Sign in with
Sign up | Sign in
Your question

Rock-paper-scissors and Benchmarks?

Last response: in Overclocking
Share
January 24, 2013 6:16:06 PM

When I buy something I like to buy the best I can for the best price. There is a lot of hype there and that is what I try to avoid. Recently I have been looking at OC'ing my computer and I have been looking at benchmarks. Currently I have to the conclusion that either one of the following is correct:

1) Reputable looking sites lie about their data for one reason or another
2) Reputable looking sites have no idea what they are doing

In trying to find the best CPU cooler all I have found out is a strong mistrust. It is amazing to see the data and see how some coolers vary so widely from site to site.

I guess the best way to find a cooler is to measure the size of the heat sink then count the fins, which should give you your surface area. Then times that by the amount of air that the fan pushes might give me better results then trying to look at stats that might be bias or unreliable. In addition to that there are other factors like material copper/aluminum. The surface of the heat sink that mounts to the CPU and thermal grease that allows for the transfer of the heat from the CPU to the heat sink.

In some of these test they had wide ranges of ambient room temperature, which I have no idea why someone running extensive benchmarks on some many processors with the the temperature varying up to 10*c. I did look at Tom's Hardware Charts from 2011 for CPU coolers and it didn't look like they had all that much information.

It seems like the best heat sinks are the tower type, I am guessing this is from the airflow of some cases.

Sites I visited for information:
-Hardwaresecrets.com - This site varied up to 13*c in room ambient temperature for different tests on the same system. Their final values were based off the max temp - ambient temperature.
-Techradar.com
-bit-tech.com
-tomshardware.com

Has anyone else gone through this frustration, this is my first post here and I am sad it is a bit raging.
a b K Overclocking
January 24, 2013 7:01:00 PM

Best cooler for what? A cheap cooler with reasonable results or the all out best performance regardless of cost? Something in between?

Edit: Try rock-paper-scissors-lizard-Spock instead.
m
0
l
January 24, 2013 7:45:42 PM

I am looking for he best for the price. Right now I am looking at the ThermalTake Water 2.0 Pro or Performer. Also I am looking at the other end ETS-T40 HSF for only $35.

The Spock in me says go w/ ETS-T40, the geek says go water pro. It looks like Toms Hardware gave the ETS-T40 a good rating, which seemed to get a lot of good reviews for the most part. I was taking a look at the Hyper 212 Plus, though according to my calculations the T40 blows the 212 away as it moves more air and has more heat dissipation w/ a larger heat sink. It is only $5 more expensive.

I like the look of the Zalman 9900 Max, though at the price even w/ a $20 rebate I think the ETS-T40 will make my system run cooler.
m
0
l
Related resources
January 24, 2013 7:52:46 PM

Frostytech.com

Only does heatsink reviews. With a synthetic load. The results are consistent.
m
0
l
a b K Overclocking
January 24, 2013 8:12:08 PM

Don't go with the +, the Evo has a solid contact plate and performs better. Regardless, a quick google shows the T40 is a great cooler as well.

With this http://www.bit-tech.net/hardware/cooling/2012/12/06/ene... as a reference, the cheap T40 get you 32C better than stock. The 200% more expensive H100i on "loud" get you another 10C.

Is that 10C worth another $75? If it is, then get high end water :) 
m
0
l
January 25, 2013 1:21:28 AM

itzsn - Thx for the link

Jed - I been reading that having the heat pipes in direct contact is better than having a solid plate. Though at this time I might stay away from liquid cooling.
m
0
l
a b K Overclocking
January 25, 2013 1:27:38 PM

Err, I used the wrong terminology there - the pipes are still direct contact, there just aren't big gaps in between the individual pipes.
m
0
l
!