conroes degrading?

I can not think of another way to describe this...

setup 1:
p5b
2*512 5300 crucial
6600
big typhoon

At first I was running 3.7ghz 1.65v 24/7 for a week or two then it started BSODing

Set 3.6 and I was fine

Switch to water and I still can't run 3.7

setup 2:
xbx2
2*1024 6400 2gbhz and the crucials from setup 1
3060 xeon
big typhoon

When I got the chip I could:

orthos 3.75 1.5v
3946mhz 3d 1.55v
4ghz 8m 1.55v 1333 strap

I ran 3.75ghz for a while then one day it rebooted so I just stuck with 3.6

Without changing anything I vmch modded my board so I could run 4ghz on 1066 strap.

4ghz won't post.

Try 2507 bios, same thing.

Try 3946mhz again and it crashes 3dmark. 1.575v and 3946mhz works.

Try 4ghz and it black screens after post flash screen.

Previously, 4ghz would boot fine on 1333 strap and 1066 strap would hang at the windows loading screen, not it just goes to a black screen.

I remounted the typhoon twice with the same amount of pressure, same TIM (ceramique), same ambient, same CPU load temps, same BIOS settings, same EVERYTHING, and no go with my old settings. What is going on?

3.7ghz 1.65v 24/7 for a week or two could have been the issue with the 6600, as for the xeon, I am not sure unless there is an issue with your mch mod.

When you run an overclock that high any number of things can contribute to instability. It could be your RAM, the motherboards, the power from the wall, anything. I wouldn't instantly assume it was the CPU based on such minimal evidence.

Minimal evidence?

I've used

2 boards
2 cpus
3 sets of ram

Same thing happens

Well, any piece of silicon will degrade if its pushed beyond its comfortable limits for extended periods. Perhaps you've found that limit for your chip. Cant take that much voltage I guess, and the Xenon maybe can since its a heartier tested chip.

This is pretty disconcerting for sure.

Quote:

---

Originally Posted by Gautam

This is pretty disconcerting for sure.

---

Ya really! What the hell:( . I guess im staying @ 1.45Vcore for now on.

Quote:

---

Originally Posted by newls1

Ya really! What the hell:( . I guess im staying @ 1.45Vcore for now on.

---

Huh..first report after 6-7mo with conroes???

who is to say the 2x 1gb ram and 2x 512 ram is not playing as happy with time....

also how long were they orthos stable? small fft, large fft, blend?

I would think that if these chips were degrading we would be hearing alot more about this here at XS...many are running 1.55v+....

maybe air cooling is part of the issues as ambients maybe are changing too...

also vmodded mobo..any chance you could a mobo problem from this..not really getting stable and consistent vcore?

1.325 vcore---1.55v =17%
1.525=15%
1.5v=13%
those values are not considered extreme....

It was an mch mod, stable vcore isn't a concern with that.

What do you mean 2gb vs 1gb, I used both.

2:30 hour orthos small ftt, then 24/7 WCG stable for weeks until it just crapped out.

Ambients were the same.

Now that I think of it, maybe I was running 1.565v and not 1.55, 1.565 almost completed 3dmark and the fan I'm using now is slightly weaker than the m1a @ 12v.

I really can't even tell at this point, this board is totally fkd. xbx2 is the worst board I've ever touched, hands down.

Electromigration and Sudden Northwood Death Syndrome returns

Don't the voltage mods put extra strain on the motherboard power components? Maybe they are failing? NFI really, just came to mind.

Quote:

---

Originally Posted by afireinside

It was an mch mod, stable vcore isn't a concern with that.

What do you mean 2gb vs 1gb, I used both.

2:30 hour orthos small ftt, then 24/7 WCG stable for weeks until it just crapped out.

Ambients were the same.

Now that I think of it, maybe I was running 1.565v and not 1.55, 1.565 almost completed 3dmark and the fan I'm using now is slightly weaker than the m1a @ 12v.

I really can't even tell at this point, this board is totally fkd. xbx2 is the worst board I've ever touched, hands down.

---

I thought you were running...all 4 sticks together??

I was questioning whether the mods affect the mobo long term....

