• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

3500+ scalding hot

Ecchi

n00b
Joined
Aug 9, 2004
Messages
19
I've re-seated the retail HSF over and over again to no avail. I think the main problem in this situation is the server-size case, is not pushing out the hot air the cpu produces (yet the case's temp is about 18-20 Celsuis). I have 4x 80mm fans going (2 intake, 2 outtake), but I think the airflow path does nothing for the CPU in this case. Using a MSI K8N Neo2 Plat board in a Antec 1040 case. I've been playing ignorant to the "issue" for the past few weeks, by using CoolNQuiet.. yet the other day when I actually have to use "full power".. I was greeted with a lovely 72 Celsius average, with a peak of 73. Hot hot hot. It seems as though another intake fan is going to be needed (on the case-door/right around the cpu area), but I worry that even then the retail HSF will "put along" terribly like it is now. Suggestions? (aside from watercooling - I dont trust myself with water and computer parts).
 
I seen a lot of people in these forum talk about how that motherboard is horrible at reading temps even after many bios updates. As long as you are sure the heatsink is on properly and you have enough thermal paste, run something like prime95 for a while and see if your computer locks up, because there is no way a cpu could run at 72 degrees stablely.
 
MetsFan978 said:
because there is no way a cpu could run at 72 degrees stablely.

I was surprised to lookup & find that the max IHS temp for the 939 chips is 70c.
So yes, that truly way too hot to be running that processor.

I would suggest you create a duct from one of the intake fans to blow right onto the heatsink's fan - that alone SHOULD drop your temps around I'd guess 15c. If your fan mount points have a bunch of metal finger guards built in, cut them out, as they restrict a lot of airflow. If you quickly pull off the sidecover and stick your hand into the case while running, is it really warm inside?
 
As for being horrible at reading temps, I have to disagree. My NB temp is spot on (which I've verified), and the cpu temp seems about right to me ~37c at idle, ~45c 100% load. That's running with a water temp of ~27c. I'm using the original bios.
 
It's a known issue with virtually every socket 939 board. Something changed in the chips that has caused problems with temp readings. On my 3500 that is watercooled on an Abit board I can get readings as low as 14 degrees and as high as 57 degrees. I don't bother with it anymore, I know my system is stable and everything is seated properly.

Warden
 
v3rt1g0 said:
As for being horrible at reading temps, I have to disagree. My NB temp is spot on (which I've verified), and the cpu temp seems about right to me ~37c at idle, ~45c 100% load. That's running with a water temp of ~27c. I'm using the original bios.

You're the anomaly then as most people note that their temp readings are way too high with that board.
 
Unless you measure with a calibrated temp probe, its hard to tell what your system temps are. If it says 70c a good way to check is to CAREFULLY place your finger on the side of the HS. Discharge any static first by grounding yourself somewhere. If its warm or even hot, its not 70c.

I have a temp probe attached to the base of my HS on my PC, and it reads about 36-40c idle, 45-49c load depending on ambient temps (ambients are 20c to 25c in room)
 
i has a K8N Neo plat and NEO-FSR S754 that were 'supposed' to have temp problems, but mine were fine aswell.
 
Just to clarify, what utility are people using to read these "crazy tempuratures" ?
I used the nVidia system utility to check my CPU/NB temp.
The tool notes "CPU die" on the readout. Just a label, or where the reading is really coming from?
 
Back
Top