapon of Choice: M4 Duramesh Pad Logitech X-240 2.1 speakers 
OtherOther
Sunbeam Rheosmart 3 Creative X-Fi 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › AIOs Why they should only be considered a short term product.