My C2D E6600 Idles at 50/47 according to TAT with a Scythe Ninja

[gamefreakgcb]

My current case is a LIAN LI PC-60BPLUSII with the Asus P5W-DH mobo and the C2D E6600 at stock with a Scythe Ninja. My temps according to TAT. Are these normal? I am thinking of changing to a more "airy" case.

 

[tylerdustin2008]

Well something is wrong. The case should have good enough flow, what is intake fan you have, and the fan on the HS? Also what thermal paste do you have applied? And you also you need to check and see if the cps's IHS is bowed.

 

[Ika]

You must have some really high ambients or something, because my load temps on a Big Typhoon barely pass that (55ºC full Orthos load @ 3.2ghz). That case shouldn't be too bad for airflow, though...

 

[gamefreakgcb]

According to Intel TAT? I heard that TAT will report actual die temps and the other apps report IHS temps or something. Ill be taking apart the entire box and rebuilding it. (I am not looking forward to it)

 

[Ozoneman]

Originally posted by: gamefreakgcb
According to Intel TAT? I heard that TAT will report actual die temps and the other apps report IHS temps or something. Ill be taking apart the entire box and rebuilding it. (I am not looking forward to it)

It's hard to say which program will measures the cpu temperature correctly. I have four programs open now and this is how they read.

TAT 21C on both cores.
Core temp 0.95.4: 36C and 35C.
Speed fan: 20C and 20C.
Everest: 21C and 20C.

I've been told that Speed fan reads 15C low. That would be true if Coretemp is correct.

Anyway, if you are getting nearly 50C with TAT, then you probably have a problem. With the results I get, your 50C TAT would be about 65C with CoreTemp.

 

[gamefreakgcb]

OK a complete rebuild is up then.

Core temp is also displaying 50/47 as well BTW

 

[Ozoneman]

Which coretemp version are you using?

 

[gamefreakgcb]

0.95.4

 

[Ozoneman]

Originally posted by: gamefreakgcb
0.95.4

That's weird that my coretemp (same version) reads different than the other programs. But I don't believe the other programs because those temps measured were below the room temperature. There's no way my cpu can be cooler than the room air.

 

[gamefreakgcb]

YES!!!! A little cleaning in the case and temps dropped by a few C's, i am waiting for the AS5 to arrive so I can realign the heatsink, think I need to remove and reapply the thermal goop?

TAT also reports 40/40

 

[BonzaiDuck]

Es-cuse, please -- hon-aber BonzaiDuck wish to conferl on this mattel.

I have the nVidia chipset, and TAT won't run on it. My understanding was this: TAT generates its own stress test. My understanding could be wrong, however, a THG article notes that -- as a rule of thumb -- ORTHOS testing produces approximately 80% of the thermal stress and loading under TAT.

For sure -- I know that CoreTemp only measures temperatures. But you'd expect higher temperatures from TAT than you'd get with ORTHOS + CoreTemp.

I'd still encourage you to do as you plan.

 

[gamefreakgcb]

Originally posted by: BonzaiDuck
Es-cuse, please -- hon-aber BonzaiDuck wish to conferl on this mattel.

I have the nVidia chipset, and TAT won't run on it. My understanding was this: TAT generates its own stress test. My understanding could be wrong, however, a THG article notes that -- as a rule of thumb -- ORTHOS testing produces approximately 80% of the thermal stress and loading under TAT.

For sure -- I know that CoreTemp only measures temperatures. But you'd expect higher temperatures from TAT than you'd get with ORTHOS + CoreTemp.

I'd still encourage you to do as you plan.

So I should re-seat my HSF than? Also torn between AS5, Arctic Cooling MX2 or the IC Diamond (if it becomes available)

 

[BonzaiDuck]

There's nothing wrong with a 3-C-degree "spread" between the cores.

HOW- -- ev--er!!

If the E6600 is not overclocked, you should find the core temperatures closer to 25C at a room ambient close to 75F, so something is definitely "not right."

And it's "not right" given that you have the Ninja cooler. The Ninja comes in just behind the TR SI-120 on the thermal resistance benches. The Ninja actually surpasses the SI-120 at low fan rpms -- around 800 rpm. But I am about to make a very important point here -- at those fan speeds, the thermal resistance of both coolers is somewhere between 0.17 C/W and 0.20 C/W. With these higher thermal resistance specs, the idle-to-load spread for processor temperatures would be greater. And FURTHERMORE -- the idle temperatures would be noticeably higher -- I just don't know by how much. You could check the benches done by Joe Citarella at Overclockers.com on both coolers -- the thermal wattage should be specified in those reviews -- and comparing the review thermal wattage to that of your E6600, it should give you an idea about where those temperatures would fall.

BUT!! I mean "BUT" -- DO YOU . . . . HAVE A FAN . . . . ON THAT NINJA? Because my little dissertation above implies something about temperatures if you don't.

As for the thermal paste -- you have more important sources of high temperatures to resolve before you move on to that, but for $5 or $6 at HeatsinkFactory and enough of the stuff for about one graphics processor and your CPU plus some extra, the IC Diamond still seems worth it to me. It should be worth 2C degrees lower temperatures across the board (pun).

 

[gamefreakgcb]

There is a Yate Loon spinning at 1300RPM on the Ninja, what I gathered from my deduction was that I had more fans exhausting than for intake (120MM intake vs. 120mm+80mm+PSU exhaust), I just added another 120MM on the side intake such that half of it blows under the heatsink above the 7900GTX and the other half of it is blowing under the 7900GTX (cant do anything about it), Ill try to post a pic once I get back home tonight. Plus my mobo dropped 8C's after that fan was put there according to speedfan from 50 to 42.

 

[BonzaiDuck]

I'm a promoter for pressurizing your case and ducting your motherboard. You tell me enough about your progress to entrench my convictions even more.