HardwareLogic

Go Back   HardwareLogic > General Discussions > User Guides/Reviews
Home Forums Rules All AlbumsBlogs Donate Subscriptions Register Mark Forums Read vBExperience

User Guides/Reviews Want to be a published writer? Want to praise or flame something you just bought? We want to know what you think.

Reply
 
LinkBack Thread Tools
Old May 26th, 2007   #1
..
Points: 2,434, Level: 29
Points: 2,434, Level: 29 Points: 2,434, Level: 29 Points: 2,434, Level: 29
Activity: 0%
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Mar 2007
Posts: 452
Default Does Thermal Compound Matter

Thermal Compound is the unsung hero of CPU cooling. It is the material that sits between your massive heatsink and CPU, making sure that at all times a good contact has been made. Over the years thermal compounds have been the source of discussion when it comes to how to apply it and which is best.

Currently in the industry the recognized standard of Thermal Compound is Artic Silver. Some have claimed as much as a 5c temperature difference by using the correct compound, are these claims true?

For purposes of these tests we will use three different compounds, Zalman ZM-STG1, Arctic Silver Ceramique and a generic white compound from a local computer store.

For purposes of this test temperatures where not recorded until the system had been running for 3 days to allow the compound to setup. The system used was an E6600 overclocked to 3.2 Ghz with a Thermaltake Big Typhoon VX cooler in an Antec Nine Hundred Case. To ensure the load of the cooling was on the cooler, thus testing it's contact, the fans on the case where set to low. Testing was performed in a room temperature of 30c. All testing was recorded when the processor was under load for at least one hour using Folding@Home dual core client. before applying the compound for each test the CPU and heatsink where cleaned with Arctic Silver Cleaning solutions.



Zalman ZM-STG1 is a bit different from the typical thermal compound. Rather than coming in a tube that your squirt out, this compound is applied with a brush from the jar.

The compound is bit thinner than other compounds tested and so the brush system works really well. Zalman's instructions call for the compound to be spread lightly on the heatsink base and the CPU.

The compound applies cleanly and smooth, making it the easiest of the three tested to install.



Artic Silver Ceramique is considered one of the premier thermal compounds on the market today, it is one of the standards the others are judged by.

The compound comes in tube dispenser as a thick compound. The application directions call for a bead to be applied to our dual core processor, vertically on the processor.

The tube took a bit of effort to force out the compound. The bead method is easy to apply but leaves a bit of guesswork on how thick. You must be careful to not let the compound pool up at the start or end of your bead.

Finally we have our generic compound. This compound claims it is silver based and comes in a small pouch. Unlike the other two it only has enough for a single application and has no way to reseal if you did not use it all anyway. For purposes of application you cut a hole in the pouch and squeeze it out.

With no instructions I decided for testing to use the tried and true method of spreading the compound using a credit card on the CPU. This compound fell between the other two in terms of how thick it was. It was the messiest and slowest to install of all the compounds.

Now that we have our three compounds lets see what each delivered. Temperatures reported are the highest core temp readying using Speedfan.
  • Generic Compound: 52c
  • Zalman Zalman ZM-STG1: 47c
  • Artic Silver Ceramique: 46c
To be honest the results surprised me. I firmly expected that the choice of compound would make only minor differences in the cooling of the CPU. When the test results of the Zalman hit I was really surprised at the level of drop that had taken place.

Now while this does show a nice drop from a generic compound, not the only 1c difference between the Zalman and Artic Silver. I think it is a fair conclusion to draw that at the higher end of the thermal compound market the differences will be less noticeable.

When you realize that even some of the highest end compound sell for around $10 the effects become even more dramatic. I mean for around $10 you can cut around 6C from the temperature of your CPU under load. We spend more than this many times to cut less with a new case fan.



Now in this testing I did use a heatsink cleaning solution, ArtiClean. This cleaner is a two step compound that REALLY works. Step one will eat any thermal paste or compound for lunch and let it easily wipe away. Step two follows this and finishes the cleaning, making the surface as clean as is possible.

Some will claim that the use of alcohol is enough, and it likely is but for my money this is without a doubt the best way to clean a heatsink and CPU of the generic junk thermal pad the stocks heatsinks come with. $10 buys you enough to do a few heatsinks and CPUs, if you are not making a lot of changes to your system go in with some buddies and buy a community cleaner set, it really is worth it.

I would suggest from this test that no heatsink be installed using stock or generic compound material. A little extra time at the beginning of the build will ensure a cooler running PC.

As for which compound I think is best? On that matter I am torn. Ceramique delivered the best numbers and Artic Silver has the best reputation. However the Zalman I really felt was the easiest to apply correctly. In the end I think I would consider this a tie and recommend either product without reservation. If you are going to use a Zalman 9700 cooler for your CPU the good news is this comes free with the cooler.



