Overclock.net › Forums › Cooling › Air Cooling › LAWNoob's Air Cooler VS. AIO Water Cooler Review
New Posts  All Forums:Forum Nav:

LAWNoob's Air Cooler VS. AIO Water Cooler Review

post #1 of 8
Thread Starter 
I have always wondered if the CPU cooler reviews are legit or not. How can one cooler beat the NH-D14 in one site and lose to a H60 in another?

Have you ever wondered how well that AIO cooler can really perform if the reviewer added another fan?
Do you want to know how well the coolers can perform when they are using the same fans at the same speed but have a hard time finding such reviews?
Do you always complain about how the reviewer is not overclocking hard enough to push those cheap coolers off the edge?


I am here today to present you an air cooler / AIO review and it will be done the way I wanted a review to be done.
I will be changing my setup and getting more coolers soon, think of this as a preview.
Once I get my new setup, I will redo the tests again to ensure accuracy.
The Setup (Click to show)
Ambient temperature @25C
XSPC K2 TIM
ASRock 970A Extreme3 mobo
FX6100 CPU > 65C Max (stock setting: 3.3Ghz @1.22V)
1 stock of Corsair XM3 DDR3 RAM at 9-9-9-24 1333 1.5V
Toshiba laptop harddrive
OCZ Stealth Stream 2 500W PSU
3 XSPC 1650RPM fans
Zalman CNPS10X Performa
Corsair H50
Corsair H60 (2011)

For this review, I have a 120mm fan blowing down on the VRMs of my motherboard because the tiny heatsink is no match for 180W+ of overclocking awesomeness.
I have all power saving features off, overclocking by only increasing Vcore and CPU muti.
Estimated TDPs of FX6100:
141W at 4.0Ghz, 1.35V (set to 1.30V in BIOS)
191W at 4.4Ghz, 1.50V (set to 1.45V in BIOS)

Disclaimer:
  • Your own results may vary. FX6100 DOES NOT need 1.35V at 4Ghz, nor 1.5V at 4.4GHz. I'm only using those settings to make my life easier.
  • Since I am short on time and I will be changing hardware soon, this time I'm only going to test coolers with dual XSPC fans running at 12V straight from the PSU. I'll definately include stock fan and low speed fan performance the next time around.

AMD Stock HSF (Click to show)
Meet the stock AMD HSF. It is tiny, pathetic, and gets real loud for cooling a CPU at stock speeds, that's why my friend calls it "the lawn mower".



The base have a small round copper heat pipe, the copper was flat until I scratched it somehow.

The mounting is quite simple, even a blind hippo can put that thing on right the first time.
All you need to do is put the two rings under the thing that sticks out of that black bar thing near you CPU socket, then locate the blue lever and slam it down hard.

Here are the results of the stock HSF cooling a FX6100 @ 4.0Ghz at 1.3V set in BIOS


The stock HSF is not so bad after all, but keep in mind that 3800RPM fans is driving me mad.
Since the CPU core temperature is getting close to the danger zone, I decide to leave the stock HSF alone for now.

My rating for the stock AMD HSF: terri-bad --
Zalman CNPS10X Performa (Click to show)
This is an alright mid-ranged 5-heatpipe air cooler, in fact I think it is the best cooling performance I can get out of a sub $40 cooler (used ones doesn't count).
It's big enough to cool well (better than the 212 Evo), yet still small enough to not interfere with your RAM.







As you can see, the cooler itself will not interfere with any RAM for the AM3 platform. The base is not polished, but at this price range you can't ask for much more. The cooler comes with an 2000RPM fan with an low noise adapter.
Moounting (Click to show)
Mounting is quite simple if you are doing it right.

First, you take off the 2 black bars beside the CPU sicket by unscrewing it. Then you take the backplate provided, insert the 4 bolts into the parallel bars (the curved edges should be facing you!). After that, push the black plastic bolt covers into the backplate fully.

Now prepare the heatsink by finding the two straight chrome bars, loosen the bottom of the heatsink and slide the two bars in. Once it's in the right position (should slightly curve downwards), tighten them.

