Noise Levels

For many enthusiasts upgrading cooling the goal is maximum stable overclock, and they will live with the inconvenience of a louder system. For other users silence is the most important factor, and these users will forgo maximum overclocking if that increases system noise levels. Those who expect silence to be the domain of high-priced cooling solutions are in for a very pleasant surprise with both these Arctic Cooling solutions. It is unusual to see this much attention paid to noise reduction in value coolers, but the use of elastomeric bushings and low noise bearings show that Arctic Cooling takes noise reduction very seriously. Do these features work?

Level - 6\

Level - 24\

The Arctic 7 (with PWM) and the Freezer 7 Pro turn out to be two of the quietest coolers we have tested so far in our new cooling test bed. All four test results are at the floor of our system noise, and the new test system noise floor is equivalent to a suburban bedroom at night in noise level.

Both Alpine coolers remain silent, even when pushed hard with load testing and overclocking, which increases fan speeds. It is rare to see such low noise levels on any cooling fan, but it is particularly impressive in coolers designed to sell for less than $25.

Neither of these sub-$25 coolers will really challenge the top of our overclocking performance results, but in their effective range - which is wider than you might expect - both coolers are extremely quiet. The frameless fan, low noise bearings, and elastomeric fan mounts are obviously doing their job very well.

Cooling at Stock Speed Overclocking and Performance Scaling
Comments Locked


View All Comments

  • spidey81 - Tuesday, January 29, 2008 - link

    I was looking into the Freezer 64 Pro for an AM2/AM2+ build and was wondering if anyone knew if the performance for the AMD version performed similarly. I noticed the specs seem to be slightly different as the 64 Pro fan maxes out at only 2200 rpm pushing 40 cfm where the 7 Pro maxes out at 2500 rpm pushing 45 cfm. They appear to be of the same design varied only by the mounting hardware. Any input would be greatly appreciated.
  • Martimus - Tuesday, January 29, 2008 - link

    I have a Freezer 64 Pro. It dropped the temps on my CPU by 10C instantly from a Zalman 7000. It now runs 20C cooler after the Thermal compound cured. It is definitely a nice heatsink, although it is difficult to attach unless you remove the MB from the case.
  • mixim - Tuesday, January 29, 2008 - link

    The design is exactly the same, only the fan circuits differ to reduce the max fan rpm, this because AMD has put out low-watt CPU's for a couple of years now , when the Intel version should still be able to cope with Pentium 5/D's and so on...

    The lower maximum RPM will help with the fans longevity, but actually in my C2D system it has never went up to full speed anyway...
  • dlock13 - Tuesday, January 29, 2008 - link

    I cannot find the Freezer 7 anywhere under $25. I can't find it anywhere under $30. I checked Newegg and their price is $32.

    I am really in need of a new heatsink, and this is just perfect.
  • Spanki - Wednesday, January 30, 2008 - link

    It's $24.99 here:">

    or.. you can save some bucks and get the Cooler Master Hyper Tx 2 (same or better performance than the Freezer Pro 7 - slightly larger, quieter, slower spinning fan) for $9.99 at the same place:">

  • BladeVenom - Tuesday, January 29, 2008 - link

    Newegg has it on sale for $21.99 quite often. It can also usually be found for $21.86 on sale at eWiz.
  • 75below0 - Tuesday, January 29, 2008 - link

    I understand there's a new test bed and temp utility, but +7C for the Ultima 90 w/SFLex under load compared to last review seems inordinately steep...
    Anyone have any comparisons of CoreTemp vs. NVIDIA Monitor and/or Intel TAT (Thermal Analysis Tool) ?
  • whatthehey - Tuesday, January 29, 2008 - link

    The motherboard and BIOS change among other things so the fan speeds may not ramp up as soon but when they ramp up they might do so to a larger amount. Look at the Intel stock cooler which drops from 56C to 50C with the new testbed. They covered a lot of the changes here:">

    If you look at the results you'll see that TAT and CoreTemp both register higher than the old EVGA board by several C. Meanwhile the Ultra 120 X goes from 36C to 42C idle and 47C to 59C load. Obviously there are lots of variables. I'm not sure what margin of error is with CoreTemp and such either but it may be several C.

    The important thing IMO is to look at noise levels and max OC along with temps. Nothing exists in a vaccuum so don't focus on one metric.
  • phaxmohdem - Tuesday, January 29, 2008 - link

    I just ordered the Freezer 7 yesterday night... before this article came out. I feel WAY better now about my decision... though it is going onto a Pentium-D which may insult it :P
  • RamarC - Tuesday, January 29, 2008 - link

    freezer 7 pros are great for pentium Ds. i've got a 3.4ghz pd945 that it keeps in the mid 50s. warm? yes, but quiet since the mobo only asks the freezer to spin at 500rpm.

    i can't believe it took anandtech this long to review the freezer pro 7 considering mine is pushing 2 years old now.

Log in

Don't have an account? Sign up now