Overclock.net banner

How come no one talks about CPU degradation? My 920 D0 is degrading.

14K views 177 replies 84 participants last post by  i7monkey 
#1 ·
I always knew that CPU's have a certain lifespan, but I never knew they were that short. I've had my 920 D0 for a year and a half, and it's already showing signs of degradation, so I'm temporarily running it at stock speeds now. I never ran it high voltages (1.27V) and now it's degrading. But I did stress test it a lot, and I think that's why.

How come no one here really talks about this? Seems like everyone goes on their marry ways and overclocks their machine to hell and back without even thinking about the consequences. Many people even have their vcores set much much higher than mine's, but what are you gonna do, every chip's different I guess.

This @$%@$% sucks. I spent a lot of money on a loop and a new case, and I now I'm afraid to overclock it cause it's gonna help speed up my CPU's already imminent death.

Intel has a 3 year warranty on their chips. That means that it's guaranteed to have error-free work for the next 3 years, assuming that it hasn't been overclocked. What's overclocking do? Kill that time in half? What's the point of overclocking if our chip's are gonna degrade or die off in a year and a half or two? What's the damn point? I thought I was gonna be able to run my chip at 4Ghz 24/7 for at least the next 5 years, but that's not happening anymore.

I always knew about CPU degradation and longevity in the back of my mind, but I never knew the consequences of even moderate overclocking were that bad. This is warning to the other relative noobs like me around here, don't expect your CPU to last a while if you overclock, even if it's a moderate voltage and you have watercooling.

This sucks!
 
See less See more
1
#3 ·
I'd say it's made pretty obvious in ANY overclocking guide or discussion that what we do has risks, including damage to hardware and voiding of warranty.

When you start moving parts out of spec, you accept the risks involved. If you didn't have a benefit from overclocking, you should have moved back to stock if you can't accept risk.

It may sound blunt, but nobody is going to feel sorry for you. We may be sorry you had a failure, but nobody here should pretend that you can run hardware out of spec without accepting risk.
 
#5 ·
Well honestly lots of people DO talk about CPU degradation, but most don't go that quickly. Yours could have to do with a lot of factors, and could even be something like the board capacitors degrading, who knows? Pushing a chip so much faster than it was meant to run will have downsides, but most people that do push their chips that far don't really worry about it because they don't use the hardware more than a couple years.

I had a C2Q a few years ago that I pushed to the limits, but never worried about it, because within 2 years I got rid of it. I will probably change out my current processor in another year or so, no biggie. Just remember, when you push hardware, it always has consequences.
 
#7 ·
Quote:


It may sound blunt, but nobody is going to feel sorry for you. We may be sorry you had a failure, but nobody here should pretend that you can run hardware out of spec without accepting risk.

I'm not asking for sympathy, and I don't need it.

And I understood that overclocking comes with risk, but this soon???? With such a mild voltage???? Ridiculous.

*Waits for "I don't care your'e a clueless noob!" comments*
 
#11 ·
Quote:
Originally Posted by i7monkey;12150478
I'm not asking for sympathy, and I don't need it.

And I understood that overclocking comes with risk, but this soon???? With such a mild voltage???? Ridiculous.

*Waits for "I don't care your'e a clueless noob!" comments*
Are you 100% sure it's your CPU that is degrading?

Try testing your RAM with memtest.
 
#14 ·
The Q6600 I have has been running at 3.4Ghz @ 1.5v for the past 3 years. It's still perfect.

You just got unlucky.
 
#15 ·
ive had my 920 D0 running at 4.2ghz 24/7(maybe shut off for a total of 2 days since ive owned it) @ 1.38v for about 2 years. its been watercooled its whole life. and never seen temps above 50*c stress testing(except when applying indigo extreme... it saww 99*c) i think that its blown out of proportion. i dont know if watercooling the board and processor has helped its life by very much. but i really dont see it dying anytime soon. i feel like the board is more likely to degrade then your processor. i still have socket a processor that could still run overclocked today.
 
#16 ·
Quote:


Originally Posted by Strider_2001
View Post

Its like a sports car....You cant have performance of twin turbos and NOS along with relability at the same time...Gotta pay to play...

false you just need to start with an unlimited budget and get everything spec'ed out right!
 
#17 ·
Quote:


Originally Posted by i7monkey
View Post

I'm not asking for sympathy, and I don't need it.

And I understood that overclocking comes with risk, but this soon???? With such a mild voltage???? Ridiculous.