Computer Ed
Core2 Duo E6600 | Gigabyte 916P-DS3 | 4 Gig Corsair XMS2 | ATI HD 2900XT
X Fi Xtreme Gamer | WD SE16 32 Gig |Liteon 20X DVDRW SATA | Bose Companion 2.0
Antec Nine Hundred | Thermaltake Toughtpower 1KW | BenQ FP202W | Vista Ultimate 64
Computer-Ed is offline   Reply With Quote
Old May 26th, 2007   #2
HardwareLogic's Otis Campbell (retired from Mayberry)
Points: 9,449, Level: 65
Points: 9,449, Level: 65 Points: 9,449, Level: 65 Points: 9,449, Level: 65
Activity: 74%
Activity: 74% Activity: 74% Activity: 74%
 
stinger608's Avatar
 
Join Date: Jul 2006
Location: 1936-2006
Posts: 2,413
Blog Entries: 1
Default Re: Does Thermal Compound Matter

Good article Computer-Ed!!!!



stinger608 is offline   Reply With Quote
Old May 26th, 2007   #3
Points: 11,593, Level: 70
Points: 11,593, Level: 70 Points: 11,593, Level: 70 Points: 11,593, Level: 70
Activity: 9%
Activity: 9% Activity: 9% Activity: 9%
 
Tyreal's Avatar
 
Join Date: Jan 2006
Location: British Columbia, Canada
Posts: 2,554
Default Re: Does Thermal Compound Matter

Great article!

But what do you mean by Zalman was the easiest to apply? I'd say any AS product is the easiest. On an IHS simply slap a small rice sized grain on and you're done.




I Like Watercooling. D-Tek Fuzion, MCP655, MCR220
Tyreal is offline   Reply With Quote
Old May 26th, 2007   #4
The Final Word
Points: 30,935, Level: 100
Points: 30,935, Level: 100 Points: 30,935, Level: 100 Points: 30,935, Level: 100
Activity: 100%
Activity: 100% Activity: 100% Activity: 100%
 
Capper's Avatar
 
Join Date: Dec 2005
Location: Las Vegas, NV
Posts: 5,382
Blog Entries: 6
Default Re: Does Thermal Compound Matter

Ty, have you ever used the Zalman Thermal Compound?

Zalman ZM-STG1 Thermal Grease Review

The stuff is incredibly easy to use, both applying it and removing it....just "paint" it on with the included brush



INTEL QX9650
ASUS Maximus Extreme
4GB DDR3-1600
Sapphire HD 3870X2
Danger Den Torture Rack (Custom W/C)
Seagate 750GB HDD
BFG ES 800W PSU
Capper is offline   Reply With Quote
Old May 26th, 2007   #5
The time has come for Revenge of the Nerds!
Points: 12,437, Level: 72
Points: 12,437, Level: 72 Points: 12,437, Level: 72 Points: 12,437, Level: 72
Activity: 44%
Activity: 44% Activity: 44% Activity: 44%
 
One4yu2c's Avatar
 
Join Date: Jan 2006
Location: Land of the Lounge Lizards
Posts: 2,487
Blog Entries: 3
Default Re: Does Thermal Compound Matter

Quote:
Originally Posted by Tyreal View Post
I'd say any AS product is the easiest. On an IHS simply slap a small rice sized grain on and you're done.
Or, according to their updated instructions, a thin line down the middle for dual-core processors. Too much or too little though, and it becomes counter-productive.



One4yu2c is offline   Reply With Quote
Old May 26th, 2007   #6
Points: 11,593, Level: 70
Points: 11,593, Level: 70 Points: 11,593, Level: 70 Points: 11,593, Level: 70
Activity: 9%
Activity: 9% Activity: 9% Activity: 9%
 
Tyreal's Avatar
 
Join Date: Jan 2006
Location: British Columbia, Canada
Posts: 2,554
Default Re: Does Thermal Compound Matter

Quote:
Originally Posted by Capper View Post
Ty, have you ever used the Zalman Thermal Compound?

Zalman ZM-STG1 Thermal Grease Review

The stuff is incredibly easy to use, both applying it and removing it....just "paint" it on with the included brush
I've seen it around and read about it. Personally I find just putting a small dot or a line is much easier than "painting" it on. I'm impressed by it's performance, I remember reading a long discussion about it somewhere.

As for the thin line method, I haven't heard about it until now. Thanks for letting me know, I might try it next time.




I Like Watercooling. D-Tek Fuzion, MCP655, MCR220
Tyreal is offline   Reply With Quote
Old May 26th, 2007   #7
ButtHead
Points: 19,285, Level: 87
Points: 19,285, Level: 87 Points: 19,285, Level: 87 Points: 19,285, Level: 87
Activity: 55%
Activity: 55% Activity: 55% Activity: 55%
 
Jokerswild's Avatar
 
Join Date: Mar 2006
Posts: 4,928
Default Re: Does Thermal Compound Matter

Try toothpaste and vegemite they will both work as thermal grease for about a half hour before they go bad and give decent temps.



Jokerswild is offline   Reply With Quote
Old May 26th, 2007   #8
Points: 7,009, Level: 55
Points: 7,009, Level: 55 Points: 7,009, Level: 55 Points: 7,009, Level: 55
Activity: 8%
Activity: 8% Activity: 8% Activity: 8%
 
Join Date: May 2006
Location: Rhode Island USA
Posts: 1,716
Default Re: Does Thermal Compound Matter

Quote:
Originally Posted by Jokerswild View Post
Try toothpaste and vegemite they will both work as thermal grease for about a half hour before they go bad and give decent temps.
Another reader of DansData I see



Opteron 64 165--1.5GB DDR--ECS KA1 MVP(thanks HL!)--x1800GTO 256MB--Seagate 320GB SATA--Antec 550 Watt--Antec P180
Lead Head is offline   Reply With Quote
Old May 26th, 2007   #9
..
Points: 2,434, Level: 29
Points: 2,434, Level: 29 Points: 2,434, Level: 29 Points: 2,434, Level: 29
Activity: 0%
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Mar 2007
Posts: 452
Default Re: Does Thermal Compound Matter

While the bead seems easier there are issues when doing it. First the compound could be cooler and thus thicker meaning more pressure is needed to get the paste to come out. A glob at the front can block the paste at first. All of these are really minor but they can easily result in way to much paste being applied.

Also the bead or drop does not always detach itself from the tube easily and you can end up dragging it around fairly easily.

Now I am not saying the difference is HUGE, I am saying for the average user wanting to make 100% sure he does it right with no muss that the Zalman is a little easier out of the gate.



Computer Ed
Core2 Duo E6600 | Gigabyte 916P-DS3 | 4 Gig Corsair XMS2 | ATI HD 2900XT
X Fi Xtreme Gamer | WD SE16 32 Gig |Liteon 20X DVDRW SATA | Bose Companion 2.0
Antec Nine Hundred | Thermaltake Toughtpower 1KW | BenQ FP202W | Vista Ultimate 64

Last edited by Computer-Ed; May 26th, 2007 at 12:00.
Computer-Ed is offline   Reply With Quote
Old May 26th, 2007   #10
Deus Sol Invictus
Points: 11,869, Level: 71
Points: 11,869, Level: 71 Points: 11,869, Level: 71 Points: 11,869, Level: 71
Activity: 100%
Activity: 100% Activity: 100% Activity: 100%
 
polobunny's Avatar
 
Join Date: May 2006
Posts: 4,278
Blog Entries: 5
Default Re: Does Thermal Compound Matter

You didn't let the silver compoud cure, though. Silver compounds need curing.
I'm not sure it would change something however, it's still cheap compound.

BTW, instructions for dual core AMD and dual core Intel processors differ on Artic Silver website. Thin line for Intel, small dot for AMD.

Edit: Oh and in my honest opinion, Zalman is the easiest to apply. It's true you can easily botch the application of Artic Silver ceramique (or any thermal compound in a tube for that matters) but it would be really really hard to apply Zalman paint on compound in a wrong way. It doesn't give the "best" performances but it's still more than decent.




If there is a final hour, let's hope for a higher power. One by one, and two by two, I have ammo, what about you?
polobunny is offline   Reply With Quote
Reply

  HardwareLogic > General Discussions > User Guides/Reviews


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
The HardwareLogic Thermal Compound Round-Up Capper Cooling 77 March 11th, 2008 12:23
IC Diamond 7 Carat Thermal Compound Giveaway Capper Cooling 183 March 10th, 2008 10:39
Arctic Cooling MX-2 Thermal Compound Review One4yu2c Cooling 5 November 16th, 2007 15:42
Arctic Silver 5 Thermal Compound Xtremeoverclocking Troubleshooting 9 July 26th, 2007 05:02
Arctic Cooling MX-2 Thermal Compound Capper Cooling 5 July 10th, 2007 16:50


All times are GMT -8. The time now is 23:32.


Powered by vBulletin® Copyright ©2000 - 2008, Jelsoft Enterprises Ltd.
SEO by vBSEO 3.1.0
© HardwareLogic 2005 - 2008. All Rights Reserved


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45