Swiftech H240 X2 Configurable AIO Liquid Cooler Review

[Crashman]

I guess you never heard of www.frostytech.com

They use a reference heat plate, not a CPU and test all heatsinks side by side using the same heat source. You'll have a list of 100 heatsinks tested side by side.

Not just them, most sites that do real cooler reviews have thermally isolated testing environments with proper heat loads, default fan and preset forced convection, and acoustic testing. On the other hand, few review sites have as comprehensive a GPU power draw setup and PSU testing, so I wouldn't really rail too much on their (current) inability to properly test CPU coolers.

And the closer Tom's Hardware got to that optimum, the more readers demanded "real world" conditions. So we got a case and started using it. You still get testing consistency, which means the numbers you see in one article are comparable to the ones you see in the next.

 

[photonboy]

Simulated vs real-world doesn't matter provided both are done correctly.

A simulated setup would simply use a resistor to produce the same HEAT output that a CPU would in either idle, default, or overclocked.

I've not monitored simulated testing but they would need a sealed case, power supply (it has a fan to exhaust heat) etc.

No point in doing outside a case.

 

[ingtar33]

I guess you never heard of www.frostytech.com

They use a reference heat plate, not a CPU and test all heatsinks side by side using the same heat source. You'll have a list of 100 heatsinks tested side by side.

Not just them, most sites that do real cooler reviews have thermally isolated testing environments with proper heat loads, default fan and preset forced convection, and acoustic testing. On the other hand, few review sites have as comprehensive a GPU power draw setup and PSU testing, so I wouldn't really rail too much on their (current) inability to properly test CPU coolers.

a heat plate fails to simulate the heat spikes that an actual CPU puts out. for all the flaws of using energy efficient cpus lightly overclocked like THG, I would rather that over a testing methodology that fails to simulate actual Computer conditions like those heat plate/open bench tests.

 

[gondo]

As long as the heat output is consistent and high, then you can have a benchmark to consistently compare all heatsinks with exactly the same conditions. Frostytech also cranks the heat level to see how the heatsinks handle high heat levels to check if the heatsink saturates or if it can handle the extra heat of full load.

I have no problems with their testing methods. Idle, full load, and noise levels at high and low speeds, and exactly the same bench setup for each heatsink. Makes comparing different heatsinks a breeze. The only problem is they don't have all heatsinks reviewed. Many of them but not all of them.

 

[shane_l_s]

You need to be careful with this one. It is an awesome looking, great performing cooler, however it has issues. I have had 3 now. All three have leaked between the fill tank and the clear reservoir. SwiftTech has been good about RMAing these with a quick cross-ship (advanced replacement) method without unnecessary hoops to jump through. But this cooler has left puddles on a, fortunately, empty drive bay. Pic here: http://shepardtech.com/Shared/Images/H240X2_PRESTIGE/Leak4.PNG