AS5 vs MX-2 vs MX-3?

Joined
Jul 8, 2006
Messages
763
I've finally exhausted all of my thermal paste and need to purchase a new tube; its been 5 years since I last purchased any thermal paste and what I found is most people go for the new MX-2 and MX-3 paste by Arctic Cooling. The reviews I've read show a max of 1-2C difference between all of these compounds.

So my question is, which one degrades the least over time? I put this stuff on systems I build, Xbox 360's, video cards, ect and I don't want to ever replace it if I don't have to. I've been told you can drop temps as much as 5C by replacing 2 year old AS5, is this true? Will MX-2 degrade as much over time? I also would like something that keeps as well as AS5 did as I had my 2 tubes of it for 5 years.
 
Last edited:

repoman0

[H]ard|Gawd
Joined
Apr 7, 2006
Messages
1,205
Not sure about the Arctic Cooling compounds but AS Ceramique will not degrade over time and performs just as well as anything else. I got a giant tube of it about 6 years ago and have not worried about thermal paste since. Also not conductive or capacative so will not destroy anything electronic you happen to get it on.
 

investinwaffles

Limp Gawd
Joined
Apr 26, 2010
Messages
495
I switched from AS5 to MX-2 and I have seen no difference in temps and the package says MX-2 has a ~10yr life.
 
Joined
Jul 8, 2006
Messages
763
Has anyone noticed temp drops after reapplying aged MX-2?

I know what the package says....but manufactures often over exaggerate a lot.

I've also observed AS5 performs poorly in lower temperature applications because curing time is greatly increased, do any of these TIMs perform worse in high temperature applications like 90C load GPU's?
 
Last edited:

investinwaffles

Limp Gawd
Joined
Apr 26, 2010
Messages
495
I reseated my Zalman VF-900 on my 4850 when I was removing the LED's from the fan ( I re-applied mx-2) and my temps didnt budge at all. That was after about a year, maybe more, of general use. Not long term, but that is the longest its been since I have reapplied TIM on any of my gear.
 
Joined
Jul 8, 2006
Messages
763
I reseated my Zalman VF-900 on my 4850 when I was removing the LED's from the fan ( I re-applied mx-2) and my temps didnt budge at all. That was after about a year, maybe more, of general use. Not long term, but that is the longest its been since I have reapplied TIM on any of my gear.

Good enough for me, guess I'm ordering some MX-2.

If I'm applying to a small GPU core should I use the spread method? I know line/dot is preferred for large surfaces due to air bubbles but it seems Arctic Silver recommends spread method on small surfaces still.
 
Last edited:

ThumperSD

Limp Gawd
Joined
Jun 19, 2010
Messages
297
By a tiny margin and costs twice as much and is much thicker which is something I dislike ;)

I'm not worried about 1C difference in performance.

yea im just answering the question :p

but I'd happily pay an extra $7 for it over AS5 but probably not over the MX-2
 

Jodiuh

Supreme [H]ardness
Joined
Feb 3, 2004
Messages
5,275
I HATE MX-3. It's near impossible to spread as it clumps all over the place. I doubt even a high pressure heatsink has enough force to make it as thin as possible when using the dot method.

MX-2 is 1/2 the cost and MUCH easier to spread. The difference in temp is negligible. I know dot method is best, but those GPU spreaders are HUGE and I'm not comfortable unless I use spread.

Does anyone use the dot method on a big GPU? Any tips? Apply, check coverage, use same amount?
 

Indeterministic

Weaksauce
Joined
Jul 26, 2010
Messages
95
Keep in mind you don't need the paste to cover the entire surface if it's a heat spreader. As long as most of the central area is covered and making contact with the heatsink. Biggest issue I run into is that heatspreaders are rarely very flat, usually with a slight raised ridge around the outside.
 
Top