What causes system insablity?

OzyOly

OzyOly

OzyOly

Soldato
Joined
5 Feb 2006
Posts
3,386
Location
15,000ft.
Hey all, I haven't overclocked before now. I have my E6700 at 3.02GHz. In sp 2004 (2 instances of course) it was fine until the second run when on the last leg of the test one of the cores had 1 error.

My temp is 63C on full load after 1/2 an hour. I'm going to be water cooling the pc in 4 weeks (want to get up to 3.2GHz).

But for now, is this one error something to worry about?

Also my Ram is clocked at 906MHz (timings 4-4-4-12), I have no idea how to change the divider so don't know how to lower this speed.

Should I change my CPU clock to 2.97GHz? The temps aren't that much different.

Also, last question. What causes system instability? Is it the heat?

Oh, just the mention, I haven't touched the voltages at all yet.

Thanks guys.
 
I haven't overclocked at all so far ( but will do very soon! ). My guess is that system instability is caused simply by the components running so fast that it can't keep up? Just a guess though and i could (and probably am :-p) wrong
 
Right. Well, it is giving me another error.

Just wondering now, is it 'safe' to run my PC if it gets errors in orthos, I'm using that niow.
 
OzyOly said:
Also, last question. What causes system instability? Is it the heat?
Click to expand...

I am no expert, and only have an overclocked P4 Northwood C (old skool ;D) and overclocked memory at the same FSB, but I would say instability is cause by not enough volts to the CPU/MEM, components just not being able to run at the overclocked speed and temperature to.

Also if your getting an error during torture tests I dont think its "unsafe" it just means your system isn't stable enough.
 
It's 'safe', as in, you're not gonna break anything. It may be unstable, though, and there's a small risk of data corruption.

System instability has many causes, but in most cases the root cause is that some component (CPU, RAM, chipset etc.) is running at too high a speed in relation to its voltage. Temperature also reduces the attainable speed, but it's definitely not the main cause - a CPU can be stone-cold and still unstable if it's clocked too highly.

I would back it down to 2.97. It should eliminate the Orthos error, and you won't notice the difference.
 
Mattus said:
It's 'safe', as in, you're not gonna break anything. It may be unstable, though, and there's a small risk of data corruption.

System instability has many causes, but in most cases the root cause is that some component (CPU, RAM, chipset etc.) is running at too high a speed in relation to its voltage. Temperature also reduces the attainable speed, but it's definitely not the main cause - a CPU can be stone-cold and still unstable if it's clocked too highly.

I would back it down to 2.97. It should eliminate the Orthos error, and you won't notice the difference.
Click to expand...

Ok, lowered. Just about to run orthos again. Hope it's stable this time. Yeah I did the get error, on the torture test, but it passed the first one, it's only ever on the second round.
 
Ok it ran othorso for an hour, it had rebooted plenty of times fine last night. Turn mu PC on this morning and it wont even boot.

So cleared the Cmos and all is right again, but still, why does it do this? How do I make it stable?
 
As I previously said, you might need to increase the volts to your CPU, and your memory if your overclocking both.

When I test stability, I run a benchmark application for 24 hours. I ran Everest's "system stability test" for 24 hours when I overclocked my PC and havent had any problems with the overclock and its about 8 weeks old now.
 
Not sure abt upping the temps, the cpu is alread on 63c on full load and upping the voltage would increase this further which could result in damaging the cpu :o
 
ozyoly; Do you have the Intels stock CPU cooler on your c2d? If so I would suggest you buy a better cooler.
 
hmm it seems rather hot for a mere 400mhz overclock, do you have a fan on your scythe ninja? What about case cooling, do you have case fans?
 
I have the syth ninja mini, still 120mm fan, but slightly less fins in the heat pipe.

I have 4 case fans, all 80mm, 3 exhaust and one intake.

All at max speed. At the moment at 2.91GHz it's 43 idle and 60C load.
 
Last edited:
3 exausts, aswell as a 120mm fan on your PSU exausting? Would it be possible to change some of the exausts for intakes? It might help with your air flow, its worth a shot if its not to much hassel.

On my old case I used to have 2 80mm intake, 2 80mm exaust and a 120mm PSU exaust. New case has 2 120mm intake, 1 120mm exaust and PSU 120mm exaust - without the second 120mm intake my temps were a lot higher.
 
fobose said:
3 exausts, aswell as a 120mm fan on your PSU exausting? Would it be possible to change some of the exausts for intakes? It might help with your air flow, its worth a shot if its not to much hassel.

On my old case I used to have 2 80mm intake, 2 80mm exaust and a 120mm PSU exaust. New case has 2 120mm intake, 1 120mm exaust and PSU 120mm exaust - without the second 120mm intake my temps were a lot higher.
Click to expand...

Sure, Ill change two of them to intakes then. One intake at front (fast), 2 intakes at side (slow) and leave one outtake at the back (fast).
 
Tbh you shouldn't need to touch any CPU or chipset voltages for 3.0ghz with an e6700. Raising the fsb to 300mhz should be enough.

What's your memory voltage, and which motherboard are you using?
 
Cob said:
Tbh you shouldn't need to touch any CPU or chipset voltages for 3.0ghz with an e6700. Raising the fsb to 300mhz should be enough.

What's your memory voltage, and which motherboard are you using?
Click to expand...

Memory voltage is 2.1V and mobo is Dfi infinity 975x
 
Ok another problem, this morning my PC again refused to post.

Yesterday at my new clock speed of 2.91GHZ, the PC was boot and reboot fine, on and off all day and 3 hour orthos blend test with zero error.

This morning (like the 2.97GHz speed) the PC refused to post and I had to reset the cmos again.

Do any of you guys have an idea to what's causing this???

What is so damn different about the day time apart from it being hotter hmm???
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom