Sign in with
Sign up | Sign in
Your question
Closed

Best Thermal Paste?

Last response: in CPUs
Share
May 30, 2012 2:36:10 PM

I'd love to see Tom's Hardware do a thermal paste round up and test to find out which thermal paste is the best at keeping the CPU cool for the best price.

So, what's the best thermal paste at the best price?

More about : thermal paste

a c 159 à CPUs
May 30, 2012 2:40:20 PM

Artic Silver 5 is very good for price and keep the temps in a decent range.
Score
0
May 30, 2012 2:49:46 PM

I don't have any experience dealing with thermal paste at all. On my last build several years ago by buddy helped me set-up my CPU and CPU cooler to make sure it was done properly.

I was reading the article below and it discusses some thermal pastes such as: Freeze Extreme and Coollaboratory's Liquid Pro

"Move up to an average overclock of 4.6GHz and the difference is even bigger with 15 degrees for the Freezer Pro and a massive 20 degrees difference using the Liquid Pro. On top of all that, higher stable overclocking speeds without thermal throttling were reached at lower Voltages using either of the replacement TIMs, although once again the best results were achieved using the Liquid Pro which allowed a stable 5GHz overclock on air at 1.55V. We should mention that the rather large and expensive Thermalright Silver Arrow SB-E CPU cooler was used for these tests, so mileage might vary depending on cooler choice, but even so, the results speak for themselves......"

"Ivy Bridge proven to suffer from poor thermal grease"
http://vr-zone.com/articles/ivy-bridge-proven-to-suffer...
Score
0
Related resources
a c 159 à CPUs
May 30, 2012 2:52:19 PM

josejones said:
I don't have any experience dealing with thermal paste at all. On my last build several years ago by buddy helped me set-up my CPU and CPU cooler to make sure it was done properly.

I was reading the article below and it discusses some thermal pastes such as: Freeze Extreme and Coollaboratory's Liquid Pro

"Move up to an average overclock of 4.6GHz and the difference is even bigger with 15 degrees for the Freezer Pro and a massive 20 degrees difference using the Liquid Pro. On top of all that, higher stable overclocking speeds without thermal throttling were reached at lower Voltages using either of the replacement TIMs, although once again the best results were achieved using the Liquid Pro which allowed a stable 5GHz overclock on air at 1.55V. We should mention that the rather large and expensive Thermalright Silver Arrow SB-E CPU cooler was used for these tests, so mileage might vary depending on cooler choice, but even so, the results speak for themselves......"

"Ivy Bridge proven to suffer from poor thermal grease"
http://vr-zone.com/articles/ivy-bridge-proven-to-suffer...


You need compare prices at this point, Gelid GC2, MX4, TX3 all those are very good for the price and give you good overclock range.
Score
0
May 30, 2012 3:10:20 PM
May 30, 2012 3:42:17 PM

I'd say Artic Silver 5, too
Score
0
May 30, 2012 3:56:12 PM

You want to stay away from Coollaboratory products. Not because they are bad (they are the best performers with temps), but because they are literally liquid metal that bonds to your cpu and heatsink. They corrode aluminum and will void your warranty, and you will have to sand the cpu and heatsink to remove the old paste whenever you replace it.

I'm currently using MX-2 and it's working great. I didn't go for the Arctic Silver 5 because of the electrical conductivity.
Score
0
a b à CPUs
May 30, 2012 3:59:08 PM

Maximum PC did a test on thermal paste and Tuiq TX-4 can out on top.
Score
0
May 30, 2012 5:30:38 PM

I'm using Arctic Silver ceramique, it's a little better than AS5 but not by much.
Score
0
a c 116 à CPUs
May 30, 2012 5:32:35 PM

http://benchmarkreviews.com/index.php?option=com_conten...

80 TIMs benchmarked, only 5C difference from best to worst, 3C if you discard the three worst entries. And despite all the raving about silver-based compounds, good old zinc oxide (one of the cheapest and oldest thermal paste materials in the books) is used in three of the top-20 entries. Aluminum oxide fills up 14 of the remaining top-20 positions including a tie for #1 against AS5. From best to worst across the top-20, the temperature difference is only 0.7C, which means roughly 0.01C/W thermal resistance difference.
Score
0
a c 116 à CPUs
May 30, 2012 6:34:23 PM

josejones said:
This article is from 2009 - does that make any difference? Has thermal paste changed much since then?

If you look at the chart, most of the compounds recommended today are the same that were recommended back then, not much change there.

Zinc and aluminum oxide have been used for thermal paste for the past 30+ years and are still the two most common "active" materials today, not much change there either. Although pastes featuring fancier materials like silver and copper are on the market, they do not exceed conventional materials' performance by large margins.

