Dolphin, the GameCube and Wii emulator - Forums

(04-29-2012, 06:24 PM)MaJoR Idiot Wrote: [ -> ]

Squall Leonhart Wrote:the stock tim on a stock cooler is usually permanent, which is why you should scrape that shit off and repolish the heatink base unless you want the cooler ripping the cpu from the socket the next time you want to install a new cpu,.

...You've never built a PC before have you. In ten years of building custom PCs and doing PC repair as a side job, I have never, ever ran into "thermal glue". Never. I've taken apart Dells, Emachines, Compaqs, Laptops (that was for fun though, it's still in a million pieces in my closet), on and on, and built tons of machines for myself and others. If TIM is used, it's the same old stuff, just twist it a little and the heatsink comes right off. And did you know that on all modern systems, the heatsink completely covers the CPU? You cannot get access to the clamps that hold the CPU in place. If the thermal paste did glue the heatsink to the CPU as you claim, no one could replace their CPU, ever. That would be a pretty dumb design for a user installable part now wouldn't it? It's so dumb, even OEMs don't do it.

You've never experienced a moment of intelligent thought before, have you?.

Quote:As for "scraping it off and repolishing the heatsink", you have obviously never worked with TIM and heatsinks. Just put some alcohol on a rag and it wipes right off. No "polishing" required. Stock fans, purchased fans, they are all the same, just wipe it off you have a nice smooth surface to put some Arctic Silver onto.

This stuff CANNOT be wiped off, no matter how much alcohol you wipe it with.

I have seen this stuff not only rip off IHS from intel and amd cpu's, i have also seen it rip pre LGA socket cpu's from the socket. It cannot be wiped off of the die and melts its way into any gap it can.

MaJoR Idiot Wrote:

Squall Leonhart Wrote:I have to question the sanity of anyone who would think for a second that cpu cooling would have stopped using a form of TIM...ever.

Stopped? No. Hindered? Yes. Thermal Paste is not as good a conductor as solid metal. Here's a tidbit from that article you didn't read.

http://www.overclockers.com/ivy-bridge-temperatures Wrote:A solder attach could have a heat conductivity in the range of 80 W/mK. A TIM paste could have a heat conductivity in the range of 5 W/mK. That’s your problem right there!

Thermal Paste allows you to have a temporary connection between two metal pieces. It's not the best, but if you keep it down to just one TIM bridge, it's not that bad. The problem with having TIM connect the die to the IHS is that you are increasing the number of times thermal grease is used, and decreasing the thermal dissipation rate. THAT is why this is bad.

Completely out of the context i was going. reference was towards IHS/DIE contact to the mounted cooler.

Shamino has verified that the high temps with Ivy Bridge are due to the use of TIM between the die and heat spreader instead of conductive solder

Squall Leonhart Wrote:This stuff CANNOT be wiped off, no matter how much alcohol you wipe it with.

Does showing a random unrelated picture of TIM make you right? Not really no. But since you want images. Here.

http://img710.imageshack.us/img710/8771/img2155dn.jpg

To test your theory, I took out my C2D stock fan. It was still sealed in its original packaging, since I just use my Zalman for my C2D. What's on the rag, not on the heatsink? Thermal Paste. Just put alcohol on the rag, and wiped it off. Done. Didn't even take 30 seconds. Every heatsink I've ever used was exactly like that. Are you satisfied?

Squall Leonhart Wrote:You've never experienced a moment of intelligent thought before, have you?.

I don't know why people still using Arctic Silver 5
Arctic MX-2 is better than AS5 and MX-4 is better than MX-2

Noctua is the current second best TIM you can get.

even beats that diamond shit in some cases

@Mr "This stuff cannot be wiped off": The other day I dismantled and old pentium 4 based computer (as I felt it was more respectful than just throwing it to the tip) and the thermal paste mainly washed of in hot water. I'm assuming some of that was because it was it was so old, but not even alcohol was necessary.

some guys dont know the difference between heatspreader and cooler...
cpu(flux/tim) -> heatspreader(tim) -> cooler

If you are referring to me, then I can say that there was no heat-spreader on that pentium 4. There was, however, a blob of something non metallic, yet rock solid. I'd say it was plastic, except I doubt that plastic would be blobbed on top of the CPU die, as it isn't famous for its thermal conductivity.

Although squall usually ends up being right about these things I have to go with MaJoR on this one. I've never had issues taking off TIM on stock or aftermarket coolers using alcohol. And I have never heard of or seen someone have difficulty doing it. Perhaps there is a specific uncommon type of TIM that behaves the way squall describes?

Quote:I have to question the sanity of anyone who would think for a second that cpu cooling would have stopped using a form of TIM...ever.

I'm confused. Who implied that that we stopped using TIM for heatspreader/die->cooler contact?

Quote:Thermal Paste allows you to have a temporary connection between two metal pieces. It's not the best, but if you keep it down to just one TIM bridge, it's not that bad. The problem with having TIM connect the die to the IHS is that you are increasing the number of times thermal grease is used, and decreasing the thermal dissipation rate. THAT is why this is bad.

It has nothing to do with the fact that it is being used for two connections. It has to do with the fact that the surface area of the connection between the die/heatspreader is much lower than the surface area of the connection between the heatspreader/cooler. Thus it is very important to use solder for the die/heatspreader connection. You would get the same effect if you used a small contact base on the cooler that only covered a fraction of the surface the the heatspreader.

(04-29-2012, 10:13 PM)admin89 Wrote: [ -> ]I don't know why people still using Arctic Silver 5
Arctic MX-2 is better than AS5 and MX-4 is better than MX-2

Some people are broke 0.0

Quote:You would get the same effect if you used a small contact base on the cooler that only covered a fraction of the surface the the heatspreader.

To my understanding, it is better to have it as Die>TIM>Heatsink than to have Die>TIM>IHS>TIM>Heatsink. Sure the heatspreader helps the TIM, giving it a large surface area; but having a small to large connection, then a large to large connection... just seems like it would be better to stick with the low surface area alone, especially with TIM hindering things. But I don't really know for sure. Hmm, I should ask on Overclock.net.

Quote:I don't know why people still using Arctic Silver 5
Arctic MX-2 is better than AS5 and MX-4 is better than MX-2

Some people bought a giant tube years ago and want to use it up. Besides, it's still within a degree or two of the best stuff; there is no reason to throw it away and replace it.