Do All E8400s Have Faulty Temp Ratings?

[Coolio_alert]

Do All E8400s Have Faulty Temp Ratings? - I Plan To Buy One..

 

[dagger]

Do All E8400s Have Faulty Temp Ratings? - I Plan To Buy One..

How can they all be faulty? If you get one, just rma it.

 

[husky mctarflash]

From the reading I have done on the message boards, not all of them are faulty, but a huge portion are--such a huge portion that if everyone RMA'ed, there would be a lot less left in the market.

There are apparently some workarounds that let you recalibrate your temp settings. (can't reset a broken probe, but can recalibrate it if you know how off the readings are).

 

[Coolio_alert]

What would of caused these faulty temp readings?

 

[kad]

Two month back after deep search I found that E8400 chips have more than 60 errata plus the wrong temps reporting
Hope the new batches arriving will be better
I think we should not run after the first batches of any newchip hitting shelves
You all know the diffrence between the first release of Q6600 B3 stepping and the later G0 stepping

 

[MU_Engineer]

What would of caused these faulty temp readings?

The programs that read the digital thermal sensor info get a numeric value from the DTS that is *not* actually the temperature of the core. The value is sort of a measure of thermal intensity and needs to be fed into a formula along with the proper calibration offset to get the correct temperature value. The offset is called T_junction (IIRC) and it varies by CPU line. The E8000 series do not have their T_junction figures published, so the programs like coretemp have to guess at what the T_junction is. They seem to be guessing wrong as people are reporting some very wacky temperatures.