• 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.

Who has damaged/broken a CPU from overclocking?

Ive seen several accounts of e8*00 chips degrading above 1.45V, though I believe only on air.

Black's equation

29115js.png


MTTF mean time to failure, from electromigration specifically
j is the current density
T is the absolute temperature in K
w is the width of the metal wire

The other letters are constants found empirically for a given processor. So this will not tell you how long the chip will last, however it does show some relationships. Increases in current density will decrease lifespan, as current density increases with voltage, voltage will decrease lifespan.

W is the 45nm/32nm etc, decreasing this reduces time before failure. Hence 1.5V fine for a q6600 but a bit risky for a e8*00.

Note the exponential temperature dependence. This is the 'strongest' term in the equation which suggests heat is the biggest risk. Hence 1.7V under phase doesn't seem to matter all that much.

The only other thing I have to add is that at voltage V, the D0 stepping x58 i7 draws more current than the C0 stepping, so the D0 stepping chip will suffer more from excessive current than the C0. It may be negligible in both cases.

This doesn't tell you how hard a processor can be pushed though :)
 
That's quite interesting JonJ678! :)

I'm pretty kak at maths so formulas go right over my head heh :o

One thing I see mentioned in your post which I believe is true is the load temps being the thing that can cause problems, having a CPU with high vCore pulsing through it running with substandard cooling at 90°C 24/7 over long periods isn't gonna do the silcon any favours! :D

If I was rich I would have several testbeds set-up with some chips running on ridiculously high voltages for several weeks/months just to see what happened, mad scientist style! ;)

Still hoping we get some posts from people who killed a chip from overclocking, I think in the absence of anyone having a big problem from added voltage we can conclude that a lot of the info floating around the inter-web is mainly second hand smoke kinda?
 
Last edited:
Haven't killed any CPUs personally - but I do know a couple of people who have killed 65nm core 2 CPUs (which I generally consider almost indestructable) by putting too much on the vtt/pll voltage. I've run some 1.6+ vcores (watercooled) for long spells without killing the CPU as well... one of my E6600 was running 24x7 @ 3.825gig 1.65vcore watercooled doing folding, etc. for ~2 years and is still working fine at 3gig in my media PC today.

I've had some RAM die (D9GMH) from using 2.1+v for prolonged periods... one of my g.skills 6400HZ kits lasted just under a year at 2.28v 1333MHz CAS5 stable.

Had a few motherboards die as well... killed 3x DFI 975X/G boards from the same overclocking related BIOS bug :( and quite a few Asus P5 series boards didn't make the grade when OC'd.
 
Big.Wayne said:
That's quite interesting JonJ678! :)

I'm pretty kak at maths so formulas go right over my head heh :o

One thing I see mentioned in your post which I believe is true is the load temps being the thing that can cause problems, having a CPU with high vCore pulsing through it running with substandard cooling at 90°C 24/7 over long periods isn't gonna do the silcon any favours! :D

If I was rich I would have several testbeds set-up with some chips running on ridiculously high voltages for several weeks/months just to see what happened, mad scientist style! ;)

Still hoping we get some posts from people who killed a chip from overclocking, I think in the absence of anyone having a big problem from added voltage we can conclude that a lot of the info floating around the inter-web is mainly second hand smoke kinda?
Click to expand...

I thought I killed a 6400 when using Phase.

I had crazy voltages through it, the max the mobo would allow and it was running at 4.5ghz IIRC stock speed of 2.13ghz:D

All of a sudden my PC wouldn't boot.

was convinced I had killed the chip.

I gave it to a mate to test and Lo and behold there was nothing wrong with it.:D
 
Couple of weeks ago I would never of though of overclocking. The reason was because I thought it was extremely complicated and could easily go wrong.

I also thought that if it did go wrong I would have wasted a lot of money.

Once I started reading on here and articles I realized that most of the time they have safety functions and also that it is not that hard to overclock.

On that bases I would have thought that people who say your quotes Big Wayne are people who can actually overclock, but have not been brave enough to push it far enough to do damage. For example, it is logic that constantly using high revs does not do a car engine any good.
 
I've not had any CPUs die on me through overclocking but I have had one video card die. I would bet money that my x1900 died from overvolting. I think I was just unlucky when my 3870 died.
 
Last edited:
I've toasted a few XP/Athlons in my time but all my own fault and all were disposable at the time due to age/circumstance. I still have 10+ y/o box's running over clocked and over volted without issue but the tolerance on modern hardware are significantly tighter.
 
None die on me todate, but i am interested to hear more on that E8600 though, as iv pumped 1.55v through my E8400 @ 4.5ghz and its still alive, was only for few mins though, boot up, run superpi, screen shot, save, then reboot and back to normal.
 
I think the only CPU I've killed by overclocking was my P4 1.6A @ 2.6ghz. It lasted about 9 months I think, the RMA replacement (P4 1.8A @ 2.7ghz) is still going strong at my dad's house 6.5 years later.

I did also kill a Cyrix 6x86 PR200+, but that was as much because the cpu fan kept failing as my tweaking.

Generally speaking most hardcore enthusiasts that are hammering their chips with huge voltages etc are usually the types who upgrade frequently anyway. It really makes no odds to them if they reduce the lifespan of the cpu from say 10 years to 3 years.
 
ive had an old 3200xp barton break on me, had it at 3ghz for about a week with 1.8v

started to smell and motherboard components glowed white.

removed the heatsink and the chip had a big crack down it
 
HangTime said:
Generally speaking most hardcore enthusiasts that are hammering their chips with huge voltages etc are usually the types who upgrade frequently anyway. It really makes no odds to them if they reduce the lifespan of the cpu from say 10 years to 3 years.
Click to expand...

Or, from 10 years to 1 year perhaps

I have a rule of thumb - overvolt a CPU by the lesser of 5% or .05v
But I like my CPUs to last 5-6 yrs minimum as they are relegated to lesser roles and passed to family. You'd never catch me putting anywhere near 1.5v through a Q6600 :rolleyes:
 
I killed a couple of 900MHz Celerons a very long time ago. In many ways it was a mercy killing....
 
I've never killed a cpu from overclocking it, but i have killed 2 during the process of clocking them. I chipped the core of an XP2500 Mobile mounting the evap onto it, then did exactly the same thing mounting the evap onto a "naked" Opty 170.

Needless to say, all my cpu's since then have had the IHS's left firmly in place :D
 
Last edited:
I degraded my first E8500 C0 with 1.52v. It had nothing to do with temps as it only hit 52 degrees at max load. It also had nothing to do with PLL voltages as they were only set to the lowest setting +1 notch. Max stable clock was originally 4.2Ghz with 1.52v 24 hours prime stable. Then it would'nt boot and after some playing would only clock to 4Ghz. Things got progressively worse until it would only do 3.4Ghz. At this point it was ripped out and replaced with my old trusty E6600. I tested the E8500 in another board and it was perfectly fine at stock settings so eventually sold it on the bay to a guy who was on his 5th or 6th wolfdale in a week. I told him about the degrading and explained about what had happened to it and he still wanted it.
 
You must log in or register to reply here.
Back
Top Bottom