-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
1.6v long-term not bad for E8400s?
- Thread starter VirtualLarry
- Start date
error8
Diamond Member
Probably around 567-632 days.
Really now, who do you think is able to respond to that question? I'd estimate that 1 year will survive for sure, maybe two years . After that, nobody will ever know.
Usually a very few people keep an overclocked cpu, more then two years. After that, it becomes obsolete and they'll buy another one. Try it and see how long it lasts and when it dies, make a thread here so everyone will know. 😉
That thread is about vtt, but the post is about vcore. two different things that have different thresholds. But with either one at 1.6v a 45nm wolfdale will not last very long.
1.6 vcore, 6-12 months imo before you notice serious damage (eg. prime errors at stock).
1.6 vtt will eventually kill the cpu on a cold boot imo, where as 1.6 vcore could endure longer.
Kraeoss, 65nm can take a lot more than 45nm. 1.55 vcore is the absolute max on 65nm, while it's 1.45v on 45nm, absolute max being already out of safe margins.
1.6 vcore, 6-12 months imo before you notice serious damage (eg. prime errors at stock).
1.6 vtt will eventually kill the cpu on a cold boot imo, where as 1.6 vcore could endure longer.
Kraeoss, 65nm can take a lot more than 45nm. 1.55 vcore is the absolute max on 65nm, while it's 1.45v on 45nm, absolute max being already out of safe margins.
Tempered81
Diamond Member
I've ran about 1.45v real for over a year on two e8400s. One E0 for about 4 weeks, and one C0 since Jan 08
Vtt is the real killer. (also you don't want extreme vcore ~1.6ish if you cannot keep the cpu cool).
Even with the best aftermarket air coolers ultra 120, and core contact freezer, once i go above 1.45ish, I cannot keep the cpu cool, no matter the mhz.
Vtt is the real killer. (also you don't want extreme vcore ~1.6ish if you cannot keep the cpu cool).
Even with the best aftermarket air coolers ultra 120, and core contact freezer, once i go above 1.45ish, I cannot keep the cpu cool, no matter the mhz.
Could be sample variability. Aigo got the weak chip that was going to die an early death even if it were at stock, this other dude got the cream of the crop +3-sigma outlier portion of the tail and his chips would last 10yrs even at 2V if he pushed it.
We can't really use statistics like this when speaking to a distribution's characteristics (average, stddev, etc). Only thing we can do is take it for what it is worth.
Aigo has demonstrated a chip can/will die on as little as 1.57Vcore under constant load. Yours may die at 1.47Vcore under identical conditions, or might live for a decade. But if you are an "err on the side caution" kind of individual you would take Aigo's datum point and set that as an upper limit to what you will expose your CPU to.
It is personality driven as to what you do with the data when limited statistics are involved.
E2200 is not a 45nm.
I can not KILL a E2180 no matter how hard i tried. (you give too much voltages it wont boot, you try to overclock too high, it wont boot).
I had an allendale up at 1.6V and still couldnt kill it.
the E8400 which is the first release of the wolfdale processor i ended up killing.
I got the processor for almost free, on the pure speculation of see how far and how long i could push it.
It was loaded on Prime95 for 1 month duration. In real numbers i dont think any of you guys will load up your processor that hard and long.
But the day i shut it off, and went to lower the overclock as i was throwing in the towel, the cpu refused to boot up afterwards and was dead.
No not really, intel gives you guidelines, so just follow them. They tested a far larger sample and for much longer than any of us could.
For 45nm Wolfdales: 1.3625V maximum VID, 1.45V absolute maximum.
Maximum VID is the voltage at which the cpu will operate safely without compromising long term reliability.
Absolute maximum is the voltage at which the cpu will operate safely, but compromising long term reliability.
Anything above: voltage at which the cpu will not operate safely and will compromise long term reliability.
So you see it's simple. Step 1 if you want to preserve your investment, Step 2 if you want to take advantage of it, Step 3 if you want to abuse it.
And what statistical proof do you have? A couple of forum members with killed cpus reporting software monitored voltage readings doesn't prove jack shit.
Statistically you have a much better chance of having your cpu survive if you follow intel guidelines, rather than trusting on the fact that your cpu could be a "statistical outlier" that is either weaker or stronger than the average.
/thread
Tempered81
Diamond Member
I'm with Jag on this. 1.45v is about the max i would ever use on aftermarket air cooling.
1.488v in bios gives me 1.424 on the desktop.
1.488v in bios gives me 1.424 on the desktop.
I found these couple of pages written nearly a year ago by Kris to be a rather accurate description of the situation:
http://www.anandtech.com/showdoc.aspx?i=3251&p=5
The contents of the link would appear to favor Gillbot's interpretation of Intel's use of the designation of max processor voltage when taken in context to the backdrop if why retail chips have 3 yr warranties (versus say 5 yr, or 50 yr, etc).
Warranties exist because chips will die even when maintained within the operating specifications provided by Intel.
These so-called "early fails" are an unavoidable reality of any set of physical samples adhering to Weibull Distribution statistics.
It's not rocket science, but you can't build rockets (or CPU's) without knowing this stuff. So it should come as no surprise that Intel, and the folks who set warranty periods as well as maximum allowed voltage specifications, know exactly what they are doing when they make their specifications.
You can read whatever you like into those specifications, but if you don't understand the science and industry that exists behind the scenes that makes those specifications exist in the first place then you are naturally going to allow yourself to convince yourself that they mean whatever you feel they should mean...
http://www.anandtech.com/showdoc.aspx?i=3251&p=5
The contents of the link would appear to favor Gillbot's interpretation of Intel's use of the designation of max processor voltage when taken in context to the backdrop if why retail chips have 3 yr warranties (versus say 5 yr, or 50 yr, etc).
Warranties exist because chips will die even when maintained within the operating specifications provided by Intel.
These so-called "early fails" are an unavoidable reality of any set of physical samples adhering to Weibull Distribution statistics.
It's not rocket science, but you can't build rockets (or CPU's) without knowing this stuff. So it should come as no surprise that Intel, and the folks who set warranty periods as well as maximum allowed voltage specifications, know exactly what they are doing when they make their specifications.
You can read whatever you like into those specifications, but if you don't understand the science and industry that exists behind the scenes that makes those specifications exist in the first place then you are naturally going to allow yourself to convince yourself that they mean whatever you feel they should mean...
Markfw
Moderator Emeritus, Elite Member
I think this is related..... I have RMA'ed every Gskill PC-8000 set of memory I have had over the past 2 years, a total of 6 times, until the last 2 times, they ran out of stock (thank god). They all dies I THINK because the stock voltage was 2.3-2.4, and thats where I ran them. I think the spec was just over-volted, factory overclocked sticks, and they died because of it.
And I agree that 1.6 is insane for that chip. I feel a little worried that I have mine at 1.375 ! Its just a matter of time (unless your computer is off a lot)
And I agree that 1.6 is insane for that chip. I feel a little worried that I have mine at 1.375 ! Its just a matter of time (unless your computer is off a lot)
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 25K
-
-
-
