Intel 2600k Overheating

[Layback Bear]

When you over clock the memory you are over heating the cpu because the memory controler is onboard the cpu.
Yes jacking up the settings on the ram is concidered overclocking.
Here is Intels specs.
http://ark.intel.com/products/52214/

 

[bobkn]

The board lists the 2133 ram because it also accepts the Extreme Edition CPU's which have a faster memory controller.

Can you provide an example of a Sandy Bridge Socket 1155 Extreme Edition CPU?

 

[bobkn]

When you over clock the memory you are over heating the cpu because the memory controler is onboard the cpu.
Yes jacking up the settings on the ram is concidered overclocking.
Here is Intels specs.
http://ark.intel.com/products/52214/

I'm skeptical.

While clearly DDR3-2133 is outside the Intel specs, how much additional power is dissipated in the CPU? (I have no idea.)

Simple enough to test. Revert to the default settings, which will run the RAM at DDR3-1333.

 

[AlexRD]

Funnily enough, i reverted to automatic settings ( 1333 ) and the CPU still overheats when under stress (80C)

 

[Layback Bear]

No two computers are the same. Most certainly no two cpu's are the same. Applies and Oranges.
If you are using Prime 95 100% torture test at 80 C that IMHO is okay. Your chances of working your computer that hard under normal use are slim. That doesn't mean that better cooling wouldn't help. Keeping in mind that no two systems are the same. When I clock my computer a lot I gain about 10c in temp over stock (default) with 100% load to about 80 to 82C. I'm at default now because the ambient temps are very hot 32-33C. No one can tell every thing over a Forum because your computer is where you are and were not. Use your computer as you would normally and monitor your temps. If you want better temps, my best guess would be better cpu cooling. Other than fun I really haven't seen any advantage overclocking today's computers.

 

[mjf]

Funnily enough, i reverted to automatic settings ( 1333 ) and the CPU still overheats when under stress (80C)

Not good AlexRD.
Be bold try the thermal paste solution. You don't need to remove the CPU chip just the heatsink fan. Just be patient and work slowly.

To add: try to get the two surfaces totally clean as possible. Small particles such as lint will reduce the heat transfer.

 

[Maguscreed]

With a modern copper core heatsink (which is oem with the chip) the heat shouldn't even really be that bad if no thermal compound was used at all.

 

[AlexRD]

Alright, i'll first try the thermal paste solution today, then if it doesn't work i'll try an aftermarket cooler.

Will any paste do? Or is there a specific brand i should be looking out for?

 

[Layback Bear]

Their are many types and brands to read reviews. Like in post # 3, the one that has always worked for me is Artic Silver.
http://www.arcticsilver.com/as5.htm
Their are some that try many and then test. If you want to keep taking your cooler on and off for testing different ones, it's a pain. It's very important to apply thermal compound properly. To much compound will raise the temps.

 

[Maguscreed]

The truth of the matter is I don't think the brand of thermal compound is nearly so important as we make certain there is some there.

That chip shouldn't be hitting those temps especially not now that we've even reduced his ram to speed.

 

[mjf]

But (I'm sounding like a broken record) don't apply too much thermal paste and make sure your surfaces are very (lint free) clean. Coffee filter paper + high purity Isopropyl alcohol.

Of course your stock cooler is working? Have you checked the RPM.

 

[ICIT2LOL]

But (I'm sounding like a broken record) don't apply too much thermal paste and make sure your surfaces are very (lint free) clean. Coffee filter paper + high purity Isopropyl alcohol.

Of course your stock cooler is working? Have you checked the RPM.

Hum MJF I use that Arctic cleaning kit which has a very good final prep solution too as getting the old stuff off was easy and although the chip looked clean the final clean came up still showing traces of grey compound.

I did try that nano diamond stuff too (No7 mix) - very difficult to spread and it is supposed to fill in even smaller gaps than the silver. Anyway the end result was about the same temps as Arctic Silver.

But as per previous post I found the Arctic Freezer 13 Pro gave me the best returns for temps
Arctic Cooling Freezer 13 Pro CPU Cooler [AC-FRZER-13PRO] - $49.00 : PC Case Gear

 

[mjf]

I've only done this a few times myself and read a bit about it. I think some grey are probably ok as long you don't have lint type particles on the surface. In Australia Dick Smith used to sell ISO PRO which is 99.7% pure isopropyl alcohol which is what I use. I think in the US drug stores sell high purity isopropyl alcohol and it's quite cheap. I wouldn't use tissues or absorbent paper towels. The coffee filter paper is great.

The way I see it is the clean surface which when magnified is a jagged surface. The Artic Silver people recommend that once the surface is clean you apply a very light smear to the surface then wipe off as much as possible without the cleaning fluid. This fills in the "valleys". Then apply thin lines over the core positions. Heat will then spead it. The objective is to get the thinnest layer between all parts of the CPU top case and the heat sink. Too much thermal paste increases the thermal resistance. As I say this is my view and putting on great mounds of the stuff is counter productive.

 

[ICIT2LOL]

I've only done this a few times myself and read a bit about it. I think some grey are probably ok as long you don't have lint type particles on the surface. In Australia Dick Smith used to sell ISO PRO which is 99.7% pure isopropyl alcohol which is what I use. I think in the US drug stores sell high purity isopropyl alcohol and it's quite cheap. I wouldn't use tissues or absorbent paper towels. The coffee filter paper is great.

The way I see it is the clean surface which when magnified is a jagged surface. The Artic Silver people recommend that once the surface is clean you apply a very light smear to the surface then wipe off as much as possible without the cleaning fluid. This fills in the "valleys". Then apply thin lines over the core positions. Heat will then spead it. The objective is to get the thinnest layer between all parts of the CPU top case and the heat sink. Too much thermal paste increases the thermal resistance. As I say this is my view and putting on great mounds of the stuff is counter productive.

Now I think about MJF what about those cloths one can get for cleaning spectacles they are like a silky based type material and always leave my specs crystal as, and there are some ones that have a rubbery feel about them too and I couldn't agree more with the amount you put on - I found those plastic artists spatulas are the bees knees as they come in all shapes and sizes and not expensive either.
I usually spread real thin on the cooler (stock ones) and just a real thin layer on the heatsink with a tiny dot in the centre where the cooler will be, cos I really dislike that mess you have to clean off some off them a they must back the ruck up and pump it. Same goes for NB and adaptor heatsimks if you can get to them eh?