• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Do o/c CPUs actually degrade?

eddiew

eddiew

eddiew

Soldato
Joined
26 Jan 2007
Posts
2,541
Location
Leeds
Hey guys,

Question in the title, really. Short version is that I've had a 2500K since a few months after they launched, running at 4.5GHz on a hair under 1.3v.

Only it doesn't seem to want to run at that anymore :( Getting increasingly frequent lockups while gaming. The weather's warm of late, but temps are never above 75 outside of torture tests.

Took the clock down to 4.4 yesterday, fingers x'd that'll be enough to stabilise it again, but it's worried me that I've pushed the old girl too hard for too long and that she might be on the brink of a slow slide down towards epic failure.

Wondered if anyone else has had similar experiences, and how it worked out in the end?

Cheers
 
My i7 950 is the same. Ran it at 4ghz for ages, one day it simply just had enough. Stability just went out the window, started with a few lock ups/bsods infrequently then got progressively worse, had to back off to 3.8 with less volts and then again down to 3.6ghz when I introduced a 290 into my system (it really didnt like that for some reason). Any ways to answer your question:

Yes they do degrade at high clocks and more so @ higher volts. You don't get increased performance for free, you can and do shorten the life of a CPU by running it at the edge with higher volts for years at a time. Some CPUS will degrade more than others, some will hardly be noticeable. A lot depends on how much volts (and heat) has been through them for how long.

Just do as you have done, back off the OC and volts a bit. You probably got plenty years left on the life of the CPU-
 
anyone have a theory (or know) what degrades, I'd assume the material degrades due to thermal stress of extreme heat and / or heating / cooling cycles stressing the material??

would it be better to keep it on running OVER a specified temp to reduce thermal expansion / contraction stress?

do the materials have a temperature NOT to exceed ie running at 80degrees makes no difference running at 90 degrees significantly reduces the file of the part?
 
I had a E8500 degrade really quickly due to giving it too much voltage. I got one when they first came out and being used to 65nm cpu clocking gave it 1.5v in a attempt to get 4.5ghz stable. In the space of two weeks it went from being stable to needing more voltage at lower clocks until it would only run at stock speed with extra voltage.

At lower voltages a cpu overclock can degrade over time. People go on about heat but it's the voltage that does the damage even if running at sub zero temps.
 
Yes, generally. Though to what extent and how quickly depends on what frequency, voltage and temps you're running them at. Typically higher voltage and temps will degrade them more quickly.

I imagine a lot of the heavily OC'd Ivy and Haswells will degrade very quickly due to poor heat dissipation (no solder), especially on ones which haven't been delidded.

Also, it appears that SOI (AMD) chips degrade less quickly.
 
I know you said the temps don't look any worse but it's also worth checking the state of your fan/heatsink. If it's been on there a while there's a good chance it has a load of dust in there that could be reducing its effectiveness.
 
Qitarah said:
My i7 950 is the same. Ran it at 4ghz for ages, one day it simply just had enough. Stability just went out the window, started with a few lock ups/bsods infrequently then got progressively worse, had to back off to 3.8 with less volts and then again down to 3.6ghz when I introduced a 290 into my system (it really didnt like that for some reason). Any ways to answer your question:
Click to expand...

What rating (silver / gold / plat etc) and wattage is your PSU, and how old is it? Sounds like the (presumably) extra wattage the 290 is taking, or its pattern of consumption caused more jitter / fluctuations in load and / current on what the PSU is delivering to the CPU. If you upgrade the PSU, you may well find that you can go back up on the CPU.

A really good PSU that delivers 'clean', variance free power is another major factor in prolonging the life and function of your CPU.
 
Yer there is for sure an issues it's the PSU. it's a 650w corsair TX.