Finally, apply TIM on your CPU, find a hard and flat surface, stick your backplate to the back of your mobo the right way, put your big and shiny heatsink on top of your CPU, and screw the screws into the bolts in a 8 pattern.

It shouldn't take you more than 15 minutes to do it yourself.

The results:


It looks like the Zalman is a total overkill for a 140W CPU.


Even at 190W, the Zalman still have a little room for further overclocking. Or perhaps you could pair it with slower fans? Anyways, dual 1600-ish RPM fans sounds loud in a quiet living room, and even louder considering that it is sitting 20cm away from my face.

My rating for Zalman CNPS10X Performa: Best Value @Sub $40 with More-Than-Alright Cooling!
Corsair H50 (Click to show)
This is an entry level All-In-One water cooling kit. I do not have high hopes for it, partly because it looked so bad in all the other reviews. This should cost less than $50 right now.
The main benifits of the H50 over the Zalman is that the H50 can be used in small form factor PCs, it will be even easier to install, and it will not add too much weight onto your motherboard.





The water block's cold plate is still not polished. The tube is sort of flexible. The cooler comes with a 1700RPM fan.

The mounting is quite simple. First, you insert the chrome bolts into the rubberish AMD backplate. Then, you find the AMD plastic inserrts and put that into the AMD bracket. After that, all you need to do it to put TIM on your CPU, screw the screws through the bracket and into the backplate, and you are done!

Since 4Ghz is too easy for the Zalman, I'm going to test my H50 at 4.4Ghz right away.


Surprise surprise, the 3-year-old entry-level all-in-one closed-loop-cooler manages to beat the value king by 5C! The H50's pump noises cannot be heard when the pump runs on the 12V from PSU with those 1650RPM fans.

My rating for Corsair H50: Better Performance and More Form-Factor Possibilities for $50-ish!
Corsair H60 (Click to show)
This All-In-One kit is still entry level, but it is supposed to be better than the H50 by a little. This kit can be had for $50-ish from newegg.ca.
The main benifits of the H60 over the Zalman is that the H60 can be used in small form factor PCs, it will be even easier to install, and it will not add too much weight onto your motherboard.





The water block's cold plate is still not polished. The tube is sort of flexible. The cooler comes with a 1700RPM fan.

The mounting is even easier than the H50. You no longer have to remove the stock backplates and the black pastic bars around the AM3 socket. All you need to do is to install the correct bracket onto the water black. After that, slide the ring side of the two black screws into the two latches on the black bars, then slide brackets (with the water block) through the screws, then take the screw tops and screw them down.

Here are some 4.4Ghz results.


Noticed how the H60 performed slightly worse than the H50? I'm not sure why, but my fans are running a little slower when I'm testing the H60 (still at 12V from PSU though).

The H60's pump makes a whinny noise when I run it on the 12V rail of my PSU. I think it is the sound of the water bubbles, so I try to move the radiator up and down but the noise remains. It's probably not Corsair's fault since I got my H60 second hand. I'll see if the problem fixes itself the next time I use it.

My rating for Corsair H60: An H50 with Easier Mounting and A Smaller and Lighter Block!

Temperature results:


Conclusion:
Any cooler over $20 will be miles better than the stock cooler for sure. Even if you are not going to overclock as all, buying an aftermarket cooler would help a lot to reduce the noise levels. The Zalman CNPS10X Performa can be had for $30 and it does a great job cooling a 190W CPU, and that's why if you want something cheap and won't ever leak, go for the Zalman. However, if you want something that can fit into smaller form factor than a mid tower ATX, then the AIO coolers are your best bet. The Corsair units not only help your CPU to stay cooler than the Zalman, but they also have more straight forward mounting mechanisms.

That's all I have for you guys today, any feedback will be helpful smile.gif ,

LAWNoob
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
post #2 of 8
Thread Starter 
Just got more toys! biggrin.gifWarning: Spoiler! (Click to show)




Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
post #3 of 8
What is the air temperature going into the cooler? Not the room ambient temperature. The temperature of the air 20-30mm in front of cooler intake. wink.gif


Reason I ask is I've tested cooler monitoring intake air temp and room temp and they do not remain the same even on an open bench. This is especially true of downflow coolers. Usually turning the fan to pull air up out of cooler instead of pushing down into cooler lowers CPU temps by several degrees.

