Vcore measurement is low - Is it a problem?

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit,alt.comp.hardware.overclocking,alt.comp.hardware.pc-homebuilt (More info?)

I have a IC7-MAX3 motherboard and a P4 3.2C Northwood CPU. Using the
Motherboard's Voltage Measurement tool from the BIOS setup menu, I
noticed that:

Vcore measures 1.5V even though it is set (by default) to 1.55V. Now
if I adjust the setting to 1.60V then Vcore measures 1.55V. So it
appears that there is a 50mV difference between Vcore(set) and
Vcore(measured).

I am wondering what to make of this. There are a few possibilies:

1. Vcore measurement is not accurate. It is actually higher than
measured and is probably correct in actuality.

2. Vcore measurement is accurate. But due to high load, Vcore
(measured) will always be slightly lower than Vcore (setting). So what
I am observing is normal and expected. Vcore(measured) is reading what
is normally expected (50mV lower than where it is set at).

3. The Vcore regulation is off and is slightly low. That is although
Vcore is set to 1.55V, there is an -50mV "offset" in the regulation
scheme. Thus in order to compensate, making Vcore (setting) = 1.6V is
the way to go to make Vcore (measure)=1.55V which is the specified
typical Vcore needed.

So what is happening?

Thanks!

(Please respond here. Email is invalid)

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.abit,alt.comp.hardware.overclocking,alt.comp.hardware.pc-homebuilt (More info?)

Email Invalid wrote:

> I have a IC7-MAX3 motherboard and a P4 3.2C Northwood CPU. Using the
> Motherboard's Voltage Measurement tool from the BIOS setup menu, I
> noticed that:
>
> Vcore measures 1.5V even though it is set (by default) to 1.55V. Now
> if I adjust the setting to 1.60V then Vcore measures 1.55V. So it
> appears that there is a 50mV difference between Vcore(set) and
> Vcore(measured).
>
> I am wondering what to make of this. There are a few possibilies:
>
> 1. Vcore measurement is not accurate. It is actually higher than
> measured and is probably correct in actuality.
>
> 2. Vcore measurement is accurate. But due to high load, Vcore
> (measured) will always be slightly lower than Vcore (setting). So what
> I am observing is normal and expected. Vcore(measured) is reading what
> is normally expected (50mV lower than where it is set at).
>
> 3. The Vcore regulation is off and is slightly low. That is although
> Vcore is set to 1.55V, there is an -50mV "offset" in the regulation
> scheme. Thus in order to compensate, making Vcore (setting) = 1.6V is
> the way to go to make Vcore (measure)=1.55V which is the specified
> typical Vcore needed.
>
> So what is happening?
>
> Thanks!
>
> (Please respond here. Email is invalid)

The answer is 4: the Vcore regulator is generating a variable voltage
according to the Intel P4 specifications where the 'set' Vcore is the
absolute maximum, which a typical processor never actually sees, with
actual Vcore being lower depending on the current draw.

If we presume your processor is behaving 'typically' then I'd guess you're
measuring it when the processor is idle with an average current draw of
around 11.5 amps or so (give or take the tolerances of the measurement).
Bring it up to full load and Vcore should drop 150mv, or more.

'Typical' Vcore is a straight line with 0 (zero) amps being
Vcore(set)-.025V and 70 amps being Vcore(set)-.177V

 

[TomG]

Archived from groups: alt.comp.periphs.mainboard.abit,alt.comp.hardware.overclocking,alt.comp.hardware.pc-homebuilt (More info?)

shouldn't be a problem as you don't know if the Vcore is low for sure or if
the monitoring port reads low. could be either. if you had random crashes,
then I'd start by raising that voltage a tad and see if it helps.

--

Thomas Geery
Network+ certified

ftp://geerynet.d2g.com
ftp://68.98.180.8 Abit Mirror <----- Cable modem IP
This IP is dynamic so it *could* change!...
over 130,000 FTP users served!
^^^^^^^




"Email Invalid" <email_invalid@mail2world.com> wrote in message
news:8a9be796.0406281808.5d9d307@posting.google.com...
> I have a IC7-MAX3 motherboard and a P4 3.2C Northwood CPU. Using the
> Motherboard's Voltage Measurement tool from the BIOS setup menu, I
> noticed that:
>
> Vcore measures 1.5V even though it is set (by default) to 1.55V. Now
> if I adjust the setting to 1.60V then Vcore measures 1.55V. So it
> appears that there is a 50mV difference between Vcore(set) and
> Vcore(measured).
>
> I am wondering what to make of this. There are a few possibilies:
>
> 1. Vcore measurement is not accurate. It is actually higher than
> measured and is probably correct in actuality.
>
> 2. Vcore measurement is accurate. But due to high load, Vcore
> (measured) will always be slightly lower than Vcore (setting). So what
> I am observing is normal and expected. Vcore(measured) is reading what
> is normally expected (50mV lower than where it is set at).
>
> 3. The Vcore regulation is off and is slightly low. That is although
> Vcore is set to 1.55V, there is an -50mV "offset" in the regulation
> scheme. Thus in order to compensate, making Vcore (setting) = 1.6V is
> the way to go to make Vcore (measure)=1.55V which is the specified
> typical Vcore needed.
>
> So what is happening?
>
> Thanks!
>
> (Please respond here. Email is invalid)