Is your 45nm degrading?

[Glow9]

So read a few things on the 45nm chips degradation going on so I'm curious over all if it's as bad as some people are making it out to be. Clearly on this forum people push their chips so I figure this would be a good place to find out.

[Brahmzy]

I've got 2 high-end boxes at home here. One's got an E8400 at 4.0GHz - it degraded quite a bit very quickly but then quickly stopped, I'm still able to maintain 4.0 with very acceptable volts.

My main box has had an E8600 at 4.4GHz the day they came out. Not one hint of degradation at all in months. Possibly because of the E0 stepping?

So, yes and no.

[Leeghoofd]

Like mentioned in the other thread I smoked my first Qx due to inadequate knowledge of the mobo and CPU at that time... pushing 65nm volts through it for a whole weekend during an SLI benching session made it degrade quite a bit... yet the CPU still runs strong at 4ghz with +/-1.35volts...Before it needed around 1.32 for prime stability at 4ghz... My error totally (one of the reasons I have a DMM now lol)

All the other CPU's stayed stable after the burn in period...some even got better ( better voltage regulation on the P5Q dlx )

One E8500@Ghz at 1.25volts, one E8400@4ghz 1.21volts, QX9650@4ghz at 1.26volts and the E8600@4.5Ghz at 1.36volts... Wolfies all at 500FSB, Qx at 450FSB. All are on air just the Qx is on water... E8600 is pretty new so maybe too early to judge... other CPU's are over 7 months , with the Qx being the oldest...

[cabbspapp]

I have a Xeon E3110 sitting at 4005 Mhz with 1.328 Vcore and it's been just fine, 10 hour prime stable after ~4-5 months. That's not a lot of voltage though.

[{GOD}Raptor22]

Wheres the option for 90nm (PD820)...

[CryptiK]

My e8400 degraded rapidly over about 4 weeks, at first needing only 1.304v for 4GHz stable, then it became unstable and needed 1.32v, was ok for a week or so then became unstable and needed 1.328v, was ok for another week then became unstable and now requires 1.336v. Seems to have not degraded any more but it might, no way to tell with these flakey 45nm cpu's.

[oli_ramsay]

My e8400 requires 1.465V for 4GHz and it's not even orthos stable at that. It's stable for everyting else I use it for though (gaming, music, internet etc). I'm hoping to get an e0 stepping of it when the prices fall after Nehalem launch.

But it hasn't degraded at all.

[carepolice]

Originally Posted by CryptiK

My e8400 degraded rapidly over about 4 weeks, at first needing only 1.304v for 4GHz stable, then it became unstable and needed 1.32v, was ok for a week or so then became unstable and needed 1.328v, was ok for another week then became unstable and now requires 1.336v. Seems to have not degraded any more but it might, no way to tell with these flakey 45nm cpu's.

Sounds pretty sad, I bet that you was even in safe limits of all voltages

[CryptiK]

Yep, vcore as stated above and I never went over 1.336v, vFSB 1.11v, vPLL 1.50v. I never abused it, stayed well within spec on all voltages, and degradation still occurred.

[rge]

E8400 was 15hrs prime stable 4.2ghz with 1.36v bios loadline enabled (1.36cpuz), ran it 4 months that way, and was still 15hrs prime stable after that and some benching at 1.5V.

Then ran it at 1.425 bios, loadline off, 1.4v cpuz 4.3ghz. Initially prime 15hrs stable, then after 4 months of normal usage, started getting random crashes, esp. in games, never crashed once before...rechecked and stopped prime in 2-3 seconds repeatedly.

Originally needed 1.3V bios for 4hz, now running at 1.36v bios (1.33cpuz) for 4hz. But no further degradation since.

[Monkey101]

LOL i like the ammount of people who have their 45nm CPU at stock speeds :P

[Ozzfest05]

Well have my E8500 at 3.8 for 24/7 nothing special, had it up to 5ghz on air at really high volts C0 stepping but havent noticed degradation its a decent chip

[BenchZowner]

Degradation ?
What's that ?

None of my CPUs is showing any signs of degradation yet.
But then again, I'm not pushing high voltages through them ( I always stop at 1.9V on air / water cooling )

[eaa315]

E3110 @ 4.0gigs w/ 1.32 volts for about 6 months now with no issues.

[rge]

Actually looking at my old orthos 10-64ffts stable pic from several months ago (left pic) then running all exact same settings (right pic), nearly the same stability...not counting the first two attempts at orthos as it rebooted.

