[section_title title=”Installation and Testing”]

Installation and Testing

Installing the Freezer i32 was a very simple process. Screw the brackets into the baseplate, then, with the motherboard in between, screw the cooler onto the backplate, and it really was that straightforward. Being such a narrow and light cooler helped when it came to fitting but there’s not much more to that story, it took just a few minutes and was the least fiddly installation I’ve gone through since attaching the last stock heatsink.

With the heatsink in place, the fan is held in place by the wire brackets and it’s time to see if it works.

Arctic Freezer i32 installed 2

My setup for testing the Arctic Freezer i32 consists of:

  • CPU: Intel Core i7 4770k
  • Motherboard: ASRock Z87 OC Formula/AC
  • Memory: 16GB (2x8GB) Team Group 1800MHz
  • Graphics Card: MSI Gaming G1 NVIDIA GeForce GTX 980 Ti
  • CPU Cooling: Arctic Freezer i32 Semi Passive
  • Thermal Paste: Noctua NT-H1

As I mentioned earlier, the Arctic Freezer i32 comes with a sachet of MX-4 thermal paste but we won’t be using that for this review – for consistency we use Noctua NT-H1 so that we see the difference in performance between the hardware and not the difference in the effectiveness of thermal paste. Also for consistency, a performance fan controller curve is used in each case.

Thermal performance

Thermal performance is judged on four factors overall; the idle temperature at stock and overclocked speeds, as well as the loaded temperatures at stock and loaded speeds. During the testing, whether that be the idle or loaded testing, the system is either at idle or full load for a duration of five minutes before any readings are taken. The average temperature across all four cores is then noted down and used as the results. All room temperatures are recorded beforehand as well as during the test to ensure that the delta temperature is as accurate as possible.

Arctic 3.9GHz Idle

Arctic 3.9GHz Load

Arctic 4.5GHz Idle

Arctic 4.5GHz Load

As an additional test case, I also checked the CPU temperature at idle with the fan turned off to see what temperatures I would get from the passive nature of this cooler. The delta temperatures were around 25-30’C at idle and low intensity usage and it was silent since the fan wasn’t running.


This is not an overclocking cooler – it’s a semi-passive, minimal noise affair with a 1300rpm fan, but even so I was able to run our Prime95 test at 4500mHz and get a very respectable result.

With the spare fan brackets and splitter on the main fan allowing easy installation of a second fan it’s easy to see that those temperatures can come down somewhat and though I considered doing just that I decided to stick with what’s in the box as far as this review goes since I didn’t have access to a spare Arctic F12 fan and using another brand just opened up too many variables.

Previous articleNZXT Grid+ V2 Fan Controller Review
Next articleASUS Z170 ROG Maximus VIII Hero Alpha Motherboard Review


  1. Hi, thx for review! I have strange question… I have i5 6500 and case with 3 intake and 3 outtake fans. Can I use this heatsink without cooler?) And maybe you will recomend me somehing? thx a lot.

    • Hi Denys, It’s a while since I used this cooler for the review but I recall the fan was only really necessary once the CPU reached about 30’C. I think that the i5 6500 runs a bit hotter than that so if you are relying on purely passive cooling from the 6 fans you already have then it will depend on their placement and how the air flows internally. It will also depend on what you are using the PC for whether or not the temperature of the CPU climbs. Since you have so many fans already I don’t think another one attached to the CPU cooler would be problematic though if they are all connected to the motherboard you may have to start using a hub for some of them. If you try it set up purely as a passive cooler though I would be very interesting in hearing your findings.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.