Sign in with
Sign up | Sign in
Your question

Arctic Silver 5 or Arctic Cooling MX-3?

Last response: in Overclocking
Share
June 13, 2010 10:12:41 PM

So I currently have an am using Arctic Silver 5. It seems nice for the most part. I don't over clock or anything. I just take apart my computers, or take apart my grandfather's computer and apply Arctic Silver 5. I removed the thermal paste from my Dell Latitude D620, and applied AS5. I made sure everything was all set, and it seems to work good? Using the stock CPU and heatsink would I notice an improvement at all? What if I reapplied the thermal paste on a MBP? Would AS5 or Arctic Cooling give me lower temps. All I do is remove the stock thermal compound an apply some AS5, and put it together.

For a desktop CPU with a heatsink and heatspreader, I apply a pea sized bad, put the heatsink over it, and let it spread it out for me. I twist the heatsink before securing it to kill those annoying air bubbles. But on a laptop, it different. This is where I want some advice. AS says to apply it to the bare die on the CPU, and spread it out with a credit card (or a smooth clean plastic edge). As hard as a try, I cant't get an even coat. I could use a razor blade, but I don't want to scratch the CPU die. So I put a really small amount of AS5 on the center of the die, and put the heatsink on, and let it spread it out. It seems to cover the entire thing, and I don't notice much improvement over spreading it. Maybe a 3F temp drop when letting the heatsink spread it. I use so little that it doesn't spill over the edges of the die, but enough to fill the die up when the heatsink spreads it.

So for just removing the stock thermal paste, which should give me better results? Arctic Silver 5 or Arctic cooling MX-3? I know that MX-3 has an advantage because if you use too much (like Apple does) and spill it everywhere it won't short out the system unlike AS5.

Also, whats the best method for a bare die on a mobile CPU/GPU? Should i apply a dab, and screw the heatsink down, and let it spread by its self, or should I spread it my self? The heatink spread method on the mobile chips seems to do a cleaner job, just like a desktop method.

a b K Overclocking
June 13, 2010 10:28:15 PM

I usually see a 2-3C decrease when I use MX-3 vs AS5, but If you don't OC and you've already put AS5 on, I'd just leave it. It gets the job done.

I've tried spreading it with a credit card/baggie, but never got better results than just applying a dot/line and letting it do the job for you.

My preferred method is putting a dab on in the shape/size of a cooked grain of rice, the CPU is not a perfect square.

As far as laptops go... I don't know. I would assume the same methods apply and AS just recommends spreading it.

Here's a good video: http://www.youtube.com/watch?v=EyXLu1Ms-q4&feature=fvw
m
0
l
June 13, 2010 10:52:39 PM

But for a mobile CPU that has no heatspreader, just a bare die, should I just apply a pea size/half grain of rice, and screw the heatsink down, and let it spread it for me, or should I follow AS5's instructions?
m
0
l
Related resources
a b K Overclocking
June 13, 2010 11:47:12 PM

"As far as laptops go... I don't know. I would assume the same methods apply and AS just recommends spreading it."

Gonna have to see if someone else knows =/
m
0
l
June 13, 2010 11:53:37 PM

Yes, because AS says to spread it with a flat thin object, like a hard plastic card. But I like to apply a dab, and screw the heatsink down, and let the heatsink spread it for me. Its more cleaner, and it seems to do a slightly better job. Apple uses one whole syringe, and then screws the heatsink down making a Core 2 Duo sandwhich with white mayo on it.
m
0
l
June 14, 2010 8:59:30 PM

anyone have anything that they would like to add? is it okay to apply a small amount to a bare die, and screw the heatsink over it, letting the heatsink spread it? So same method as a desktop cpu, but w/o a heatspreader.
m
0
l
June 16, 2010 3:10:03 AM

I recently inherited a batch of older IBM Thinkpads and a few of them had the bare core cpu's as you are working on. I used as5 as you did by putting a rice to pea sized dab directly on the CPU core. I removed a couple after seating to see what kind of spreading I got. It turns out that at least for me I had 100% CPU coverage with excess pushed of to the sides of CPU core. This has worked best for me on all chips and tim materials including up to the core I7 intels. I have used as5 , Diamond Ice, Ceramique , and other generic white tim materials. The only tim I have used that I spread on a CPU is that really thin Zalman stuff that comes with an applicator brush in the bottle like finger nail polish. I hope this helps you.
m
0
l
!