A64 3200 + AC Silener 64 TC = 70 degrees C !?!?!?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
A64 3200 + AC Silener 64 TC = 70 degrees C !?!?!?
I just installed an A64 3200 in a machine that needed the motherboard replaced. This machine was running previously with the same CPU and HS/F for a couple months.
I'm in the BIOS and I notice that the CPU temps are at 70C?!?!?! And the CPU fan says it's only spinning at about 1000RPMs.... what's up with this? I know the 64 TC is a variable speed fan, but shouldn't it be kicking it up a notch once the temps get that hot??? Is there a way to over-ride it, so it runs at full speed, cause this CPU is gonna fry in no time at that rate.
WATYF
I'm in the BIOS and I notice that the CPU temps are at 70C?!?!?! And the CPU fan says it's only spinning at about 1000RPMs.... what's up with this? I know the 64 TC is a variable speed fan, but shouldn't it be kicking it up a notch once the temps get that hot??? Is there a way to over-ride it, so it runs at full speed, cause this CPU is gonna fry in no time at that rate.
WATYF
Does it work again after it is cooled down? What kind of temps where you getting on the other board? Although unlikely to be a cause with a cpu with an integrated heat spreader, have you tried re-seating the heatsink? Perhaps it is not installed properly or maybe something got in between the heatsink & cpu...
This is the second mobo already... it's an RMA for the one that previously went bad.
Cooling down does nothing... I left it off all night and still did the same exact thing this morning.
I tried unplugging and reseating everything... I guess I can try reseating the HS/F next.
Even if it was booting, I still don't know how to get that HS/F to spin faster.
WATYF
Cooling down does nothing... I left it off all night and still did the same exact thing this morning.
I tried unplugging and reseating everything... I guess I can try reseating the HS/F next.
Even if it was booting, I still don't know how to get that HS/F to spin faster.
WATYF
NewCastles. This combination shows up to 20 degrees too much, depending on cold/warm boot. Luckily v1.4 BIOS seems to fix this issue, now it seems to fluctuate only 1-2°C between cold/warm boot.Tyrdium wrote:Hmmm... Are you using an MSI K8N Neo Platinum, by any chance? The older BIOS versions have an issue with certain Athlon64 chips (the Clawhammers, I believe) where their temperature measurements are inaccurate,
To the OP: Maybe the HSF has a bad heat sensor? From what You told it seems to be a temp-variable HSF.
Upgrading (or downgrading) the BIOS could fix the high-temp issue, as happened in my case. But since the board doesn't even boot properly, I'd RMA it... again.
Cheers,
Jan
Ttat temp is normal while booting, and will drop when it gets to windows. The CG steppings (dh7-cg) read over 20C high as stated earlier. It can run prime95 well over 85C indicated.
Any way try simplifying the installation to see if it can get through the boot process. You will have to repair your windows installation unless your old board had an nF3 chipset. (To many new parts for windows to boot)
Your psu might not have enough 12v amperage to boot it.
Come to http://forums.pcper.com/forumdisplay.php?f=46 and we will do our best to get you going.
Any way try simplifying the installation to see if it can get through the boot process. You will have to repair your windows installation unless your old board had an nF3 chipset. (To many new parts for windows to boot)
Your psu might not have enough 12v amperage to boot it.
Come to http://forums.pcper.com/forumdisplay.php?f=46 and we will do our best to get you going.
Tried a reset... still the same... idles between 60 and 70.
I've flashed the BIOS to the latest (that I know of)... 10/19. Lots of talk on the net about incorrect temps for this board, but those were from months ago and everyone said the latest BIOS fixed their issues.
It's kinda weird, because right when I start it up (after being off for hours) it immediately shows 50 or so, and just goes up from there. Maybe the temps are showing wrong... I dunno... it's just kinda annoying, cause if it really is running it at these temps, it'll die a lot sooner than I'd like it to.
WATYF
I've flashed the BIOS to the latest (that I know of)... 10/19. Lots of talk on the net about incorrect temps for this board, but those were from months ago and everyone said the latest BIOS fixed their issues.
It's kinda weird, because right when I start it up (after being off for hours) it immediately shows 50 or so, and just goes up from there. Maybe the temps are showing wrong... I dunno... it's just kinda annoying, cause if it really is running it at these temps, it'll die a lot sooner than I'd like it to.
WATYF
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
It's not well know, but the P4 actually has two temp diodes in it. One provides the temps that we're all familiar with; it's located in a corner of the die and tends to under-report core temps, although it may be a pretty close match to casing temps. Intel refers to casing temps when discussing HS C/W performance requirements, CPU temp safety and so on.
The second sensor appears to be closer to the center of the die where the temps are much higher. I recall reading that complete shutdown occurs when this sensor reaches 130C. Or thereabouts.
How is this relevant to A64?
I've examined a couple of 754 boards / A64s and found the reported CPU temps to be about in line with typicxal P4 temps-- maybe a bit lower. Recently, I obtained samples of two different Soltek A64-939 boards. With either a Winchester or Clawhammer 939 A64, on both boards, one temp sensor reads 65~70C on boot. This sensor is extremely sensitive and jumps 15~20C within a couple seconds of starting CPUBurn. In fact, it has risen as high as 90C, but at this temp, an external sensor problem wedged in against the edge of the heat spreader indicates less than 60C.
One of the Soltek boards (an SFF actually) has a second CPU sensor of some kind that appears much closer to the temps reported by the 754 boards. It is typically 12~20C lower than the other sensor readout.
My guess:
1) The A64 may also have 2 thermal diodes, with similar functions as in the P4. According to soem AMD tech docs, complete CPU reset is triggered at 125C.
2) If 1) is true, some boards may be monitoring the "wrong" temp diode.
Will try to get more info on this. If anyone else has any insights into this, please let us know.
The second sensor appears to be closer to the center of the die where the temps are much higher. I recall reading that complete shutdown occurs when this sensor reaches 130C. Or thereabouts.
How is this relevant to A64?
I've examined a couple of 754 boards / A64s and found the reported CPU temps to be about in line with typicxal P4 temps-- maybe a bit lower. Recently, I obtained samples of two different Soltek A64-939 boards. With either a Winchester or Clawhammer 939 A64, on both boards, one temp sensor reads 65~70C on boot. This sensor is extremely sensitive and jumps 15~20C within a couple seconds of starting CPUBurn. In fact, it has risen as high as 90C, but at this temp, an external sensor problem wedged in against the edge of the heat spreader indicates less than 60C.
One of the Soltek boards (an SFF actually) has a second CPU sensor of some kind that appears much closer to the temps reported by the 754 boards. It is typically 12~20C lower than the other sensor readout.
My guess:
1) The A64 may also have 2 thermal diodes, with similar functions as in the P4. According to soem AMD tech docs, complete CPU reset is triggered at 125C.
2) If 1) is true, some boards may be monitoring the "wrong" temp diode.
Will try to get more info on this. If anyone else has any insights into this, please let us know.