Cooler Master Hyper 212 Plus CPU Cooler Review Comment Thread

[AkG]

As Skymtl promised here is another set of charts, this time with the NF-p12-1300 fan used on the 212+.

As we said in the review the Prolimatech is still the king of cooling. Will this change with our next cooler? Who knows! But for right now one thing is for certain the fan which accompanied our 212+ is one serious fan. At stock and moderate loads the OCZ Vendetta 2 and the 212+ showed very similar results but as the heatload was turned up, the 212 w/ its 4 (smaller) heatpipes did pull ahead. Does this change our results in the least? No. The 212+ is still a Dam Good Value CPU cooling solution.

[SKYMTL]

Whattdaya know.....exactly what we were predicting.

[miggs78]

Thanks AkG.. We appreciate your hard work..

First of all, interesting to see the Vendetta 2 actually having a very slightly unnoticeable lead over the Hyper 212+ w/Noctua fan, on avg temp only by 0.1. but still.. Also the V10 passes it by a small margin on MAX OC.

But look at the difference in temps with the Noctua compared to the Stock CM fan, 3C degrees on idle, and 2+C on load, now I sure agree that's some serious fan. CM sure didn't give cheap stuff out, they gave some quality equipment. Taking all these results in to mind, I would certainly either use that CM fan for real, or buy an even higher RPM fan.

Only one thing comes to mind, I've seen lots of fans by manufacturer prove solid at first go, but as time goes on the fan start dying, meaning I hope this CM Fan does last in the long run.

[AkG]

Thats an excellent question miggs78!
IF CM really is shipping the 212+ w/ Rifle Bearing fans it should last a long time. I prefer FDB or hydro bearings but RB's are a good improvement over the old sleeve bearing when it comes to longevity. Only time will tell on what makes it to market but I hope it is the same as ours. You can easily tell the difference just by holding the fan in one hand and spinning the blades with the other. If it feels and sounds "click clack" or "tick tick" its most likely RB.

[miggs78]

Thanks for clarifying, I am looking at getting this, hopefully it comes out soon in Canada. I don't overclock all that much, so water is a pain for me, especially to clean the loop every few months, good airflow.. no time man..

Sorry one last question, the fan wire clips that come with the cooler, can they accommodate a 38mm fan by chance..

:)

[AkG]

They will easily work with most fans. Unlike some cooler setups which go through the front edge of the fans holes these guys go through the holes nearest the cooler's face. As long as the fan you want uses the more common open style corners and not the tunnel style you are golden. :)

[unclewebb]

Originally Posted by SKYMTL

Now you may ask yourself: "why don't you use peak temperatures". That's because every now and then a temperature monitoring program will have a millisecond spike which while high does not reflect the actual load temperature of the processor on a regular basis. Averaging things out over 15 minutes allows us to avoid such oddities so they don't impact the final temperatures in a meaningful way.

I don't know about Core Temp but I do know that RealTemp logs the data to a file and also keeps track of the Maximum temperature within the GUI and not once have I ever seen a single random peak in recorded temperatures.

If you run a test like Prime 95 and use the Small FFTs option, it will load the cores equally and the core temperature tends to stabilize fairly quickly when testing with an air cooler on the bench or in an open case. The data when plotted is mostly a flat line.

Here's an E8400 at 4000 MHz running Prime 95 Small FFTs using the RivaTuner / RealTemp plug-in:



In less than 2 minutes the peak temperature had pretty much stabilized at 64C. During the next 9 minutes it oscillated a little bit between 64C and 65C but there were no random peaks. There never is when I'm testing. If you are using Core Temp and are having problems then maybe it's time to change software.

RivaTuner makes testing easy. Select an area of the graph and it automatically calculates the Minimum, Maximum and Average. In this example, it shows the range over a 9 minute interval while running Prime 95 between 64C and 65C and the average is 64.17C on both cores. That's excellent consistency. Come up with a standardized test procedure and this software will provide you with consistent and repeatable results.

If you want some more examples check out the recent cooler reviews at X-Bit Labs. They are very happy with the results they are getting from RealTemp and the RTCore.dll plug-in.

You can download RealTemp and the plug-in here:
http://www.fileden.com/files/2008/3/...alTempBeta.zip

I also included i7 Turbo and i7 Turbo V8. These programs are the most accurate way to monitor the CPU multiplier using the Intel recommended method as outlined in their November 2008 Turbo White Paper. This data can be used to watch for turbo throttling at full load which is a problem on some motherboards like the Asus P6T Deluxe when pushed hard.

[SKYMTL]

Thanks for all the info. Once the latest round of CPU cooler tests are done (AkG has a number of units which are in the middle of testing), we will carefully look at including peak numbers in our tests as well as average numbers.

We will make sure that there aren't any randomly high numbers like were showing up with our Q6600 system.

However, if we do go ahead with it, peak numbers will only be added to our i7 results.

[SKYMTL]

For those of you who were wondering about the fan used on the 212+, I have attached information from Cooler Master regarding the specifications and a few technical documents. There is a PDF at the very end of this post as well.

[SKYMTL]

Update: Cooler Master is predicting the Canadian retailers will have this cooler in stock on Monday next week.