I really doubt 1.55 vs 1.565 is that big an issue..esp. on water

I just question whether this is a real issue or isolated(Electromigration and Sudden Northwood Death Syndrome returns..I sure hope not!!!)..I have seen no other reports, am I wrong

Quote:

---

Originally Posted by afireinside

Minimal evidence?

I've used

2 boards
2 cpus
3 sets of ram

Same thing happens

---

Yeah, and "a week or two". Not enough time to blame the chip unless those high voltages fried them.

You guys gotta remember that as CMOS processes that are used to produce these chips shrink they become more and more sensitive to voltage, perhaps by the last process' standards 17-20% more voltage wasn't a major issue but its not unreasonable to think that with their current 65nm that it might be. Remember how it was with the P4 2.4C when everyone started OC'ing them and pushing the volts up and several months later we started getting reports of them spontaneously dying and/or getting degraded overclocks? Personally I won't go over 1.5v with my E6300 no matter what the cooling as its really the volts that will end up severely degrading the life of your chip (heat does have an impact to mind you, but you have to run the chip at or over the recommended limit for extended periods of time before it really becomes an issue).

Quote:

---

Originally Posted by mesyn191

You guys gotta remember that as CMOS processes that are used to produce these chips shrink they become more and more sensitive to voltage, perhaps by the last process' standards 17-20% more voltage wasn't a major issue but its not unreasonable to think that with their current 65nm that it might be. Remember how it was with the P4 2.4C when everyone started OC'ing them and pushing the volts up and several months later we started getting reports of them spontaneously dying and/or getting degraded overclocks? Personally I won't go over 1.5v with my E6300 no matter what the cooling as its really the volts that will end up severely degrading the life of your chip (heat does have an impact to mind you, but you have to run the chip at or over the recommended limit for extended periods of time before it really becomes an issue).

---

Exactly. 1.3525v is the stated limit of the chip. Running 1.65v through it and expecting no repercussions is rather short-sighted.

Personally, I need my system to last so I shy away from extreme voltages.

Hmm, I'm running 3.73GHz, 1.5v now. Maybe I should notch it back down to 3.6GHz, 1.45v.

Quote:

---

Originally Posted by afireinside

What is going on?

---

See below.

Quote:

---

Originally Posted by afireinside

I was running 3.7ghz 1.65v 24/7 for a week

---

Mystery Solved!

Quote:

---

Originally Posted by Anonymous

See below.


Mystery Solved!

---

Thanks for the brilliant observation Mr. Anonymous. It was more like 1.55v because my p5b wasn't vdroop modded.

Quote:

---

I thought you were running...all 4 sticks together??

I was questioning whether the mods affect the mobo long term....

I really doubt 1.55 vs 1.565 is that big an issue..esp. on water

---

no, not running 4

The mod was on for all of 2 minutes and voltages were not changed with it

1.55 vs 1.565 wasn't on water

can we get a mod to move this into main Intel section

I am finding it hard to believe that 1.55v is causing an issue...way too many around here running that vcore with no reported issues

My 4 months old E6700 is running 24/7 at 1.75V in BIOS with watercooling, still at 3.9GHz, no degrading, no unstability ... i changed only many mobos, cpu is the same ...

Quote:

---

Originally Posted by Cracker

When you run an overclock that high any number of things can contribute to instability. It could be your RAM, the motherboards, the power from the wall, anything. I wouldn't instantly assume it was the CPU based on such minimal evidence.

---

100% agreed, Conroe does not degrade Sean.

Quote:

---

Originally Posted by Anonymous

See below.


Mystery Solved!

---

Ive run 1.67v measured for a week(only owned X6800 for a week :D), 24-7 and all was great :)
I should have about a year of 24-7 at over 1.63v between my last three cpus, all are 100% functional.

Quote:

---

Personally I won't go over 1.5v with my E6300 no matter what the cooling as its really the volts that will end up severely degrading the life of your chip (heat does have an impact to mind you, but you have to run the chip at or over the recommended limit for extended periods of time before it really becomes an issue).

---

Heat degrades electronics, voltage also can, but a simple 25% increase or so in voltage wont hurt anything long-term. (unless you expect to be 80 years old and still using it, or your grand children to be using it!)

