Wrong frequency with 0.99.4 and 0.99.5 beta

[imposter]

We have been getting a few reports of incorrect frequency detection with Core Temp versions 0.99.4 and 0.99.5 beta. The Coolest is aware of the situation and has it on his top priority to finding a solution. A fix is coming soon

Also:


For Intel users there is an alternative clock detection method that can be activated in settings. This may a work as a temporary fix.

Go to tools --> settings --> and check "use alt. frequency detection (if avail.)"


If CnQ or EIST are turned on, Core Temp's frequency detection might not work properly.

This thread will be updated when public beta is released.

Thanks

[The Coolest]

Fixed beta is here:
http://www.alcpu.com/forums/viewtopic.php?p=2401#2401
Please read the description before using it.

[CoreTemp-User218]

Hi,

CoreTemp 0.99.5.20 reports different speeds for the two processors inside my Q9400 rev. R0 (using F3 -System information). For two cores the readings are OK 2667 (333*8 ) but for the other two I have 2000 (333*6 ). It doesn't matter if is idle or full load the readings are the same. "On the fly bus clk detection" is activated and Intel Speed Step is disabled.
The cores of the second CPU have a lower multiplier 6 instead of 8 ??

I have tried Core temp on two other machines (E 6400 and Q9550) and the readings are OK.

Now I am not sure which is the real speed since Everest and Sandra report 2666 each for core !?! Do yo have an idea why this is happening ? I don't think is possible to have different multipliers.

P.S. MB ASUS P5W-DH, i975x, bios 2801 (last version available)

P.P.S Real Temp v.3.0 reports my CPU: Q9400 2334 MHz 333 * 7 Is getting funny ...maybe I will found some program reporting a Phenom inside

Update:

I have changed the FSB to 350 and now I get 2667 (350* 7.5) for two cores and 2000 (350 * 5.5) for the other two. Without changing anything else....so there must be a problem while reading the multipliers.