I plan on doing a system upgrade in the next few weeks anyways, with the old parts (minus the 290) going into a media server.
 
everything degrades over time

board/ram/cpu/psu/gpu

it could be any of those,it could also be bios updates/windows updates putting extra load on the cpu or needing slightly more voltage than previous

warmer weather can push a stable oc to the edge of stability aswell

I just think you need to tweak/revise your oc again due to the warmer weather
 
Cheers guys, interesting answers all round :) Though it does look like I might have to be a bit less aggressive with the overclock. Didn't think 1.3v was too much for a Sandy, but I guess it's done a lot of work in its time. I'll keep it down at 4.4 for now, see if that's enough to settle it.

Ah well, worst case I'll have an excuse to make an mATX DC build - can't justify it while my rig keeps working :P


cheesyboy said:
I'd favour degradation of motherboard or power supply causing the drop in overclocked stability, over the CPU itself.
Click to expand...

Almost certainly not the PSU, got a new 550W Seasonic about 3 months ago. Plenty for a 2500K and an nVidia 760 :) Motherboard... possible, it's as old as the CPU.

nails666 said:
I know you said the temps don't look any worse but it's also worth checking the state of your fan/heatsink. If it's been on there a while there's a good chance it has a load of dust in there that could be reducing its effectiveness.
Click to expand...

Seemed ok last time I cut myself to shreds on the heatsink (a few months back), and I have filters on the intakes so it should be pretty clean still. Will check though.

wazza300 said:
everything degrades over time

board/ram/cpu/psu/gpu

it could be any of those,it could also be bios updates/windows updates putting extra load on the cpu or needing slightly more voltage than previous
Click to expand...

PSU and GPU are both less than 6 months old, doubt they're the culprit. Not done a BIOS update, ever I think. Never had stability issues before so didn't want to rock the boat :P

Temps are fine while gaming, usually only in the high 60s at the point it crashes (2nd monitor with temp gadget). Creeps to 80 during a Prime 95, but happily ran for at that for 90 minutes last night without error.
 
Last edited:
my sandy is running on constant 1.45 vcore@5ghz from day one. So thats 27months now. And nothing changed so far. Before that i had phenom [email protected] dont remember but high constant voltage to for over 2 years to.

In my opinion you have to be unlucky to burn out cpu. And my pcs have hours and hours of benching and stability testing.
Running pc for DAYS without turning off. Rendering videos ect.

I think as llong as you keep temperatures at sensible level you are ok for at lest 2 years :) then warranty ends and time to get new stuff :D Like my Sandy never goes above 70c and phenom was below 60 if i remember.
 
Last edited:
I see the ops running 32gb of ram,could need more imc voltage or vccio

ive been running my 2600k at 1.344v for nearly two years just fine,not a fixed voltage tho,through offset
 
wazza300 said:
I see the ops running 32gb of ram,could need more imc voltage or vccio

ive been running my 2600k at 1.344v for nearly two years just fine,not a fixed voltage tho,through offset
Click to expand...


Cheers, I will see if I can find those settings when I get home tonight :)

And I have actually been using offset - under gaming load, cpu-z reports 1.288-1.296. Probably got enough headroom to give it another 0.05v without being too scared.
 
eddiew fact known for 20 years :D they build stuff to last 8-10 years. I am ok with ith burning out after 3 cause i dont have electronics tor that long. After 2 eyars you can sell it for1/3 of price after 3+ years its worth as much as scrap metal off scale.
 
eddiew said:
Interesting... So in short, either volts or heat will hasten the degradation via electromigration?
Click to expand...

Yes, rather than the process of running above factory specs, as a lot of chips are the same chip, just set to run at different speeds.

Lord_ZED said:
eddiew fact known for 20 years :D they build stuff to last 8-10 years. I am ok with ith burning out after 3 cause i dont have electronics tor that long. After 2 eyars you can sell it for1/3 of price after 3+ years its worth as much as scrap metal off scale.
Click to expand...

Q6600s held their price very well! They were going for just over 1/3 their price 5 years later! Well at least for me, I sold mine either last year, or early this year for about that.
 
You must log in or register to reply here.
Back
Top Bottom