I have been running 1.61v through my cpu for 6 months on phase. No problems whatsoever. Still has stable as the day I got it.

Quote:

---

Originally Posted by nealh

can we get a mod to move this into main Intel section

I am finding it hard to believe that 1.55v is causing an issue...way too many around here running that vcore with no reported issues

---

It WAS in Intel section until it got moved...

Quote:

---

Originally Posted by fhpchris

Ive run 1.67v measured for a week(only owned X6800 for a week :D), 24-7 and all was great :)
I should have about a year of 24-7 at over 1.63v between my last three cpus, all are 100% functional.

---

Quote:

---

Originally Posted by Yoxxy

I have been running 1.61v through my cpu for 6 months on phase. No problems whatsoever. Still has stable as the day I got it.

---

Quote:

---

Originally Posted by OBR

My 4 months old E6700 is running 24/7 at 1.75V in BIOS with watercooling, still at 3.9GHz, no degrading, no unstability ... i changed only many mobos, cpu is the same ...

---

Quote:

---

Originally Posted by nealh

I am finding it hard to believe that 1.55v is causing an issue...way too many around here running that vcore with no reported issues

---

There is no consistent and reliable way to measure how these chips will degrade past the manufacturer's specs nor is there anything that suggests that they will all degrade in the same fashion/time period, only Intel would know and they don't exactly give that info. away. Not all of those P4C's started having issues at the same time you know, and there those who never had issues at all seemingly no matter what voltage they put their chips at.

Quote:

---

Originally Posted by fhpchris

Heat degrades electronics, voltage also can, but a simple 25% increase or so in voltage wont hurt anything long-term. (unless you expect to be 80 years old and still using it, or your grand children to be using it!)

---

There is no way for you to know this without NDA'd process info. from Intel, you wouldn't mind posting though if you do have it right?

the problem I see is this..it maybe that xs voltage does cause an issue
but heat is a killer as well...and we all have seen stock cpu with stock HSF..have crazy temps..which even at normal vcore would and could be as bad....those concave IHS..are bad

I am guessing the safety margin is pretty high or we would be seeing much higher failures rates here and around the net

BTW phase(sub-zero temps) does not prevent xs voltage damage...but lower temps make it more tolerable and slow the effects

concave IHS can lead to extreme temps...and the average user does not even check..if the system does not crash..they pay no attention

Heat and increased voltage are bad..together this is a horrible combination

I am not concerned at shortening like of my cpu from 10 yrs to 7 yrs or even 5yrs...C2D will be not even exist...my 2-3 yrs is a fair life

2-3yr. of use overclocked would be about the best I would expect, but what if its more like 6 mo. to a year if overvolted + a temp. of xC, there would be no way for us to know for sure what those values would be either which way which is kind've my point. We don't know so its best to be conservative unless you plan on doing suicide overclocks and don't care either which way if the chip lasts 3 years or 3 hours.

Quote:

---

Originally Posted by afireinside

can we get a mod to move this into main Intel section

I am finding it hard to believe that 1.55v is causing an issue...way too many around here running that vcore with no reported issues

---

I understand what you are saying but that doesnt mean somone out of a handful wont run into a problem at that voltage. In all truth these CPUs havent been out that long for anyone to come to a conclusion on how far you can push a random core2 before it breaks. The only way you could know for sure is if you ask an Intel eng for that data because they are the only ones who know the tolerances based on internal stress testing.

Quote:

---

Originally Posted by afireinside

I really can't even tell at this point, this board is totally fkd. xbx2 is the worst board I've ever touched, hands down.

---

My xbx2 is boxed up and ready to go back to the store. My board went from stable to flaky in less than a week. One minute it boots fine and the next it would not at the same settings. :fact:

Edit: Before I get flamed, I am not saying all of these boards are bad as a lot of members love them and have great results, but not me.

just to add my 2c, when i got my conroe i was running 100% stable at 3.5999999 Ghz, this was at about 1.55v on a triple 120 rad. I did notice over time the cpu drop, but blamed it in part on the weather and dust clogging up whatever - when i got to 3.5 at the same volts and it was still a bit unstable, i dropped the volts down to about 1.4 and the overclock to 3.3 hoping that if i did that it would not drop further. This was around xmas.

