• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Best way to apply paste to CPU?

I dont see how a grain of rice sized blob in the middle is the same amount as a line accross the cpu or spread with a credit card. waynes pic looks like he has used loads
 
I agree it does look like I was making a jam sandwich but I used a real small amount of Arctic Silver 5 (AS5) and spread it out, in hindsight I should have taken a before "blob" picture! :D

Practice makes perfect, horses for courses, and less is more! :cool:
 
Pretty sure that for AS5 due it's thickness that the credit card method is superior to the blob method - AS themselves only changed it due to the issues experienced with early Core 2 chips where the IHS was concave and the layer left using a credit card wasn't thick enough to actually touch the heatsink.
 
I used the line technique as described by AS in the pdf. All cores are around the same temperature. I hardly used any paste at all
 
i use the clingfilm wrapped around the finger and spread method and have never had any heat issue's :)
i try to spread the paste just thick enough so that you cant see the writing on the heatspreader and the cpu is fully the colour of the paste.

What do you mean direct contact heatsink.

by direct contact i think he means ones like this where there are pretty large channels that may cause the blob method not to spread properly.
 
Last edited:
i think i spread MX -2 wrong on my core i7 920. any advice guys. i though the whole cpu needs to be covered to cool the cpu better. shows what i know.
 
I've noticed that MX2 is a bit harder to spread effectively than AS5. It's rather easy to remove it entirely with a credit card rather than spreading it
 
by direct contact i think he means ones like this where there are pretty large channels that may cause the blob method not to spread properly.

Personally, the way I tackled this was to first "fill" the channels with TIM, then use the blob method. Seemed to worth rather well.

I've noticed that MX2 is a bit harder to spread effectively than AS5. It's rather easy to remove it entirely with a credit card rather than spreading it

Yea, I also found MX-2 and 3 much harder to spread than AS5.

Smooshing is the answer :D
 
i though MX-2 is better than AS5?

Cooling performance is very similar once AS5 has bedded in. MX2 has the advantage of being non-conductive electrically (so won't short circuit something if you make a mess) and requiring no curing time.

Plus it's supposed to last longer but I've never found AS5 to really need a reapplication.
 
Don't spread on any recent CPU.

If they have heat pipes like the images posted earlier you need to use a good quantity of paste - more than usual.

Otherwise, horizontal line for core 2 CPUs, vertical line for i5/i7 and blob for AMD CPUs - tho some AMDs should be done like the core 2, probably the new quad core ones - you might get good results with spreading on some AMD CPUs but without knowing the specifics I wouldn't reccomend it.
 
i though MX-2 is better than AS5?

Yes, from my experience (and some reviews) mx-2 (and 3) perform better. However, if you are a person that spreads the TIM accross the CPU heatspreader then you have a very tough time with MX-2(3) due to its consistency and tendency to blob up and be annoying.

However, if you use the blob or line methods (or any other smooshing method) then you have none of these troubles and it works great- you just need to ensure you apply the correct volume of TIM to the chip.
 
Below is how I spread my paste, [email protected] with tuniq 120 extreme, idle 35, load 65, sweet.


he's put too little on there imo the best results I have found are same method but just enough so you cant see the chip through it.
A good way is to do what he has done then add a tiny blob in the middle afterwards as well.

Anyway all chips are different some are convex and heatsinks can be too only way to be sure is trial and error on your own chip and heatsink.

If anything is slightly convex you generally need to apply more.

I suggest if people have the time to try all of them and see which works best for you.

I remember trying line, blob, and high five method on my 6420 + tuniq tower it was a good 4-5 c higher than using the spread method.

This could be down to a number of factors though such as heatsink movement when you are screwing it to the backplate and so on.
 
Last edited:
Back
Top Bottom