CPU Thermal Compound

[Ptopman]

I come here bearing but just one question. What is the best thermal interface material that we can get our hands on that wont destroy your cpu and/or heatsink? Looking at you liquid ultra... I do not care about price as long as it works beyond expectations.

Currently looking at GELID GC-Extreme but was wondering what you guys recommend.

I currently have arctic silver 5 but looking around, I am seeing that is is becoming a bit out dated.

[cesrai]

Here you go

 

http://www.tomshardware.com/reviews/thermal-paste-performance-benchmark,3616-10.html

http://forum.hwbot.org/showthread.php?t=71658

http://forums.overclockers.co.uk/showthread.php?t=18416534

[Sintezza]

paste is paste, it doesnt make sense.

 

Something like the Noctua NT-H1 paste is fine.

[Padunkadunk]

I'm just using whatever TIM comes with Noctua heatsinks and it works just fine.

[cesrai]

paste is paste, it doesnt make sense.

 

Something like the Noctua NT-1 paste is fine.

You're joking right ?

[Sintezza]

I'm just using whatever TIM comes with Noctua heatsinks and it works just fine.

 

Yeah like i said Noctua NT-H1 is totaly fine

Artic silver or any other over expensive thermal compounds, is just a big load of marketing BS.

[Sintezza]

You're joking right ?

 

No im exaly damm serious.

like i said above over expensive stuff like artic silver and what not, is just marketing blalala

It doesnt make much sense.

 

There are only a few pasts that are exaly horrible, But Noctua NT-H1 is totaly fine.

[revl8er]

Usually what comes with the heatsink is fine, but I personally use arctic mx-5 on my build.  Dropped temps a few degrees compared to the compound on my x61 cooler.

[revl8er]

No im exaly damm serious.

 

like i said above over expensive stuff like artic silver and what not, is just marketing blalala

 

I don't think $6 is overly expensive but I guess it's different for some people.  It does drop temps a few degrees, and that can mean the difference between a 4.4ghz overclock and a 4.6ghz or higher overclock.  While not important to all, some people would like to have the extra .1ghz or so.

[Sintezza]

Usually what comes with the heatsink is fine, but I personally use arctic mx-5 on my build.  Dropped temps a few degrees compared to the compound on my x61 cooler.

 

problem with pre aplied paste is that its spackled over the coolers surface.

So there will be airbubbles as soon as you push the heatsink down.

Thats why the credid card spreading method is so terrible.

[revl8er]

problem with pre aplied paste is that its spackled over the coolers surface.

So there will be airbubbles as soon as you push the heatsink down.

Thats why the credid card spreading method is so terrible.

 

It depends on the use, for most cases the pre-applied is fine.  But for everyone else you want something better.  As for the credit card method, I've used it in the past and yes it is terrible but was effective back in the day when we didn't know any better. 

[Sintezza]

It depends on the use, for most cases the pre-applied is fine.  But for everyone else you want something better.  As for the credit card method, I've used it in the past and yes it is terrible but was effective back in the day when we didn't know any better. 

 

Yeah well pre a-applied is okay.

But i would allways recommend to just put some new paste on it.

Like i said something like Noctua NT-H1. or artic silver MX4/5 if you like to pay for that... will be fine offcourse.

But in the main, there is not much of a diffrence in thermal compounds for that matter.

Its more how you apply it.

[cesrai]

No im exaly damm serious.

like i said above over expensive stuff like artic silver and what not, is just marketing blalala

It doesnt make much sense.

 

There are only a few pasts that are exaly horrible, But Noctua NT-H1 is totaly fine.

It's fine but there are differences, they use different materials and you should consider ageing the Arctic silver has amazing ageing.

[Lmp]

I use ic-diamond

[revl8er]

These days there are very little differences with the exception of aging.  I mean you might get a 1 degree difference between the compounds which isn't the biggest deal, but the cooler the better.  Don't worry about paying a premium for thermal paste though and get the $5 mx-5 instead of the $12 ic diamond (just an example).

[ThomasD]

One or two degrees of difference at full load is - practically speaking - negligible, and probably within the margin of error.

 

These tests are nice, but to be remotely statistically valid would need to be repeated multiple times.

[Frankz]

I've got Arctic Silver 4