Hardware Secrets Forums


Go Back   Hardware Secrets Forums > Miscellaneous > Content Comments



Reply
 
Thread Tools Search this Thread Display Modes
Old 05-02-2011, 09:32 AM   #1
Hardware Secrets Team
Administrator
 
Join Date Nov 2004
Posts: 5,595
Hardware Secrets Team is on a distinguished road

Default Thermal Compound Roundup - May 2011

There has been a new article posted.

Title: Thermal Compound Roundup - May 2011
URL: http://www.hardwaresecrets.com/artic...-May-2011/1249

Here is a snippet:
"Following up on our Thermal Compound Roundup - April 2011 review, we are adding five more thermal compounds to our roundup, for a total of 20 different models from Antec, Arctic Cooling, Arctic Silver..."

Comments on this article are welcome.

Best regards,
Hardware Secrets Team
http://www.hardwaresecrets.com
Hardware Secrets Team is offline   Reply With Quote
new Sponsored Links

Old 05-02-2011, 12:21 PM   #2
C.W.
Senior Member
 
C.W.'s Avatar
 
Join Date Jan 2011
Posts: 425
C.W. is on a distinguished road

Default

Results are somewhat underwhelming, but consistent. Outside of a longitudinal study over maybe a couple of years, this is pretty comprehensive.

I'm curious if the thermal compound found on the heatsinks you get with your CPU have similar performance or significantly worse.
C.W. is offline   Reply With Quote
Old 05-02-2011, 02:24 PM   #3
Rafael Coelho
Super Moderator
 
Rafael Coelho's Avatar
 
Join Date Jun 2008
Posts: 182
Rafael Coelho is on a distinguished road

Default

Quote:
Originally Posted by C.W. View Post
Results are somewhat underwhelming, but consistent. Outside of a longitudinal study over maybe a couple of years, this is pretty comprehensive.

I'm curious if the thermal compound found on the heatsinks you get with your CPU have similar performance or significantly worse.
We will review the thermal compound that comes in stock coolers soon.
__________________
There´s nothing as practical as a good theory.
Rafael Coelho is offline   Reply With Quote
Old 05-02-2011, 10:49 PM   #4
Olle P
Senior Member
 
Olle P's Avatar
 
Join Date Oct 2008
Sweden
Posts: 1,736
Olle P is on a distinguished road

Default

Quote:
Originally Posted by C.W. View Post
I'm curious if the thermal compound found on the heatsinks you get with your CPU have similar performance or significantly worse.
At least it used to be pretty good.

Tom's Hardware did a test with a P4 Prescott Extreme (notorious for running hot) quite a while back.
- The stock cooler with the pre-applied TIM was just about sufficient to provide the required cooling in a basic computer case.
- Replacing that TIM with some cheap silicon based stuff worsened the temps decidedly out of specs.
- Replacing again, now to AS5, resulted in temps about 2C below that of the original TIM.

Cheers
Olle
Olle P is offline   Reply With Quote
Old 05-03-2011, 01:35 AM   #5
C.W.
Senior Member
 
C.W.'s Avatar
 
Join Date Jan 2011
Posts: 425
C.W. is on a distinguished road

Default

If you read Newegg reviews, the overwhelming trend is that people who buy paste are satisfied with it. If they're not satisfied, it's usually because they spread too much and made a mess OR they feel the product is inferior to product X and thus downrate it. Even reviews where they give quantitative evidence, the jump is far outside what you would expect. Other reviews cite that there's very little difference between before and after (a slight increase or decrease which could also be attributed to random variation)- which actually confirms Rafael Coelho's tests.

Let's assume that a majority of thermal compounds work with +/- 2.5 degrees of the statistical mean (so far 34.33 if I'm not missing data). Given a bit of wiggle room, we'll say that anything +3 degrees above the mean is inferior. This still gives us a large amount of thermal compounds that are still adequate. At this point, it would be logical to add more variables to determine the "best" thermal compound. For example, what sort of price/g are we looking at? How much viscosity does the compound have (and would it have an effect on the experience of applying it)?



Finally, I was thinking that there would have to be some sort of literature about the compounds or something similar and how it reacts to increased heat. I bet that could provide actual evidence toward the whole set-in or burn-in conjecture. I would speculate that depending on what they are comprised of, they react differently to heat and have different thresholds (overheating resulting in decay of the product). That means one product on a hot CPU might do poorly (as it actually breaks down the compound) whereas another may retain its properties. Of course this is probably more complicated and time consuming than you'd like.
C.W. is offline   Reply With Quote
Old 05-03-2011, 06:50 PM   #6
dlb
Senior Member
 
dlb's Avatar
 
Join Date May 2010
Posts: 16
dlb is on a distinguished road

Default

Great round-up! I wasn't surprised to see some standard "top contenders" still in the top spots. However, it would be nice if some of "less modern" or maybe "previous top contenders" were included, like Arctic Cooling MX-2 and/or MX-3.

