Reboots from overheat?

R

Ripskin

2[H]4U
Joined
Jan 15, 2004
Messages
2,536
I have an Athlon 2600 non barton running at 210 x 10, on a DFI Lan party NFII Ultra B

Cooling is minly an Aquarius II kit + new Chipset cooler.


Everything has been running fine for over a month since I put in the new chipset cooler.
But recently I would get kicked out of certain games to the desktop and sometimes reboots. I watched the temp through board supplied Hardware Doctor, It didnt idle or run that much higher than normal. It is a little hotter now due to summer, but not enough to warrent a curtesy reboot I would think.

While I was doing a render and watching temps since that was the easiest way to get high temps it got to 44.5 C which seemed a little high to me, It ran fine for about 5 min and then it rebooted.

Today I refilled the system with more water and coolant, Idle temps dropped a little bit but nothing huge, around 39c. As I was rendering it hung at 43.5 for a good 10 min then rebooted. Im at a loss to whats causing this. The only thing I can think of is airbubble in the radiator, but tipping case side to side and other directions has not moved the air (if any is in there) out.

The thermal compound on the CPU cannot be the problem either, I recently made sure it was clean, just a thin layer on the core and none anywhere else, and pressed / cleaned into the block. Temps from about two to three weeks ago were about 35c on the system and 35.5 - 36c Idle. Renders got it up to about 42-44 and gaming was around the same. but no problems. Nothing changed system wise or hardware wise before the reboots started.

Thanks for any help you may be able to provide.
 
reboot isn't from overheating, it is more likely one of these possibilities:

1) Not enough Vcore for your overclock
2) Not enough Vdimm (or tight memory timings).

Try increasing Vdimm up .1V. For most people that's enough, as your FSB overclock isn't too extreme.

If that doesn't work, I would suggest booting up memtest, and seeing if your memory is getting errors--IF it is, then raise Vdimm while loosening your timings (if they're 6,2,2,2 go to 7-3-3-2.5, etc. etc.)
 
Didnt think it was from overheat...

Ive never messed with my gpu / dimm OC'ing yet and know nothing about memory timings.

I have the Vcore set to 1.75, in hardware doctor it show's that being the low limit of the status bar,(not sure if you know the program. I can post ss of what it says my power usage is for vcore / dimm) Vdimm runs at 2.60

The dimm is around the middle so I assume it was okay. The +12 is another one near the low limit, but Im not sure what that does.

I just downloaded MemTest and will run it soon as I get a chance...

Your post made me think that while its renderng instead of watching temp I should watch power and see if some fall below or above the range set. Either way I will run MemTest and check timings / Vdimm.

Thank you for the quick reply and help! I will try your advice ASAP and post.
 
MemTest Found nothing after about 30 Min of testing the unused memory.

Could the ram being in a certain slot be causing issues?
 
What about something in software?

If it has been running fine for a month and you haven't done anything to the hardware or the clock speeds of anything, my first guess would be something in the software has changed somehow.

I certainly can't say for sure that this is the case, but just want to make sure you've looked at the software side aswell.

Diazo
 
The only thing to change software wise was an install of Sim4 and an Uninstall, I needed the HDD space so took it off. That was about a week or two ago. Im not sure how that would cause reboots but It could.

Im hoping to format my drives soon so if that is the problem then it will get fixed but if not...

Ill look into any other software changes just to make sure though
 
With my system my machine will reboot once the +12v line gets below 11.80. So I'm having a hard time overclocking both my a64 and the 6800gt.
 
Would Dropping the Vcore voltage so that the proc is using the power being allocated for it help?

I guess that right now too much power is being sent toward the core than it can use? Or do i have it backwards? I dont want to take the fsb up too much more to generate more heat, but if by the same token i need to balance the power going in and consumption so that the graph appears in the middle like the other power than I will do that.
 
Ripskin said:
Would Dropping the Vcore voltage so that the proc is using the power being allocated for it help?

I guess that right now too much power is being sent toward the core than it can use? Or do i have it backwards? I dont want to take the fsb up too much more to generate more heat, but if by the same token i need to balance the power going in and consumption so that the graph appears in the middle like the other power than I will do that.
Click to expand...

Your thinking too hard... Please put back to stock specs and back away from the bios! :rolleyes:
 
Stock specks or fresh bios will leave me at 1 ghz. That is not an option and well below that a 2600 should run at.

No one was born knowing all there is about OCing, people had to learn. All I did was ask for some help, im trying to learn these things.

I had thought that 210 x 10 @1.75 would be a stable set up, and was till recently, I dont know what happend and was just asking if anyone else had some ideas or could help.
 
You must log in or register to reply here.
Back
Top