Which temperatue to trust?

[moltomateoxp]

I overclocked my x2 4200 to 2.454 at 1.375 volts with an arctic cooling freezer 64 pro. I get the following load temps using different software and methods:

Prime 95
Core temp: Core 1 59, Core 2 49
Asus Probe II: 52
Sandra: 53

Using DVD Shrink to get to 100% load:
Core Temp: Core 1 52 Core 2 48
Asus Probe II: 50
Sandra: 50

If core temp is accurate, core 1 gets extremely hot, more so than core 2. If the latter is accurate, my temps are good and my ac64 is doing its job. I also applied arctic silver so it may drop a few, but my question is, which of these methods produces a safe accurate temp reading?

I've done some research and what complicates the question is the extreme difference in cores using 2 instances of prime 95 using the appropriate affinities.

Thanks in advance for the help!

EDIT: Wow same topic title and general problem as another poster from 5 days ago, am I a noob or what. Will check other post's replies.

 

[moltomateoxp]

Dropped voltage to 1.35, dropped temps at load to 57 core 1, and 49 core 2. Much better. Now adding an intake fan. Also the search feature was very useful Also, load temps were from prime95 for 10 mins.

 

[sdy284]

well they're all reading the same sensor on the mobo....so they're all as accurate & trustworthy as that sensor is

 

[Mucker]

Core temp:

"The temperature readings are very accurate as the data is collected from a Digital Thermal Sensor (or DTS) which is located in each individual processing core, near the hottest part."

I would say THAT is as good as it gets..........