Also, I was surprised to see that Arctic Silver Céramique scored as well as it did. Not to suggest that it's a low-end TIM, but it is inexpensive, and it's white, where most "top contenders" are relatively expensive and are grey. I have a HUGE syringe of AS Céramique and I've been using Arctic Cooling MX-2 instead. Should I be using the Céramique?

(PhenomII 955BE at stock 3.20ghz w/ occasional OC to 3.6/3.8; Spire ThermaxII Eclipse cooler in push/pull)
__________________
Asus M4N72-E; AMD PhenomII X4 955BE @ 3.20ghz w/ Spire ThermaxII Eclipse cooler; 6gb DDR2; 750gb+320gb+250gb HDD; 2xDVDRW; Nvidia Zotac GTX570; PowerBird900 PSU; Win7 X64
dlb is offline   Reply With Quote
Old 05-03-2011, 10:11 PM   #7
parsec
Senior Member
 
Join Date Jul 2010
Posts: 50
parsec is on a distinguished road

Default

I just love it that you did a test without TIM, and it surprised me how essential TIM really is. I'm not familiar with the profile of the heat spreader of an i7-860, but it looks like it is somewhat concave, given the picture of the TIM "fingerprint" (I like your terminology). IMO, TIM may serve more as a medium to create a good interface between the CPU and heat sink base on a macro scale, given that many Intel CPUs have a concave IHS surface.

Thanks for referring to the product you are testing as "TIM", rather than grease or paste, the latter is unprofessional IMO.

Some of the younger PC enthusiasts may not know that TIM really had it's origin for use with large single transistors, years before micro-processors existed, like the kind still used today in (analog) high power audio amplifiers. These transistors have a metal surface that transfers heat to a heat sink. The metal surface usually is one of the main electrical connections of the transistor, so must be insulated from the metallic heat sink. A flat, thin "washer" made of the mineral mica cut to match the shape and size of the transistor was used as the insulator, and TIM was applied to the transistor, both sides of the mica washer, and the heat sink. This worked fine but was messy and labor intensive. The TIM was usually a white, silicone based compound that is rather viscous and oily.

Recently the mica washer and TIM has been replaced with the "sil-pad", a flexible silicone impregnated material resembling soft plastic, in the same shape as the mica washers. They are relatively thick (3-4 sheets of paper) and used without TIM. I've often wondered if this material may someday be used with CPUs and their heat sinks. I'd try it myself, but the pre-cut pieces are smaller and are not shaped like a CPU heat spreader.
parsec is offline   Reply With Quote
Old 05-04-2011, 03:40 AM   #8
Rafael Coelho
Super Moderator
 
Rafael Coelho's Avatar
 
Join Date Jun 2008
Posts: 182
Rafael Coelho is on a distinguished road

Default

Thank you guys.

Parsec, I remember to use mica insulators with white thermal compound in both sides, while mounting audio amplifers and power supplies. It was really messy. I doubt the sil-pads have the same heat conduction. By the way, those sil-pads are relatively easy to find on laptops.

Last edited by Gabriel Torres; 05-04-2011 at 10:16 AM. Reason: grammar
__________________
There´s nothing as practical as a good theory.
Rafael Coelho is offline   Reply With Quote
Old 05-05-2011, 08:01 AM   #9
hangfirew8
Member
 
Join Date Jul 2010
Posts: 31
hangfirew8 is on a distinguished road

Default

Great job, Rafael.

If you do another round, I'd like to nominate Innovation Cooling's Diamond 7 (aka ICD7), if only because I own some and would like to know how badly I was taken.
hangfirew8 is offline   Reply With Quote
Old 05-05-2011, 09:08 AM   #10
Rafael Coelho
Super Moderator
 
Rafael Coelho's Avatar
 
Join Date Jun 2008
Posts: 182
Rafael Coelho is on a distinguished road

Default

Thanks Hanfgirew8, we will try to get this one.

Last edited by Rafael Coelho; 05-05-2011 at 09:20 AM.
__________________
There´s nothing as practical as a good theory.
Rafael Coelho is offline   Reply With Quote
new Sponsored Links

Reply

Share This Thread

Bookmarks

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Forum Jump


All times are GMT -8. The time now is 12:46 AM.


vBulletin Security provided by vBSecurity (Lite) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
vBulletin Optimisation provided by vB Optimise (Lite) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd. ()
© 2004-12, Hardware Secrets, LLC. All rights reserved.