Nox

Quote:

---

Originally Posted by nox_uk

just to add my 2c, when i got my conroe i was running 100% stable at 3.5999999 Ghz, this was at about 1.55v on a triple 120 rad. I did notice over time the cpu drop, but blamed it in part on the weather and dust clogging up whatever - when i got to 3.5 at the same volts and it was still a bit unstable, i dropped the volts down to about 1.4 and the overclock to 3.3 hoping that if i did that it would not drop further. This was around xmas.

Nox

---

now this is very interesting and seems to support a degrading of the cpu

was your overclock run 24/7? was it prime95 stable?

Do you suppose, as would make sense, that the problem, if there is one, would be more prevalent in the E6600 rather than the X 3060? I am inferring that it would be safer to get a Xeon 3060 (not to mention better OC's from some sources ... worse OC's from others... oh well).

I dont want to highjack the thread and start talking about badaxe2 but i was wondering how such a large range of feelings about the badaxe2 is possible.

There are people like me who think its the most stable/reliable board for overclocking ever (took me less than one day to get 3.8ghz air orthos stable) passed 8hr orthos at 3.8 and testing 3.86 for the moment...

It may depend on the revision or the bios maybe? Im using bios 2333.


Anyway, as for the life of your cpu, i think its technically possible that the 1.55vcore "killed" it but i think its hard to believe. There is so much people out there pushing over 1.6v with phase cooling 24/7 since conroe released... If your temp are not going over 70c i would believe the voltage killed it..

Quote:

---

Originally Posted by nealh

now this is very interesting and seems to support a degrading of the cpu

was your overclock run 24/7? was it prime95 stable?

---

That was everquest 1 stable lol, let alone anything else. I could prime etc a bit higher. My overclock was run a lot, sometimes 24/7, sometimes *only* 12 hours a day, depended what I was using it for. It could also be something else degrading though, or it could just be bad luck, some electronic stuff does degrade when run at stock anyway. My CPU is also not the flatest, should of lapped it but was more interested in keeping the warranty intact-ish, but certainly no where near as bad as some pics i'd seen - could tell this from the pattern in the thermal paste when connecting the block. I've also switched back on that thing that drops the multiplier down when the cpu doesn't need it now, that was 2 weeks ago when I remembered it existed!

i'm not too worried, if it dies i'll get a new one, if it doesn't i'm still happy with 3.3, though if it gets to 3 then i'll start thinking about replacing it!

Nox

i think this may be an issue in general with using a smaller process coupled with an etching medium like silicone.

i remember reading on something about the gates wearing out due to electrons missing the junction and creating damage.

a smaller process would account for increased wear with highoverclocks compared to older cpus using larger process.

might have be part of the reason why the tech is moving onto new materials?


who knows. only large scale long term testing would be conclusive and i'm sure only intel has attempted anyting close.

if the chip needed 1.65v just to do 3.7ghz then it was a weak chip to begin with. very weak.

i dont care who you are or where your from you go shoving 1.65v into a chip that uses 1.3v or less for standard operation all your doing is ask for the chip to die.

its just the simple facts and anyone who thinks they can just shove those kinds of volts into any cpu and expect it to survive needs to get real and wake up.

if all it takes is a few 1/10th's of a volt to kill ram dont you think the same thing is true for a cpu ? huh ? yes it is.

go buy a new cpu and get rid of that one or keep it around for emergency back up.

Thought this may have some bearing on this topic....I found this in a PDF from Intel titled :
"Conroe and allendale Electrical, mechanical, and thermal specifications"

http://img.photobucket.com/albums/v1...maxvolts-1.jpg

Quote:

---

Originally Posted by exhausted mule

i remember reading on something about the gates wearing out due to electrons missing the junction and creating damage.

---

Yea its called electromigration, the effect gets worse and worse the more they shrink the chips which is why they get more and more sensitive to voltage with smaller processes.

Quote:

---

Originally Posted by exhausted mule

a smaller process would account for increased wear with highoverclocks compared to older cpus using larger process.

