Quote Originally Posted by xwards View Post
on my p5k-e wifi with 1012 bios i measured Vpll:
and at 1.5V set in bios, i get 1.56V measured
Quote Originally Posted by munim View Post
Could the people that have confirmed the voltage bug also tell us which revision of the board they have? I ask because the person that had 1.56v vpll when setting it to 1.5v in bios had the newer 1.02g revision board.
Xwards, your measurement is absolutely correct but you may still have the volt-bug Here's why:

Barbonenet had mentioned earlier that setting Vpll, Vtt/Vfsb, and Vmch to their lowest level is equivalent to setting them to Auto. What was not mentioned was that the Auto settings are FSB dependent, meaning the voltage levels are mapped to certain FSB frequencies.

Here's what I have discovered with my trusty DMM:

Vpll set to 1.5v or Auto
FSB-----Vpll Measured
333-----1.550
367-----1.654
400-----1.757
434-----1.861

Vtt/Vfsb set to 1.10 or Auto
FSB-----Vtt/Vfsb Measured
333-----1.111
423-----1.204
459-----1.293
495-----1.387
*Note: Vtt/Vfsb measurements for non-45nm chip may be higher.

Vmch set to 1.25 or Auto
FSB-----Vmch Measured
333-----1.261
351-----1.412
400-----1.564

These measurements were taken from my Rev 1.02g board with 1013 bios and 3110 Wolfdale. I suspect this volt-bug is present on all board revisions and bios versions, but please perform your own measurements and share your results.

Hope this is useful information to those perplexed by the strange behavior of these voltage adjustments.