Testing Thermalright AXP-100 for example was 4-8c cooler depending on fan speed.
PM'ed you the results of tests
post #4 of 8
Thread Starter 
Quote:
Originally Posted by doyll View Post

What is the air temperature going into the cooler? Not the room ambient temperature. The temperature of the air 20-30mm in front of cooler intake. wink.gif


Reason I ask is I've tested cooler monitoring intake air temp and room temp and they do not remain the same even on an open bench. This is especially true of downflow coolers. Usually turning the fan to pull air up out of cooler instead of pushing down into cooler lowers CPU temps by several degrees.

Testing Thermalright AXP-100 for example was 4-8c cooler depending on fan speed.
PM'ed you the results of tests

I'm sorry I didn;t take the air intake temps this time, I'll definitely do that next time.
All I have is some cheap thermometer which has a huge plastic plate on the back. I can't remover that plate because that's where the numbers are at.

My bench is really open air, my living room up stairs is connected to the first floor, so the room temperature should be pretty constant even after hours of benching.

I'm test your theories on downflow coolers once I buy my new setup. It looks like I'll probably go for a delided i5-4670K or a i7-4920K.
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
post #5 of 8
i did some of my own testing with aio and air coolers but didn't document it like you have.
very nice work. i'm interested to see what the top down cooler can do especially without that second fan on the vrms.
More Fans
(9 items)
 
x100e (RIP)
(7 items)
 
 
CPUGraphicsRAMHard Drive
Athlon Neo MV-40 ATI Radeon HD3200 4GB Patriot DDR2 OCZ Vertex 3 Slim 120GB 
OSMonitorMouse
Crunchbang LG Green Screen Trackpoint 
  hide details  
Reply
More Fans
(9 items)
 
x100e (RIP)
(7 items)
 
 
CPUGraphicsRAMHard Drive
Athlon Neo MV-40 ATI Radeon HD3200 4GB Patriot DDR2 OCZ Vertex 3 Slim 120GB 
OSMonitorMouse
Crunchbang LG Green Screen Trackpoint 
  hide details  
Reply
post #6 of 8
Why is everything spray painted?

I have both the H60 and that Zalman cooler. In my experience the Zalman is the better performer by several degrees Celsius.

That's with running two of the Corsair SP120 high performance fans in a push/pull configuration on both heatsinks.
post #7 of 8
Thread Starter 
Quote:
Originally Posted by Poisoner View Post

Why is everything spray painted?

I have both the H60 and that Zalman cooler. In my experience the Zalman is the better performer by several degrees Celsius.

That's with running two of the Corsair SP120 high performance fans in a push/pull configuration on both heatsinks.

Hmm, what CPU do you have and what is the TDP of that? Did you run a stress test until the temperatures start to level out?

I need more details smile.gif\

Edit: Thanks l0max, I'll test the CM Vortex as soon as I can test it on something other than an i3-550.
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
Power Bac G5
(8 items)
 
  
CPUGraphicsRAMHard Drive
i5 6600K @ 4.5Ghz R9 280X with no FreeSync support 8GB 2800Mhz Samsung 850 EVO 250GB 
Hard DriveMonitorPowerCase
2x ADATA SP550 120GB RAID 0 AOC 4k FreeSync TN Panel EVGA SuperNova NEX 750B Parts Directly On Top Cardboard Box 
  hide details  
Reply
post #8 of 8
I have an 8350 and I noticed the difference after about an hour straight intel burn test runs. Which is long enough for the chip to get with 90/95% of its maximum temperature. I was pushing the chip pretty hard last night to see how far I could get at stock voltage.
Edited by Poisoner - 8/1/13 at 11:08am
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Air Cooling
Overclock.net › Forums › Cooling › Air Cooling › LAWNoob's Air Cooler VS. AIO Water Cooler Review