---

Overclocking alone shouldn't cause any extra wear and tear on the chip so long as you keep the voltage at default and temps within spec, so if you're paranoid about your chip dying on you or just want it to last a long time just see how high a clockspeed you can get at default volts with good cooling and you'll be fine.

Quote:

---

Originally Posted by exhausted mule

might have be part of the reason why the tech is moving onto new materials?

---

No, they're switching materials and process tech to reduce leakage from the transistors which is becoming an overwhelming problem now.

Quote:

---

Originally Posted by L33T

Do you suppose, as would make sense, that the problem, if there is one, would be more prevalent in the E6600 rather than the X 3060?

---

No, they'd be equally susceptible to a degradation when over volting/over heated as they're both made on the same process, Xeons just go through a more rigorous validation process.

Well of course they're going to make it sound like 1.55V is the ultimate voltage before yielding or failure.

Consider a mechanical engineering analogy. When we design something, we have a safety factor "X", which for typical non-aerospace apps is usually 2.0. That means we build something so that it is X times the yielding stress, or in this case, twice the yielding stress. That way we have plenty of leeway.

It is the same way here. 1.55V is not the electrical equivalent of the "yielding point", nor the ultimate point. However, there is no way to tell what the X value is. It could be something insignificant, such as 1.001, which would in turn mean 1.552V is the true yielding point.

Well think of it this way, Intel has to guarantee that their chips will work for a certain min.of time and you don't see Intel selling any C2D's with a default voltage over that do you?

Quote:

---

Originally Posted by new-b

Thought this may have some bearing on this topic....I found this in a PDF from Intel titled :
"Conroe and allendale Electrical, mechanical, and thermal specifications"

http://img.photobucket.com/albums/v1...maxvolts-1.jpg

---

thanks..1.55v max...

I guess my 1.525v(bios setting) @3500 is ok....hmm...may just drop back to 3400..I prefer stability,afraid 1.5+ could lead to issues, not the life of the chip

I also saw my e6600 degrade a little.. it used to need 1.4500 in bios for 3.2ghz now it needs 1.475v bios (real volts are -0.05v)

Oh and by the way absolute max is the maximum voltage it is suppsoed to be able to take for for a short period time only. My old 7800gt has ddr3 with absolute max 2.5v.. did that mean 2.5v was fine 24/7, lol no.. however i did often bench with 2.45v..

I still find this an interesting topic....but I am not sure why we are not hearing more reports of issues given the vcore most are using

Quote:

---

Originally Posted by Lestat

if the chip needed 1.65v just to do 3.7ghz then it was a weak chip to begin with. very weak.

i dont care who you are or where your from you go shoving 1.65v into a chip that uses 1.3v or less for standard operation all your doing is ask for the chip to die.

its just the simple facts and anyone who thinks they can just shove those kinds of volts into any cpu and expect it to survive needs to get real and wake up.

if all it takes is a few 1/10th's of a volt to kill ram dont you think the same thing is true for a cpu ? huh ? yes it is.

go buy a new cpu and get rid of that one or keep it around for emergency back up.

---

Well put.

Reminds me of some DSM owners (Eagle talons, Eclipse) just upping the turbo boost without a care in the world and then wondering why they cracked their block.

Am I safe?
E6600 running 3.6 @ 1.475v :slobber: :slobber: :slobber:

Quote:

---

Originally Posted by lawrywild

I also saw my e6600 degrade a little.. it used to need 1.4500 in bios for 3.2ghz now it needs 1.475v bios (real volts are -0.05v)

---

very true.

I used to run 3.5 with 1.5V(in bios) now it needs 1.525(bios) to remain stable

My E6600 runs @ 3204 MHz, 1.33V speedfan / 1.36v ASUS Probe / 1.35v BIOS for about a month, 24/7, no degradation. In fact it looks like its upgrading itself, recently i've been able to push 3.6GHz at that voltage, orthos stable, but not for 24/7 use due to the temps.

I think it's more likely that the motherboards are wearing down (capacitors, mosfets).

