Thermal Compound Round-Up

H

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
All I'm saying about this thermal compound round-up is that, if your product scores lower than mustard, mayonnaise, pink lipstick or diaper ointment, you might want to rethink your formula. :eek:

Following up on our Thermal Compound Roundup – January 2012 review, we are adding five more thermal compounds to our roundup, for a total of 65 different models from Akasa, Antec, Arctic Cooling, Arctic Silver, Biostar, Connectland, Coollaboratory, Cooler Master, Coolink, Deepcool, Dow Corning, Enermax, Evercool, EVGA, Gelid, Glacialstars, Innovation Cooling, Masscool, Nanoxia, Nexus, Noctua, Phobya, Prolimatech, Scythe, Shin-Etsu, Spire, StarTech, Revoltec, Rosewill, Thermalright, Thermaltake, TIM Consultants, Titan, Tuniq, Xigmatek, Zalman, and ZEROtherm. In this review, we will determine if certain products are superior to others. We will also try another alternative thermal compound to see if it works.
Click to expand...
 
I think that review is almost worthless.

1. Ambient temperatures vary from 12C - 26C. They rate the compounds based on their delta from ambient.

2. The rate of convection is dependent on the temperature delta. Need I say more? Given the same heatsink & compound, a heatsink/compound will dissipate better when there's a lower ambient temperature.

Because they are changing 2 parameters (ambient temp & compound) the results are worthless.

3. I might as well use mayonaise because it's an awesome performer and cheap, but they excluded Miracle-Whip?
 
I had a tube of AS2 last me many years. worked well enough. Luckily it finally ran out and I bought a tube of AS Ceramique to replace it.

One thing I learned from the chart is do not use chocolate!
 
I'm a bit disappointed in the rating they gave for the IC Diamond 7 carat. I was under the impression that it was better then AS5? I used it for the first time yesterday on my new build (2600K w/Coolermaster Evo). Not use to such a thick paste, but so far so good. I won't be able to push it until next weekend, but as I sit here, we're hovering nicely at 34-35c running about 6 programs and 4 webpages. ... I think it just yawned at me haha
 
I find it funny that the TIM Thermaltake included with my 120 extreme is worse than mustard and matches performance with diaper cream. I'm glad I didn't use it.
 
J Macker said:
I think that review is almost worthless.

1. Ambient temperatures vary from 12C - 26C. They rate the compounds based on their delta from ambient.

2. The rate of convection is dependent on the temperature delta. Need I say more? Given the same heatsink & compound, a heatsink/compound will dissipate better when there's a lower ambient temperature.

Because they are changing 2 parameters (ambient temp & compound) the results are worthless.

3. I might as well use mayonaise because it's an awesome performer and cheap, but they excluded Miracle-Whip?
Click to expand...

100% agree. It is utterly useless. And where the hell were they testing that they had that kind of ambient temp variance. Are they homeless testing in the park? :confused:
 
Pringals said:
I'm a bit disappointed in the rating they gave for the IC Diamond 7 carat. I was under the impression that it was better then AS5? I used it for the first time yesterday on my new build (2600K w/Coolermaster Evo). Not use to such a thick paste, but so far so good. I won't be able to push it until next weekend, but as I sit here, we're hovering nicely at 34-35c running about 6 programs and 4 webpages. ... I think it just yawned at me haha
Click to expand...

I used Diamond 7 Carat on my last build and it seemed to work well. The lesson I've learned after trying several different brands is that it's really not worth stressing about. In other tests I've seen there are only a few degrees difference between the worst and best performers. Then once you factor in the prices of some of them, it just isn't worth it to me.
 
A shame they couldn't control their control variables. This could have been useful round up.
 
J Macker said:
I think that review is almost worthless.

1. Ambient temperatures vary from 12C - 26C. They rate the compounds based on their delta from ambient.

2. The rate of convection is dependent on the temperature delta. Need I say more? Given the same heatsink & compound, a heatsink/compound will dissipate better when there's a lower ambient temperature.

Because they are changing 2 parameters (ambient temp & compound) the results are worthless.

3. I might as well use mayonaise because it's an awesome performer and cheap, but they excluded Miracle-Whip?
Click to expand...

Actually, if you read carefully, you'll notice that they only tested five compounds this time around. The rest of the list appears to be tests they've done in the past. The ambient temperature in the room stayed around 26 degrees for those five.
 
Also note that they state that the margin of error for these tests is 2 degrees. This means that, since many of the results are within the margin of error, most of the top contenders are, effectively, equivalent.
 
J Macker said:
I think that review is almost worthless.

1. Ambient temperatures vary from 12C - 26C. They rate the compounds based on their delta from ambient.

2. The rate of convection is dependent on the temperature delta. Need I say more? Given the same heatsink & compound, a heatsink/compound will dissipate better when there's a lower ambient temperature.

Because they are changing 2 parameters (ambient temp & compound) the results are worthless.
Click to expand...

20C isn't a large enough spread to affect the thermal conduction to be discernable from standard error. I have confirmed many times on my own hardware that the component/ambience differential changes very little despite large variations in ambient temperature (64F-86F depending on the season). The temperature difference has to be very large to significantly change the rate of heat exchange from something like aluminum to air.

The round-up just re-affirms what most people know, that you just need some kind of thermally conductive fluid that won't smell, disintigrate, or undergo phase change.

It's all pretty much the same.
 
Couldn't it also be a factor that a better performing heat sink would benifit more from a more efficent heat transfer? If they're running on the very limit of what the air cooler can do, won't it all look the same?


I only use Noctura NT-H1 thermal paste based on this article http://skinneelabs.com/2011-tim-results/

It performs in the top 5, is an actual affordable paste like substance, performs well even when you get a bad mount, Also non conductive so I used it on my GPU.
 
You must log in or register to reply here.
Back
Top