[Viperino]

Sorry if this is a little long but...

System:
MOB: Intel D875PBZ MOB
CPU: Intel Pentium 4 Processor 3.40E with 1-MB L2 Cache (Prescott?)
HSF: standard Intel boxed (will be trying a Zalman 9500)
GPU: ATI 9800XT (just added an Antec VCool PCI slot fresh air intake fan below it)
POS: 450 WATT with small fan
CAS: Aluminum Z-Alien from Cyberpowerinc (not TAC)
FAN: 80mm in front, 120mm in back

History:
A few months ago, Battlefield 2 started to reboot the system almost everytime upon beginning to start an online game. (I did not have any temperature monitoring in place until last week.)
I have continuously updated all hardware drivers, including GPU and system BIOS to the latest versions.

Last week I installed Intel Active Monitor and Intel Desktop Control Center, which includes StressTest 1.0 to perform various tests. Using these, I realized my CPU was overheating. I first checked all my fans, which seemed to be working properly. (They are not connected to the mother board, so I can not verify speed, but they are all turning) So, I called Intel and to my surprise, they sent me a new CPU and HSF. While I was waiting I cleaned my case and blew out all my other fans, including the GPU (9800XT).

The new HSF came with the thermal pad, so I installed it that way. BUMMER, same thing still. I purchased some Silver thermal compound (Fanner 700) from a local computer store and tried that. Same results...

Couple of side notes:
1. Running BF2 triggers CPU temp warning (set at 75C) in about 30 secs.
2. My system has rebooted from BF2, probably 40-50 times in last few months.
3. I have not pushed the system to reboot stage since knowing about the overheating, but believe this is same problem that started months ago.
4. I noticed my GPU fan seems slow and makes a lot of noise. So I also used ATI Catalyst utility to monitor GPU temp. (I have a ticket with ATI to verify safe temps for GPU, but have not heard back from them yet.
5. I have never OC'd.

I ordered a Zalman CNPS9500-cu, some Arctic Silver 5 thermal compound the 2 cleaning compounds. I hope to have them by Weds (12/7/2005).
I decided to do some better testing and record the results with a few comments/concerns...

Test date: 12/3/05 - before Antec VCool slot fan
IDLE: (Ambient 23C)
CPU: 52-54C (53) - slightly high but not out of order?
Zone 1: 33-35C (34) - case temps look ok?
Zone 2: 30-32C (31) - case temps look ok?
GPU: 67-68C (67) - seems high but waiting on ATI?

50% (GPU and CPU) load for 5 min:
CPU: 67C (+14) - not really good
Zone 1: 39C (+5) - case temps still ok?
Zone 2: 35C (+4) - case temps still ok?
GPU: 68C (+1) - not bad?

100% (GPU and CPU) load for 5 min:
CPU: 84C (+21) - VERY BAD - about to reboot?
Zone 1: 45C (+11) - case temps still ok?
Zone 2: 40C (+9) - case temps still ok?
GPU: 72C (+5) - not bad, not that much increase?

Since the above tests I have:
1. Added the Antec VCool slot fan.
2. Removed and verified contact of HSF. Looked like good contact, but not great. If anything I might have had too much compound on it?
3. So, I touched it up a little and reseated the HSF.
4. Ran a few more tests recording the temps every 10 secs for the first 2 mins.
5. Installed a program called throttlewatch to see if/when my CPU throttles.

Test date: 12/6/05 - after Antec VCool slot fan
All IDLE temps are lower than 12/3 tests.
Not sure what made biggest change? (ambient, reseat of HSF, or VCool)
IDLE: (Ambient 22C)
CPU: 49C (-4 from 12/3)
Zone 1: 29C (-5 from 12/3)
Zone 2: 26C (-5 from 12/3)
GPU: 61C (-6 from 12/3)

100% (GPU and CPU) load at 30 sec:
CPU: 59C (+10)
Zone 1: 30C (+1)
Zone 2: 26C (0)
GPU: 62C (+1)

100% (GPU and CPU) load at 60 sec:
CPU: 66C (+10)
Zone 1: 31C (+2)
Zone 2: 28C (+2)
GPU: 62C (+1)

100% (GPU and CPU) load at 2 min:
CPU: 73C (+21) - kicks off Intel Active monitor alert (which is reccomended for the 3.4E Prescott)
Zone 1: 35C (+6)
Zone 2: 31C (+5)
GPU: 63C (+2)

100% (GPU and CPU) load at 4 min:
CPU: 79C (+30) - throttlewatch begins to show throttling (STOPPED test - I just don't like going any higher anymore)
Zone 1: 39C (+10)
Zone 2: 35C (+9)
GPU: 63C (+2)

System has been IDLE for about 10 hours and current temps are:
CPU: 50C
Zone 1: 35C
Zone 2: 32C
GPU: 63C

This is were I sit right now.
I am hoping that the CPU/HSF/contact is the only thing that is out of order and the Zalman 9500 will fix it but...
I am not real confident that it will.
Here are a few of my comments/questions:
1. Does anyone know where exactly the Zone 1 and Zone 2 readings from Active MOnitor are located on the D875PBZ MOB?
2. Do Zone 1 and Zone 2 look to be ok? They do to me.
3. Due to the rate of increase of the CPU, I want to believe the Zalman9500/AS5 will help/fix.
4. Is there anything else besides the CPU/HSF/contact that could cause this?

If anyone has any other suggestions or comments please post.

Thanks...