The biggest change between thermal pastes from "ancient" times and today is the finer particle size in newer compounds, which allows thinner TIM layers and better gap filling.
Score
0
a c 79 à CPUs
May 30, 2012 6:46:13 PM

josejones said:
I don't have any experience dealing with thermal paste at all. On my last build several years ago by buddy helped me set-up my CPU and CPU cooler to make sure it was done properly.

I was reading the article below and it discusses some thermal pastes such as: Freeze Extreme and Coollaboratory's Liquid Pro

"Move up to an average overclock of 4.6GHz and the difference is even bigger with 15 degrees for the Freezer Pro and a massive 20 degrees difference using the Liquid Pro. On top of all that, higher stable overclocking speeds without thermal throttling were reached at lower Voltages using either of the replacement TIMs, although once again the best results were achieved using the Liquid Pro which allowed a stable 5GHz overclock on air at 1.55V. We should mention that the rather large and expensive Thermalright Silver Arrow SB-E CPU cooler was used for these tests, so mileage might vary depending on cooler choice, but even so, the results speak for themselves......"

"Ivy Bridge proven to suffer from poor thermal grease"
http://vr-zone.com/articles/ivy-bridge-proven-to-suffer...



Those massive shifts in temp are only for a de-lidded ivybridge, most times you're looking at maybe 5 deg C difference between best in world and stock TIM (excluding soldering).
Score
0
May 30, 2012 7:35:17 PM

^ Indeed, that seems like an extreme case to me where I was under the impression that the thermal paste mentioned there were "better" than most thermal pastes. Is this not true? I suppose that's what 'InvalidError' is pointing out that they all do a fairly similar job in the end?

I'd like to know more about "Zinc and aluminum oxide" what brand name would that be under as a thermal paste and where can I get it? Has the "Zinc and aluminum oxide" been reviewed? Will it work fine on the hot Ivy Bridge CPU's? That's why I'm asking all of this is to make sure I get the best thermal paste I can for the best price.
Score
0
a c 79 à CPUs
May 30, 2012 7:52:40 PM

unless you are buying bloggs and sons super duper extra special paste, that turns out to be toothpaste with silver colouring then more or less anything will do. If you are going for extreme speeds then sure better stuff is better, but other than that for everyday use its all much of a muchness. Agree with invalid that there's not been much since that 2009 report.
Score
0
a c 116 à CPUs
May 30, 2012 8:00:15 PM

josejones said:
I suppose that's what 'InvalidError' is pointing out that they all do a fairly similar job in the end?

Basically, yes.

As noted in the thermal compound benchmark I linked, there is only a 3C difference between AS5 which was #1 in the round-up and the 77th position. CPU TDPs have not changed much over the past six years, they all have IHS so thermal transfer from IHS to HSF for a given paste would remain largely unchanged and results on modern CPUs should be roughly the same.

http://www.hardwaresecrets.com/printpage/Thermal-Compou...
In this newer roundup, even mayonnaise is a reasonable (albeit short-term) TIM, falling only 3C behind Prolimatech. There is only a 5C spread between the best and the worst option that beats lipstick.

Shows how much TIM is not really worth worrying about unless you want to overclock the heck out of a chip.
Score
0
May 31, 2012 8:18:45 PM

I'd like to know more about the thermal paste with "Zinc and aluminum oxide," what brand name it would be under and where can I get it? Has the "Zinc and aluminum oxide" been reviewed? Will it work fine on the hot Ivy Bridge CPU's?
Score
0
May 31, 2012 8:38:38 PM

i use arctic cooling mx-4 and it's excellent. I have a corsair h60 running push pull on my 15 2500k overclocked to 4.0ghz i dont trust stock paste on anything but i tried it first i was around 30c at idle which is ok but i removed the stock corsair paste went with the artic cooling mx-4 and my temps dropped 3degrees I idle anywhere from 26 to 27c and running a few games on max settings i dont go above 48c and my highest prime test temp was 55c. artic cooling makes a great paste.
Score
0
a c 116 à CPUs
May 31, 2012 9:17:17 PM

josejones said:
I'd like to know more about the thermal paste with "Zinc and aluminum oxide," what brand name it would be under and where can I get it? Has the "Zinc and aluminum oxide" been reviewed? Will it work fine on the hot Ivy Bridge CPU's?

Zinc oxide and Aluminum oxide pastes are two of the most ancient TIM materials in the world and still the two most common today. In the articles I linked, they account for more than 3/4 of entries and perform nearly on par with the best synthetic pastes produced today as those articles/benchmarks show.

Score
0
February 2, 2013 3:11:37 PM

i know this is an old post and topic and some of us still read them for others advice which in my opinion is good. any way i have tried alot of different thermal pastes and have had good results with most of them. as a ham radio tech i have used alot of the white silicone transistor compound and have had the same results as with the other compounds. white transistor compound if applied properly to the processor in my experience works as well as any other thermal compounds. just my opinion. thanks. poorguy
Score
0
!