p4p800se question

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

I have a P4P800SE system with a Northwood chip that has two problems.

1) if i put it into standby mode, when it wakes up AIBooster is reporting a
voltage of 3.3, double it's normal 1.6. Out of hibernation everything is
fine.

2) Every once in a while, the CPU temperature shoots up to 60c with no
apparent cause (as reported by AIBooster and AIProbe)--fans are still
running, etc. Simply restarting the box results in CPU temperature of low
30's making me wonder if the temperature reading is correct.

My BIOS is a 1004 rev, AIBooster is 1.01.07, AIProbe is 2.22.00.

Any suggestions?

Thanks in advance

 

[Paul]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <g-adnVaNYpU0qKnfRVn-oQ@comcast.com>, "truffle"
<truffle100@comcast.net> wrote:

> I have a P4P800SE system with a Northwood chip that has two problems.
>
> 1) if i put it into standby mode, when it wakes up AIBooster is reporting a
> voltage of 3.3, double it's normal 1.6. Out of hibernation everything is
> fine.
>
> 2) Every once in a while, the CPU temperature shoots up to 60c with no
> apparent cause (as reported by AIBooster and AIProbe)--fans are still
> running, etc. Simply restarting the box results in CPU temperature of low
> 30's making me wonder if the temperature reading is correct.
>
> My BIOS is a 1004 rev, AIBooster is 1.01.07, AIProbe is 2.22.00.
>
> Any suggestions?
>
> Thanks in advance

Given the opportunity, I bet that AIBooster is going to
toast your system After all, there are frequency settings
possible on the clockgen, that no hardware can hack. A human
operator wouldn't select those settings, but a piece of crazy
software could.

Seriously, why rely on such obviously well written
software, when a few manual adjustments will suffice ?

The 875/865 chipsets have ratios between FSB and memory,
so that you can raise the FSB and drop the memory clock
a corresponding amount - that keeps the memory in spec
and allows the (fixed multiplier) Pentium core speed to
be raised. If you do it a bit at a time, you'll be able to
tell whether it needs an extra 0.1V of Vcore or not.
One of my testing tools is a Knoppix read-only Linux
boot disk (because no hard drive is needed and you cannot
corrupt a CD if overclocking). A second useful tool is
Prime95, as it is a good indicator of whether your overclock
is stable or not.

(I would give exact instructions, but for the fact that
the BIOS on all the 875/865 boards doesn't work the same.
My P4C800-E BIOS lies about the memory speed, while I
think your board may tell the truth. That is why a
Windows utility is part of the overclocking effort -
to verify what is happening. For example, I had one board,
where the CAS setting in the BIOS, wasn't reflected in
what was seen in Windows later. That is important when
trying to interpret the results of each overclocking
experiment.)

If you get a copy of CPUID, you can verify what the BIOS
settings are actually doing. That is why you should only
make tiny changes at first - so you can determine how the
BIOS settings work. (On my A7N8X-E it only took about
thirty minutes of testing, to get a feel for it.)

http://www.cpuid.com/cpuz.php

The first step in overclocking, is exposing more manual
settings in the BIOS. "AI Overclock Tuner" [Manual] and
"Performance Mode" [Standard] should give you a good start.
And don't use Turbo, because it sets CAS to 2, and is only
appropriate if you own CAS2 memory. Many a person is
greeted with a black screen after selecting Turbo. But there
aren't too many other traps like that...

Paul

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

Thanks for the help. I'm starting to look into those issues now. A further
question however: since I'm running everything at the default settings, do
the symptoms still sound to you like a fault config? Again, the return from
standby, shows a doubling of cpu voltage while returning from hibernation is
ok. And the temperature spikes don't seem to have any rhyme or reason
behind them--I'm not using AIBooster to switch anything. Finally, is it
reasonable for the CPU temperature to fall almost immediately to normal
temps after a reboot?

Thanks--I'm a newbie at all this. If there are things you think I should
read and can give pointers, that would be helpful.

 

[Paul]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <Ge-dnVaS2pykx6nfRVn-oQ@comcast.com>, "truffle"
<truffle100@comcast.net> wrote:

> Thanks for the help. I'm starting to look into those issues now. A further
> question however: since I'm running everything at the default settings, do
> the symptoms still sound to you like a fault config? Again, the return from
> standby, shows a doubling of cpu voltage while returning from hibernation is
> ok. And the temperature spikes don't seem to have any rhyme or reason
> behind them--I'm not using AIBooster to switch anything. Finally, is it
> reasonable for the CPU temperature to fall almost immediately to normal
> temps after a reboot?
>
> Thanks--I'm a newbie at all this. If there are things you think I should
> read and can give pointers, that would be helpful.

As for what can cause a temperature spike:

1) Increased Vcore voltage. Power is proportional to FCV**2
where F is the frequency, C is the capacitance (a constant),
V is the voltage. Since the voltage term is squared, power
dissipation is pretty sensitive to voltage. More power means
more heat.
2) Increased core frequency, as per formula above.
3) Reduction in fan speed, causing a reduction in cooling ability.
4) Increase in case air ambient temp (not likely, would happen too
slowly).
5) Faulty contact between heatsink and CPU. Like when a clip
on the heatsink is loose.

Why not take AIBooster out of the picture, then repeat your
recovery from standby and hibernation tests again ?

Paul

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

thanks again--pulled aibooster out, still got doubling of vcore when
returning from standby, then put system into hibernation and restored and
vcore values returned to normal.

I'll leave aibooster out of the picture, see if the temperature issue
returns.

I've been reading about some issues with temperature tied to versions of the
ASUS bios but have yet to find anything on their site that discusses this.

if other ideas come to you, let me know--thanks


"Paul" <nospam@needed.com> wrote in message
news:nospam-1303051113440001@192.168.1.178...
>
> As for what can cause a temperature spike:
>
> 1) Increased Vcore voltage. Power is proportional to FCV**2
> where F is the frequency, C is the capacitance (a constant),
> V is the voltage. Since the voltage term is squared, power
> dissipation is pretty sensitive to voltage. More power means
> more heat.
> 2) Increased core frequency, as per formula above.
> 3) Reduction in fan speed, causing a reduction in cooling ability.
> 4) Increase in case air ambient temp (not likely, would happen too
> slowly).
> 5) Faulty contact between heatsink and CPU. Like when a clip
> on the heatsink is loose.
>
> Why not take AIBooster out of the picture, then repeat your
> recovery from standby and hibernation tests again ?
>
> Paul

 

[Paul]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <1omdnRbawfh766nfRVn-3w@comcast.com>, "truffle"
<truffle100@comcast.net> wrote:

> thanks again--pulled aibooster out, still got doubling of vcore when
> returning from standby, then put system into hibernation and restored and
> vcore values returned to normal.
>
> I'll leave aibooster out of the picture, see if the temperature issue
> returns.
>
> I've been reading about some issues with temperature tied to versions of the
> ASUS bios but have yet to find anything on their site that discusses this.
>
> if other ideas come to you, let me know--thanks
>

See the bottom post on this page:
http://abxzone.com/forums/showthread.php?t=71612&page=4&pp=15

I hope the uninstaller completely removes all signs of
AIBooster. You wouldn't want some leftover files messing
with it.

Paul

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

Be careful putting your system into stand by mode.
----------
"In standby the fan shuts down but when the system comes out of standby the
fan just makes a half spin, stops and does not come back on until the system
is restarted."
* http://forums.nvidia.com/index.php?showtopic=288

Graphics card related but still good to know.