Tolerance of motherboard voltage regulators that control CPU Vcore?

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

For example, on a Gigabyte GA-EP45-UD3, what is the actual voltage
range (tolerance, plus or minus) if one sets "CPU Vcore" to the
maximum operating voltage 1.3625 for an Intel Q9550?


Re: Tolerance of motherboard voltage regulators that control CPU Vcore?

John Doe wrote:
Quoted text here. Click to load it

Because the processor has a "load line", the voltage values you
see will be all over the place. For example, at idle, you might see
a measured voltage 0.060V above where it should be (this is actually
at odds with what the Intel datasheet says). And when the CPU
runs flat out at 100% load, the voltage can drop 0.150V (ballpark
figure) below the nominally programmed value. That behavior is the
"load line", and is different than the basic "tolerance" of the
regulator chip itself. (Unfortunately, they don't go into details
in the Intel CPU datasheet, as to what their objectives are with
regard to this subject. The objective doesn't appear to be, that
the voltage value remain the same under all conditions.)

(Pg. 24 here, for your processor, shows the load line. For whatever
reason, the voltage isn't allowed to be a flat horizontal line with
respect to load current.)

This is the regulator on my Asrock LGA775 motherboard, and while this
doc doesn't answer every question you might have, it'll give you a better
idea how complicated the answer to this question can be. This regulator
has a "tolerance" of 0.5%, which is damned good, but that tolerance
is just for the DAC and reference inside the regulator - it isn't a
spec for the whole motherboard or anything. The load line kinda makes
a joke out of that 0.5% figure.


Site Timeline