Stulid Tested all of OcUK Thermal Paste Range

Associate
Joined
20 Jan 2009
Posts
811
What are any of you guys opinions on what would be the best thermal cpu paste to use on my setup with the Skylake i7 6700K cpu and the corsair H80 closed loop cooler (only using one fan on the rad, no room for two fans in my very small itx build) ?

I will be trying my best to overclock the cpu, so need good decent temps.

The H80 comes with paste on but I have been told its not very good paste although I could be wrong, so thats why I need to try and find what would be best paste to use for these items and my requirments.

Thanks guys, please let me know, cheers :)
 
Caporegime
OP
Joined
17 Jan 2010
Posts
66,802
Location
weston-super-mare
What are any of you guys opinions on what would be the best thermal cpu paste to use on my setup with the Skylake i7 6700K cpu and the corsair H80 closed loop cooler (only using one fan on the rad, no room for two fans in my very small itx build) ?

I will be trying my best to overclock the cpu, so need good decent temps.

The H80 comes with paste on but I have been told its not very good paste although I could be wrong, so thats why I need to try and find what would be best paste to use for these items and my requirments.

Thanks guys, please let me know, cheers :)

It comes with a silver based paste, its fine.

Buy some newer fancier paste when it comes time to remount the cooler whenever that time arises.
 
Caporegime
OP
Joined
17 Jan 2010
Posts
66,802
Location
weston-super-mare
thanks for this however, any advice on to spread or not? i have been reading so many different ways of applying but which is the best?

If its a large CPU such as Haswell-E then spreading can be good to make sure all parts are covered, but with Haswell/Devils canyon/Skylake I just do the blob in the center and let the clamping force of the heatsink spread it out.

Video below shows various methods,


Cross method looks best.
 
Caporegime
Joined
14 Dec 2005
Posts
28,071
Location
armoy, n. ireland
Regarding Haswell E and paste, the small hole you see on the ihs of theese chips, does it matter if it gets blocked with paste? Fitting a new cooler earlier so had to get a fine brush and tease out a drop from this hole. MX4 so shouldnt cause any conduction issues if any got inside.
 
Caporegime
OP
Joined
17 Jan 2010
Posts
66,802
Location
weston-super-mare
Regarding Haswell E and paste, the small hole you see on the ihs of theese chips, does it matter if it gets blocked with paste? Fitting a new cooler earlier so had to get a fine brush and tease out a drop from this hole. MX4 so shouldnt cause any conduction issues if any got inside.

I've never had my X79 or X99 CPUs die from paste getting in the hole, I read ages ago its used to fill with glue and bond the heatspreader to the die (but then I also thought they were soldered) so if the hole is full of glue the chances of paste getting to a important part is 0%.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
The video above show the biggest problem with most TIM applications .. NO DIRECT CONTACT BETWEEN IHS AND COOLER BASE!!

TIM does not transfer heat anywhere near as well as direct metal to metal contact.
TIM is designed to fill the microscopic voids between metal surfaces, NOT CREATE A LAYER BETWEEN THEM.
Oversized image.


Additionally, the IHS size is as big as it is not to spread heat, but to provide a big enough area for cooler base to set flat and stable on. It is to thin to spread heat but maybe a couple mm from edge of chip.
Heat%20Transfer%20are%20fro%20IHS_zpsqhj1njj4.png~original


Intel specifications say mounting pressure of cooler on IHS should be 50-100 pounds per square foot .. that is at most, much less than 1 pound per inch(actually only 0.347-0.6944 pounds per square inch).

This means their is not a whole lot of 'pressure' being applied to spread and force unneeded TIM out so we end up with direct metal contact wherever possible.
 
Last edited by a moderator:
Caporegime
OP
Joined
17 Jan 2010
Posts
66,802
Location
weston-super-mare
Given that the Gelid isn't available in the shop - what's the current best-of-class for ease/performance for a 4690K?

