[citation][nom]jimmysmitty[/nom]How could it be more accurate when the thermal spec for a E6600 is 60.1c? If it did hit 90c it would have started to throttle which he said it didn't.[/citation]
Thermal spec (for desktop and server processors, not mobile) refers to Tcase Max, not Tjunction Max. Have a look
here. Setting a 70C Tj Max resulted in temperatures reportedly well below ambient on air. If air could do that we wouldn't have much need for water now would we
The problem with the August and October presentations is they don't seem to be consistent about this Target Tjunction. In the August IDF presentation, they had the following mentioned on the first slide about Core i7:
• Approximate temperature in °C can be calculated by:
–TJ = TJ_TARGET –DTS value
–Ex: TJ_TARGET = 85, DTS = 24, Approximate TJ ≈ 85-24 ≈ 61 °C
This indicates that Target Tj is Tj Max. In the October presentation, which was basically an extension of the August one, this was missing. The same basic logic above is used to in Core Temp to get the temperatures shown in the screenshot. If you are getting temps below ambient on air, then Tj Max must be higher than 70C on a B2 stepping CPU to make sure that you aren't breaking the laws of physics.
So, jkflipflop, that is why we can't necessarily take Intel's word for Tj Target values; at least we can't assume they are Tj Max even though Intel provides indications that they are. I am not saying that Intel is lying about these figures, but they certainly aren't clear about their purpose, or even why they were included in the presentation.