[Glow9]

If people are telling the truth it's kinda scary 1/3 of chips are degrading. Honestly I thought it was gonna be like 4 in 30

[rge]

Originally Posted by Glow9

If people are telling the truth it's kinda scary 1/3 of chips are degrading. Honestly I thought it was gonna be like 4 in 30

Mine degraded only when abusing it. If I actually cared about my chip, I could have prevented it.

Benching 1.65 for an hour and running 1.36v cpuz for months with loadline enabled, did not degrade even one notch.

Running 1.425v for 4 months degraded first time, then benching few hours this am trying/failing to get 5ghz. 4.9 easy with 1.7V, but 5ghz is no go, ram/mem had much more room but cpu cant do it on water with 10C outside air. Interestingly after few hours it would not do 4.9 anymore even with 1.8V.....waiting for i7 results before deciding on e8600 vs q9650 vs i7, so this chip is going to be relegated for backup use soon anyways.

[truehighroller]

Originally Posted by CandymanCan

Yea mine has but im not sure if its degradation or a burn in period or because i used diff voltages at the time of testing. Because at first my cpu was able to run 4200mhz with 1.288v and now it requires 1.296v but it hasnt gone down since and its running fine. 4300mhz required only 1.338v and now it needs 1.36v.

It also might be due that i was running the VFSB Termination when i first tested the cpu at 1.20v and when i relised that was kinda high i lowerd it down to 1.10v so now the cpu needs more vcore. Im not sure if this is degradation or due to fact that i simply lowerd 1 of the voltages causing it to need more voltage on something else to continue running stable. I dont really care tho my 24/7 speeds are 4200mhz because the voltage is all default or extremely low and 4200mhz is perfectly fine for me i dont need that extra 300mhz for 24/7 usage and i dont benchmark anymore so i see no point.

I guess the only way to tell is to raise my VTT back to 1.20v and lower the vcore back to 1.288v and see if it will pass 4200mhz if not then i guess i do have degradation. If it does pass then it was just because the VTT voltage was to low for that vcore.

Benchzowner: I havent fully read through your thread where you took bets on if the cpu would die or not, but the parts i did read i dont think i saw a single screen shot or video or anything from you proving that your cpu or cpu's havent failed or degraded all i see is writing and frankly its rather hard for me to belive that your putting 1.9v through a 45nm cpu and its running fine 24/7 without any proof

You can safely push it to 1.25v. Some one put out an intel thing that showed it some where on this thread. I think the gtl reference thread actually..

[rge]

Originally Posted by BenchZowner

Degradation ?
What's that ?

None of my CPUs is showing any signs of degradation yet.
But then again, I'm not pushing high voltages through them ( I always stop at 1.9V on air / water cooling )

My E8400 degraded .03v in a few hours at 1.8V vcore on water in 10C outside air.

Can you post pics of cpuz/stress testing from different time points to show it has not degraded at that high vcore, etc

[AVWar]

just killed E8500 waiting for E8600

[nugzo]

Originally Posted by BenchZowner

Degradation ?
What's that ?

None of my CPUs is showing any signs of degradation yet.
But then again, I'm not pushing high voltages through them ( I always stop at 1.9V on air / water cooling )

Originally Posted by rge

My E8400 degraded .03v in a few hours at 1.8V vcore on water in 10C outside air.

Can you post pics of cpuz/stress testing from different time points to show it has not degraded at that high vcore, etc

I'll answer for him... If you aren't familiar with his thread, the sarcasm may not have been detected. http://www.xtremesystems.org/forums/...d.php?t=198614

[nugzo]

Originally Posted by CandymanCan

I havent checked through the entire thread but from the first 4 pages i see no proof at all no proof and there is no way im beliving it, i mean everyone seems to praise this guy just for his word. He hasnt posted 1 screen shot even showing the voltage from cpu-z let alone a single prime95 or burntest....

BZ has a good rep around here as far as i know. We've both been around for quite a while and i cant recall anything negative. Plenty of useful threads and posts by BZ. My VTT is at 1.4 b/c of his thread. All is good although its still early. Screenshots dont realy prove anything pertaining to that thread. If he wanted to be deceptive, then falsifying SS wouldn't be a problem. Believe Candyman Believe!!!!! i believe!

[Glow9]

Originally Posted by AVWar

just killed E8500 waiting for E8600

What you do?

[Fungus]

I run my E8500 at 3.8ghz 24/7 for a few months now and I have had no degradation.

[BenchZowner]

Relax... if you really want it that much, I'll give you some.
You just have to wait a bit... unless you're willing to write those 14 hardware reviews that I have to get done ASAP.