Quote:

---

Originally Posted by rmd2001

Am I safe?
E6600 running 3.6 @ 1.475v :slobber: :slobber: :slobber:

---

I would think this well within reason..nice overclock..just a tad over 10%..remember amny mobos can overvolt a fair amount so I would expect Intel to take this into account


Ok..I just finished overclocking my e6600@3500(9x390) I need 1.525v stopped after 12hrs blend p95v251
after this topic..I dropped 3420(9x380) with 1.4625v....24hr P95v251 blend stable:woot:

I really like the way 3500 sound but not sure I want to push that vcore..1.525v...if 1.55v is at best a ?? long term max, at worst only ok for short times

There is no doubt if you are really pushing the mobo..high fsb and NB voltage, etc...this may very well be the problem

I for one will contine to try and lower my mobo voltages..NB@1.42v, SB 1.34v bios settings

If it is a voltage issue, maybe it has something to do with a specific set.

I ran my e6600 es at 3.6ghz, 1.55v for ~6months 24/7 under water and NEVER had a problem with it.

Quote:

---

Originally Posted by Lestat

if the chip needed 1.65v just to do 3.7ghz then it was a weak chip to begin with. very weak.

i dont care who you are or where your from you go shoving 1.65v into a chip that uses 1.3v or less for standard operation all your doing is ask for the chip to die.

its just the simple facts and anyone who thinks they can just shove those kinds of volts into any cpu and expect it to survive needs to get real and wake up.

if all it takes is a few 1/10th's of a volt to kill ram dont you think the same thing is true for a cpu ? huh ? yes it is.

go buy a new cpu and get rid of that one or keep it around for emergency back up.

---

damn lestat, cool down. we're just talkin about this, not calling each other morons.

Anyway, voltage overclocking is always a risk. You're pusing past the default. You can't guarantee any overclocking number can you? Not all conroes can hit 3.5 Ghz at certain voltages, we've all seen that. Quality won't be assured at higher voltages and higher clocks since Intel never pushed it that far. BUt that's not to say that all chips can't run at the higher clocks, as we've also seen.

And also, there is another reason for Intel to set the voltage lower: to lower temperatures and to lower power consumption. Intel's taken a lot of :banana::banana::banana::banana: for the Prescotts as being power whores.

hmm, exactly what i am experiencing...

was running 3.6 ghz orthos stable at 1.53V. then, after pumping 1.63 in for 3.7 ghz, i decided to drop back to the 1033 strap (3.6 ghz). now i cant run f@h 24/7 at 1.53v, 3.6 ghz as mentioned. i must set 1.625 to get 1.55 vcore always.

therefore, in the first week, it could run 3.6 ghz orthos stable. 1.53 v
ran that for 2 weeks, then moved up to 3.7, then moved back.
just a week ago, it kept on restarting (i have f@h as service, both cores). i thought it was vcore, so i bumped it up. now it works fine, just last night. im wondering if its the board.... i have decent air cooling: TT BT with 110 cfm and some exhaust fans. i highly DOUBT that its the cpu degrading.

Quote:

---

Originally Posted by Cracker

Exactly. 1.3525v is the stated limit of the chip. Running 1.65v through it and expecting no repercussions is rather short-sighted.

Personally, I need my system to last so I shy away from extreme voltages.

---

1.45v is Intel's specified maximum voltage in their white papers I believe. Someone correct me if I'm wrong. The 1.3525 is the maximum that they will program a chip to run at.

I've had the EXACT same issues described to a T in this thread.

Thought it was my memory, Bought another set same issues.. Now running 4GB's no problems. Wasnt memory.

Thought it was CPU, rigged up MachI for some temporary testing and brought CPU down to -23C.. Same instability at same OC.

Added better MCH cooling, problem "reduced"

Added Vmch mod, Problem "reduced"

I'll save ya the sob story about how I could run 3.850Ghz all day long running F@H 24/7 etc. I'm now down to 3.7 Ghz and it hasnt gone worse after I added the MCH mods after about 2 weeks. I had "reduced" because I still cant run 3.85 stable any longer.

In my case, the degradation happened in the time span of 4 weeks.

Oh ya, my Vcore never was above 1.43Volts actual measured with calibrated fluke DMM I use for work.

In my case, I think my problem is a weak northbridge on my XBX2. Dont listen to the people who say you dont need better cooling on the chipset... YOU NEED IT!!

