Q6600/Ultra 120X Temp Tests

[cnumartyr]

Well here is what I promised a bit ago. Did some testing on the Q6600 with a Thermalright Ultra 120 eXtreme and Cooljag Everflow 110 CFM PWM Fan. Arctic Silver 5 was used and allowed to cure for 100 hours. I will be redoing the tests next weekend so that it has it's allowed 200 hours.

After that I will replace it with MX-2 and compare temps to see whats better for air cooling. So here are some definitions of what I used so there is no confusion. Vcore is the Voltage measured by CPU-Z under a load from Prime95v255 with 4 worker threads at 100%. The Idle temperature is measured as sitting for 15 minutes after booting into windows on the desktop with nothing running. The load temperature is averaged over 30 minutes of Prime95v255 testing. Ambient temp is measured via a K-Type Thermocouple attached to a Fluke 902 HVAC Digital Clamp Meter.

All temperatures are read as average core temps from Core Temp Beta with a 100 C Max Tjunction. My cores run within 2-3 degrees of each other so max core temp will ussually be +1-2 C over the average (ie: if Core 0/1 are at 53, Core 2/3 will be at 55 or 56). I'm also hoping that lapping will fix this.

I decided with the motherboard I was using I would not be allowing vDroop to exceed 50mV. While trying to get 3.5 GHz stable I was experiencing vDroop of around 64mV. Because of this I decided to back off for now until I can get a new motherboard with a better power supply (this has a 4 phase digital PWM).

So without further BS from me.. Here are the results on 100 Hour old Arctic Silver 5!! The second measurement with dT is the difference between the temp and the ambient temperature (delta T). All temperatures are measured in Degrees Celcius.


Arctic Silver 5 (100 Hour Cure):

Clock vCore Idle/dT Load/dT Ambient
2.4 GHz 1.248 32/8 42/18 24
2.6 GHz 1.248 32/8 46/22 24
2.8 GHz 1.248 33/8 47/22 25
3.0 GHz 1.248 33/8 48/23 25
3.2 GHz 1.264 34/9 51/26 25
3.3 GHz 1.264 34/9 52/27 25
3.4 GHz 1.312 36/11 55/30 25

Well that's that. I will be redoing them (as said) next weekend and will be adding them to this post and keeping it updated. The ultimate goal is the winner of the AS5/MX-2 Shootout on a lapped CPU/Heatsink. Hopefully by then I'll have a better board for overclocking a quad. I also just might be being a bit too careful as I don't feel like replacing the motherboard right now. Just as a reference.. I use 400x8 as my daily overclock.

 

[itotallybelieveyou]

Dude........ those are damn good temps

 

[dlikhten]

Love the numbers, though I wish that fan u got would not produce such noises... 🙁

Stock fan for duo can't reach those temps at 2ghz... 🙁

 

[CompuTronix]

cnumartyr, nice work, however, it's not clear whether you're referring to CPU temperatures or Core temperatures, hottest Core. Also, the question of accuracy, calibrations and monitoring software must be considered. Have you checked out the Core 2 Quad and Duo Temperature Guide? - http://www.tomshardware.com/forum/221745-29-core-quad-temperature-guide

Earlier this year, DaSickNinja conducted a very in depth controlled study of several different Thermal Interface Materials (TIM). Although that very popular thread (Thermal Shootout) no longer exists since Tom's upgraded the Forums several months ago, the conclusions pointed out that Masscool Shin-Etsu X23 was very clearly the best TIM available for CPU cooling, and was far superior to Arctic Silver 5 by ~4c. Artic Silver 5 actually finished near the bottom of the list, while the closest contender was Arctic Silver Ceramique ~1 to 2c behind Shin-Etsu X23.

Comp

 

[cnumartyr]

The comparison is mainly for myself and I figured I'd share it with everyone.

I've read all the guides on here numerous times from the time I started thinking about overclocking until I'd gotten into it a bit more.

Temperatures were measured with Core Temp Beta showing a Max Tjunction of 100C. The temperatures are the average of the 2 cores (it's normally right in the middle, 2 cores run at say 53 the other 2 run at 55).

I'll look into X23, might be worth it. Thanks for the input

 

[CompuTronix]

Thank you for the clarifications. Again, nice work! For everyone's benefit, I would like to offer the following regarding the software utility "Core Temp". From the Guide:

"...Section 8: Tools...

...Core Temp 0.95.4 (Beta): http://www.thecoolest.zerobrains.com/CoreTemp ...

...Note 1: Core Temp is an excellent utility, however, it has a fundamental flaw in terminology, which creates confusion in the CPU temperature community, by obscuring the distinction between temperature and specification. Core Temp shows Tjunction 85c (or 100c), which is an incorrect term. The proper expression is Tjunction Max 85c (or 100c), which is the term defined by Intel as shown above in the Specifications section, and as represented below:

Tjunction = Core temperature
Tjunction Max = Shutdown

(A) Junction Temperature is a thermal measurement because it scales, thus the term TJ, or Tjunction, which is synonymous with Core temperature.

(B) Maximum Junction Temperature is a specification because it does not scale, thus the term TJ Max, or Tjunction Max, which is synonymous with Shutdown.

(C) ~ 5c below Tjunction Max Throttling is activated. If Tjunction Max is reached, Shutdown occurs, which is either 85c or 100c, and is determined by Stepping...

I have been in touch with the author of "Core Temp", Arthur Liberman, regading this terminology issue, and he has assured me that this problem will be corrected in the next release.

Comp