Temperature of XP2600 Mobile on A7N8X-E Deluxe

A7N8XE-Deluxe, Rev 2
Barton XP2600-M at 1.65V and 1666MHz (for now ;-)
Speeze Whisper Rock II, .55C/W
Kingmax MPXC22D-38LT3R 2x512MB PC-3200 CAS 2.5
MBM 5.3.7.0
WinXP-SP1

Both the Asus Hardware Monitor and MBM indicate a CPU socket temp of 38
deg idle and 41 deg at 100%, with case temp of 18 deg indicated, but 22
deg measured with a thermometer in the case. I want to go ahead
overclocking the machine (my first attempt), but the CPU temps seem kind
of low, and I'm wondering if I can trust them. I'd like to measure the CPU
temp some other way as well, as a cross-check, but the A7N8X-E/MBM
combination won't allow CPU diode readings, as far as I know. Is there
another way to do it?

And based on your own experience, does 38 deg idle/41 deg at 70% seem
about right for this CPU?

Thanks for any/all input!

On Sun, 24 Oct 2004 14:54:57 +0000, Peregrin wrote:

A7N8XE-Deluxe, Rev 2
Barton XP2600-M at 1.65V and 1666MHz (for now ;-)
Speeze Whisper Rock II, .55C/W
Kingmax MPXC22D-38LT3R 2x512MB PC-3200 CAS 2.5
MBM 5.3.7.0
WinXP-SP1

Both the Asus Hardware Monitor and MBM indicate a CPU socket temp of 38
deg idle and 41 deg at 100%, with case temp of 18 deg indicated, but 22
deg measured with a thermometer in the case. I want to go ahead
overclocking the machine (my first attempt), but the CPU temps seem kind
of low, and I'm wondering if I can trust them. I'd like to measure the CPU
temp some other way as well, as a cross-check, but the A7N8X-E/MBM
combination won't allow CPU diode readings, as far as I know. Is there
another way to do it?

And based on your own experience, does 38 deg idle/41 deg at 70% seem
about right for this CPU?

More than anything, it's the voltage that affects the cpu temp, not so
much the clockspeed. So speed it up and leave the voltage alone. You
shouldn't see much of a difference in temps even over 2000MHz.

--
Abit KT7-Raid (KT133) Tbred B core CPU @2400MHz (24x100FSB)
http://mysite.verizon.net/res0exft/cpu.htm

As far as I know the manufacture choose not to enable reading of the CPU
diode through the on board monitoring chip on the nvidia chip set and to do
so requires modifying the motherboard. Although the cpu over protection in
the bios does monitor the cpu diode through other circuits around the cpu.
There is a program (freeware) called cpuheat that will help you determine
how the voltage and frequency increase/decrease affect the heat generated in
wattage of the cpu. A good rule of thumb is to add 10°C to the motherboard
diode temperature. From what I have read as long as the cpu motherboard
diode reading at load is around 50°C everything should be ok and that
emperors above 60°C underload can cause problems. At 41°C you still have a
little headroom.

NF7-S v2 modified Merlin bios
Athlon XP1700+ (680 A-core) at 1.875V and 1921MHz on a 173MHz FSB
TT Silent Boost with a tornado 80x80 fan rheostat controlled, 0.27C/W
Micron DDR2100 at 173MHz (346MHz) 128MB+256MB CAS 3.0 (4 3 11)
WinXP-SP2
EGA 5900SE (440/880) self modified bios

"" Peregrin wrote in message
. ..
A7N8XE-Deluxe, Rev 2
Barton XP2600-M at 1.65V and 1666MHz (for now ;-)
Speeze Whisper Rock II, .55C/W
Kingmax MPXC22D-38LT3R 2x512MB PC-3200 CAS 2.5
MBM 5.3.7.0
WinXP-SP1

Both the Asus Hardware Monitor and MBM indicate a CPU socket temp of 38
deg idle and 41 deg at 100%, with case temp of 18 deg indicated, but 22
deg measured with a thermometer in the case. I want to go ahead
overclocking the machine (my first attempt), but the CPU temps seem kind
of low, and I'm wondering if I can trust them. I'd like to measure the CPU
temp some other way as well, as a cross-check, but the A7N8X-E/MBM
combination won't allow CPU diode readings, as far as I know. Is there
another way to do it?

And based on your own experience, does 38 deg idle/41 deg at 70% seem
about right for this CPU?

Thanks for any/all input!