*Waits for "I don't care your'e a clueless noob!" comments*

Send it back to Intel and play stupid...Works like a charm...
 
#19 ·
I originally ran a battery of stress tests, from Linx, to P95, to memtest86+ for many hours and they all turned out fine with a vcore of 1.2625, but over the past month or two I started getting errors in Linx after an hour, so I bumped it to 1.27 and it ran fine. Then it got errors with 1.27 so I had to bump it to 1.28, and now it won't run with 1.28 without getting errors.

I ran memtest86+ yesterday for 17 hours and 11 passes and it came out with ZERO errors.

I've had the same bios settings and no hardware has been added since (unless you count switching monitors).

I'm not blaming anyone here, and you guys are really helpful group of guys (and girls), but I'm really stunned how soon this could happen. More people should know about this, and not just "oh...if you overclock your CPU might last only 10 years instead of 20", it should be known that a year and half or two years down the line your cpu might not work as good anymore.
 
#20 ·
For those that are interested, the original thread started here. Now dont blame me for putting those "cpu degradation" ideas in his head.
tongue.gif


Link


Also, OP has tried memtest. Here is the post he made recently.
Quote:
Originally Posted by i7monkey;12138513
Here's the test I did yesterday, ran for 17 hours with no errors. Vcore: 1.2625, QPI/DRAM: 1.2625, DRAM Bus: 1.64.

I actually lowered the the voltages to the way they were before I got these errors and memtest came up error free, so I guess this rules out the ran and I should concentrate on CPU stability with Linx.

edit: And why is Memtest reporting the wrong RAM? I physically checked the ram chips and they're the ones listed in my sig.

memtest12625.jpg
 
#21 ·
Quote:


Originally Posted by i7monkey
View Post

I originally ran a battery of stress tests, from Linx, to P95, to memtest86+ for many hours and they all turned out fine with a vcore of 1.2625, but over the past month or two I started getting errors in Linx after an hour, so I bumped it to 1.27 and it ran fine. Then it got errors with 1.27 so I had to bump it to 1.28, and now it won't run with 1.28 without getting errors.

I ran memtest86+ yesterday for 17 hours and 11 passes and it came out with ZERO errors.

I've had the same bios settings and no hardware has been added since (unless you count switching monitors).

Just pump it to 1.3v.

Otherwise send it back to Intel. From what I saw somewhere on here Intel has no way of knowing if the chip was overclocked since overclocks are stored in the BIOS on your motherboard.
 
#22 ·
i SERIOUSLY doubt a i7 d0 that was perfect to begin with degraded in 18 months at 1.28v.

Either a defective chip from the start... or your running with with a stock heatsink at f@h loads 24/7 since you bought it.

My Q9550 was at 4.0-4.1 for most of its life in the 1.33-1.35 range... i occasionally let it top out at around 65C and that thing is as stable as it was when I bought it over 2 years ago.
 
#23 ·
I would be honestly surprised if this is actually CPU degradation and not some other component malfunctioning. In my experience Intel build really quality chips that can take a ******ed amount of abuse.
 
#24 ·
Instead of bumping it to maintain your current overclock, have you tried downclocking it to see just what speed the current Vcore (1.27-1.28 whatever) will hold?

Also, what speed are you running your memory at? Running 12gb of DDR3 overclocked would put a stress on your IMC as well. I can't speak for everyone, but if I was running bench tests and stressing my rig like you, I would not be running 12gb of memory.

Just my .02
 
#25 ·
Quote:


Originally Posted by i7monkey
View Post

I originally ran a battery of stress tests, from Linx, to P95, to memtest86+ for many hours and they all turned out fine with a vcore of 1.2625, but over the past month or two I started getting errors in Linx after an hour, so I bumped it to 1.27 and it ran fine. Then it got errors with 1.27 so I had to bump it to 1.28, and now it won't run with 1.28 without getting errors.

I ran memtest86+ yesterday for 17 hours and 11 passes and it came out with ZERO errors.

I've had the same bios settings and no hardware has been added since (unless you count switching monitors).

I'm not blaming anyone here, and you guys are really helpful group of guys (and girls), but I'm really stunned how soon this could happen. More people should know about this, and not just "oh...if you overclock your CPU might last only 10 years instead of 20", it should be known that a year and half or two years down the line your cpu might not work as good anymore.

Know how to check capacitors?, cause that's kinda what it sounds like is your problem.

Check for caps that have a rounded head like they popped.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top