Quote:

---

Originally Posted by SPL15

1.45v is Intel's specified maximum voltage in their white papers I believe. Someone correct me if I'm wrong. The 1.3525 is the maximum that they will program a chip to run at.

---

intel released an updated datasheet for C2D's in january. current specs are as follows:

Intel Core 2 Duo Desktop Processor E6000 Sequence – Datasheet
http://developer.intel.com/design/pr...hts/313278.htm

2.6.1 Absolute Maximum and Minimum Ratings

Table 4 specifies absolute maximum and minimum ratings only and lie outside the functional limits of the processor. Within functional operation limits, functionality and long-term reliability can be expected.
At conditions outside functional operation condition limits, but within absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. If a device is returned to conditions within functional operation limits after having been subjected to conditions outside these limits, but within the absolute maximum and minimum ratings, the device may be functional, but with its lifetime degraded depending on exposure to conditions exceeding the functional operation condition limits.
At conditions exceeding absolute maximum and minimum ratings, neither functionality nor long-term reliability can be expected. Moreover, if a device is subjected to these conditions for any length of time then, when returned to conditions within the functional operating condition limits, it will either not function, or its reliability will be severely degraded.
Although the processor contains protective circuitry to resist damage from static electric discharge, precautions should always be taken to avoid high static voltages or electric fields.

Table 4. Absolute Maximum and Minimum Ratings
VCC (Core voltage with respect to VSS) –0.3V / 1.55V (Min/Max)
VTT (FSB termination voltage with respect to VSS) –0.3V / 1.55V (Min/Max)

Notes:
For functional operation, all processor electrical, signal quality, mechanical and thermal specifications must be satisfied.
Excessive overshoot or undershoot on any signal will likely result in permanent damage to the processor.

2.6.2 DC Voltage and Current Specification

Table 5. Voltage and Current Specifications
VID (Range of VID) 0.8500V / - / 1.5V (Min/Typ/Max)
VCC (Supply Voltage) - / - / 1.5V (Min/Typ/Max)
VTT (FSB termination voltage) 1.14V / 1.20V / 1.26V (Min/Typ/Max)

Notes:
Adherence to the voltage specifications for the processor are required to ensure reliable processor operation.
Each processor is programmed with a maximum valid voltage identification value (VID), which is set at manufacturing and can not be altered. Individual maximum VID values are calibrated during manufacturing such that two processors at the same frequency may have different settings within the VID range.



Intel 965 Express Chipset Family - Datasheet
http://www.intel.com/design/chipsets...hts/313053.htm


11.1 Absolute Minimum and Maximum Ratings

Table 10-1. Absolute Minimum and Maximum Ratings
VCC (1.5 V Core voltage with respect to VSS) –0.3V / 1.375V (Min/Max)
VTT (System Bus Input Voltage with respect to VSS) –0.3V / 1.32V (Min/Max)

11.3.1 I/O Buffer Supply Voltage

Table 11.4 I/O Buffer Supply Voltage
VCC (MCH Core Supply Voltage) 1.188V / 1.25V / 1.313V (Min/Typ/Max)
VTT (System Bus Input Supply Voltage) 1.14V / 1.20V / 1.26V (Min/Typ/Max)



Intel 975X Express Chipset MCH - Datasheet
http://developer.intel.com/design/ch...hts/310158.htm

10.1 Absolute Minimum and Maximum Ratings

Table 10-1. Absolute Minimum and Maximum Ratings
VCC (1.5 V Core voltage with respect to VSS) –0.3V / 1.65V (Min/Max)
VTT (System Bus Input Voltage with respect to VSS) –0.3V / 1.65V (Min/Max)

10.4 DC Characteristics

Table 10-5. DC Characteristics
VCC (MCH Core Supply Voltage) 1.425V / 1.50V / 1.575V (Min/Typ/Max)
VTT (System Bus Input Supply Voltage) 1.14V / 1.20V / 1.26V (Min/Typ/Max)


the notes for absolute minimum and maximum ratings on top of this post, holds true for chipset voltages also!

umm those are the limits if you want it to outlive the warranty