Thermal Grizzly is.

they do a few for different types of cooling such as Aeronaut = Air, Hydronaut = water, Kryonaut = sub zero.

I dont see why you cant use Aeronaut on water or visa versa though.

Kryonaut is designed for subzero and doent freeze and crack easily.

Basically its what all the pros use to set records these days.

This dispenser - https://www.overclockers.co.uk/ther...ormance-thermal-paste-3ml-7.8g-th-004-tg.html - includes two rubber tips that makes application absolute childs play.
 
Associate
Joined
25 Mar 2016
Posts
3
Location
netherlands
Associate
Joined
15 Aug 2012
Posts
241
Great thread. I didn't realise MX-4 was worse than AS5. I've just ordered a tube of MX-4 to replace a lost tube of AS5. Still, it was less than £4 and I don't plan on doing any overclocking on the chip I'm going to put it on so no great loss.
 
Soldato
Joined
1 Jul 2011
Posts
8,632
Great thread. I didn't realise MX-4 was worse than AS5. I've just ordered a tube of MX-4 to replace a lost tube of AS5. Still, it was less than £4 and I don't plan on doing any overclocking on the chip I'm going to put it on so no great loss.
What? Data shows MX-4 at 89.75c and AS5 at 89.5c
But don't forget the so called 'testing' in this so called 'review' was done without the use of a thermometer or a record of ambient temperatures while these so called 'tests' were conducted. Simple stated, this means the CPU temps mean little to nothing because there is no baseline of air temp. We cannot calculate a delta temp because we can only guess, hoping the room was between 21c and 24c .. but that is only a guess. If this guess is close to accurate, we still have a 3c 'margin of error', meaning any all CPU temp may be 1.5c hotter or 1.5c cooler than reported .. so even though the CPU temps were taken to 100th of a degree, they could all be 1.5c warmer or cooler delta.

At best this is only 'Spread-ability' review, but it is definitely not a temperature test, and anyone who understand the basic principles needed to conduct TIM testing knows this.
 
Soldato
Joined
19 Apr 2003
Posts
13,513
How long does Arctic Silver 5 last in the tube? I've got - somewhere - a tube that's a few years old, maybe more. Will it still be usable?
If it's been stored in a cool, dry place with the cap properly secured it should be OK.

As long as it hasn't started to separate you should be good to go. I would discard the bit in the nozzle and use the TIM in the main tube.

I had a tube that was ~5 years old - and apart from a tiny bit of residual liquid on opening it was fine. Exactly the same consistency and temps as the replacement tube i bought.
 
Gangster
OcUK Staff
Joined
16 Mar 2008
Posts
7,991
Location
OcUK HQ
If it's been stored in a cool, dry place with the cap properly secured it should be OK.

As long as it hasn't started to separate you should be good to go. I would discard the bit in the nozzle and use the TIM in the main tube.

I had a tube that was ~5 years old - and apart from a tiny bit of residual liquid on opening it was fine. Exactly the same consistency and temps as the replacement tube i bought.

I think that would have been a great answer when they asked a year ago :). I wonder if they still have it.
 
Soldato
Joined
19 Apr 2003
Posts
13,513
I think that would have been a great answer when they asked a year ago :). I wonder if they still have it.
LOL - just noticed the user tag- @Quartz - (we cross paths a lot on the forums) and answered. Didn't even notice it was a sticky...

I wonder if they still have it.

...and probably not :)

*TBH - i've only just got used to acknowledging it's 2017 - kids make you lose all sense of time...
 
Last edited:
Soldato
Joined
22 Mar 2014
Posts
3,956
LOL - just noticed the user tag- @Quartz - (we cross paths a lot on the forums) and answered. Didn't even notice it was a sticky...



...and probably not :)

*TBH - i've only just got used to acknowledging it's 2017 - kids make you lose all sense of time...
I've got some old thermal paste that is about 5 years old, it's not bad tbh! When I do a new rig I'll buy some new paste :D
 
Back
Top Bottom