DescentBB

On Friday I went out and did a budget upgrade. I spent $200 at got a Barton 2800+, a Shuttle nForce2 mobo, and two 256mb sticks of PC2700 RAM. This is replacing a 2100+/MSI VIA266/512mb PC2100.

I have adjustable speed case fans, and my CPU fan is adjustable. However, when I try to adjust the speed, the computer clicks off with some sort of CPU overheat prevention feature. A red light goes on on the mobo, and you're forced to pull the power plug until it goes off to turn the computer back on. I can't even touch any of the speed adjustment dials without the computer clicking off. I pulled everything apart expecting to find something shorting, but when i put it all back together, it still does it. Has anyone seen anything like this? This makes it impossible to adjust fan speeds since I can only do it when they're off, and can't tell how fast or loud they are going.

My other issue is with CPU temperature. At idle, it runs at about 35c at it's current fan speed. Under full load, it goes up to about 45c. This is about the same as my old 2100+. I noticed today, however, that when I run winamp(or wmp9), the CPU temp shoots up to full load temps with only 2-4% CPU utilization. My old CPU never did this, though I was using an SBLive before, now I'm using the onboard audio. I would think that I'd see more CPU load for the heat to rise like that. That doesn't seem normal to me.

Any ideas on either of these two issues?

*edit, I stuck my SBLive back in for shits and giggles. The temp shoots up to full load temps, but then when I quit winamp, it doesn't settle back down to idle temps like it does with the onboard audio. Strange.

what exactly are you using to measure your CPU temps with?

A program called speedfan. It puts the temp in the system tray. I've checked it against the temps shown in the bios, and its within a degree or two.

bleh.. those programs are usually off by ten to fifteen degrees.

I'd get a seperate temp. probe with LCD readout.

Whether or not the CPU temp is actually 45c doesn't really matter. It seems to record the fluctuations just fine. What I'm asking is why playing an mp3 with 2-4% CPU utilization causes the CPU temp to rise to the same temp as 100% CPU utilization.

improper application of thermal transfer compound?

I suspect your motherboard may have hardware monitoring which has lower limits for CPU fan speed. If CPU fan-speed drops below a certain number of RPMs, then it'll power-off the mobo immediately.

Have a good dig around inside your BIOS, and see if you can disable the power-off, or see if you can drop the number of REVs required for the poweroff to function.

flip, thats a definite maybe. Just in case, I just reapplied it specifically following the instructions. I'll give it a couple days for the arctic silver to do it's stuff.

On the short problem, I tried pulling it all apart again, and it seems to have fixed it. I couldn't tell you what it was, but I can now adjust fan speeds without the computer cutting out.

Thanks.

If you're using Arctic Silver 5, you need to let it go through several heat/cool cycles for best performance.

http://www.arcticsilver.com/as5.htm

Due to the unique shape and sizes of the particles in Arctic Silver 5's conductive matrix, it will take a up to 200 hours and several thermal cycles to achieve maximum particle to particle thermal conduction and for the heatsink to CPU interface to reach maximum conductivity. (This period will be longer in a system without a fan on the heatsink or with a low speed fan on the heatsink.) On systems measuring actual internal core temperatures via the CPU's internal diode, the measured temperature will often drop 2C to 5C over this "break-in" period. This break-in will occur during the normal use of the computer as long as the computer is turned off from time to time and the interface is allowed to cool to room temperature. Once the break-in is complete, the computer can be left on if desired.

heh.. interesting fluke...