E8400/8500 degradation myth possibly busted?

[CrazyNutz]

I am having a similar issue where when I run Prime which was unstable for several runs, will one day be stable for 13+ hours. All settings are exactly the same and so is the temperature. I have absolutely no explanation for this.

Yea I know what ya mean. This is what I'm trying to figure out, and why I think a lot of people are thinking this is degradation. I don't know what to
call it yet, I initially thought it was degradation, but I'm not convinced of it.
Since it appears to be degradation one day, and the next It's all good again.
What mobo do you have, I think I have stumbled on to something that has
made an improvement, I need to test it some more before I can be certain.

[Cronos]

Why don't you guys try something else besides Prime?

[CrazyNutz]

Why don't you guys try something else besides Prime?

I have, OCCT, same story though.

[mrcape]

Isn't the symptom that you are experiencing a bit different than the standard cpu degradation symptoms? It seems your and tranceaddict's symptoms are ramdom where mine and others are not at all. If my chip fails at an old clock it fails in at the exact same time, to the minute, every time. It even scales up consistently at each clock - add .01v and it passes 5 minutes, add.025v and it passes 17 min., add .035 and it passes 30min. etc etc... I've never had it fail then pass then fail at the same clock/voltage.

I see degradation like this -

1. get a fresh chip
2. run stability tests and record settings from stock up to max stable OC (testing every notch of vcore on the way up to find min. vcore for each clock)
3. Go ahead and have fun benchmarking (in my case vcore up to 1.42v and fsb 545)
4. Test chip a week or more later with same tests at same clocks and settings
5. If chip cannot pass identical tests at identical settings, but can pass with a voltage bump of x notches, chip seems to have degraded
6. If chip needs the same amount of voltage increase to pass each test, then the degradation seems consistent. Ex. - needs .05v at every clock to pass the same tests as a week ago

If the chip already has symptoms, it's too late to apply the above method.

Of course there are other variables in the system, but in a nutshell I think the above steps could tell you something. I tried that method out of the box and that is what I found. After the fact I also tried fresh ram kit and two other mobos - P5E and P5K-Premium, to see if I could get the same low v clocks, but nope.

[gooface]

I just declocked my chip to 4.05ghz and i still get crashing in orthos.... wtf? do i need to RMA my chip? was working great at 4.25ghz till like sunday, then i got a BSOD.... so i declocked it and still errors... i also volted it down.

[tranceaddict]

I have the Asus P5K-E WiFi with 1006 BIOS and 2GB of Patriot Extreme Performance PC2-8500. What I find strange is that every other chip I had was like "mrcape" described where: the "chip fails at an old clock it fails in at the exact same time, to the minute, every time. It even scales up consistently at each clock". That is why i find this one really strange, there is no patter to it, and yes I am even taking ambient temps into account. I just find it hard to believe that I could have degraded it with runing it at 1.43V for less than 45min and the temps where in the mid 50s. The only other thing that changed since that session was me installing Vista Inspirat 2.1 with Windows Blinds 5.5. But I do not see how those OS changes can effect my OC stability.

[situman]

Perhaps start a poll to see how many people experienced this degradation? I mean if it is widespread, it could be a core problem or Intel built in anti-overclocking.

[Spyrus]

Maybe it is a bug of the core like the errata things...

[CrazyNutz]

Isn't the symptom that you are experiencing a bit different than the standard cpu degradation symptoms? It seems your and tranceaddict's symptoms are ramdom where mine and others are not at all. If my chip fails at an old clock it fails in at the exact same time, to the minute, every time. It even scales up consistently at each clock - add .01v and it passes 5 minutes, add.025v and it passes 17 min., add .035 and it passes 30min. etc etc... I've never had it fail then pass then fail at the same clock/voltage.

I see degradation like this -

1. get a fresh chip
2. run stability tests and record settings from stock up to max stable OC (testing every notch of vcore on the way up to find min. vcore for each clock)
3. Go ahead and have fun benchmarking (in my case vcore up to 1.42v and fsb 545)
4. Test chip a week or more later with same tests at same clocks and settings
5. If chip cannot pass identical tests at identical settings, but can pass with a voltage bump of x notches, chip seems to have degraded
6. If chip needs the same amount of voltage increase to pass each test, then the degradation seems consistent. Ex. - needs .05v at every clock to pass the same tests as a week ago

If the chip already has symptoms, it's too late to apply the above method.

Of course there are other variables in the system, but in a nutshell I think the above steps could tell you something. I tried that method out of the box and that is what I found. After the fact I also tried fresh ram kit and two other mobos - P5E and P5K-Premium, to see if I could get the same low v clocks, but nope.

I agree with what you are saying, but a few times I've seen you post "It's stable again" (Maybe not at your original clocks), so there may be a mixed
issue with your CPU, a little degradation, and a little of what I'm experiencing.

What I, and Tranceaddict are seeing are very similar, he also has the exact
same mobo/bios version as me. But I do agree if you are "never" able to make it stable at the Initial settings then It would likely be degraded.

Like situman suggests a poll would be good at this point, however we need to clearly define what criteria must be met to clam degradation.

[CrazyNutz]

I have the Asus P5K-E WiFi with 1006 BIOS and 2GB of Patriot Extreme Performance PC2-8500. What I find strange is that every other chip I had was like "mrcape" described where: the "chip fails at an old clock it fails in at the exact same time, to the minute, every time. It even scales up consistently at each clock". That is why i find this one really strange, there is no patter to it, and yes I am even taking ambient temps into account. I just find it hard to believe that I could have degraded it with runing it at 1.43V for less than 45min and the temps where in the mid 50s. The only other thing that changed since that session was me installing Vista Inspirat 2.1 with Windows Blinds 5.5. But I do not see how those OS changes can effect my OC stability.

What is the batch# of this CPU?

[mrcape]

I agree with what you are saying, but a few times I've seen you post "It's stable again" (Maybe not at your original clocks), so there may be a mixed
issue with your CPU, a little degradation, and a little of what I'm experiencing.

What I, and Tranceaddict are seeing are very similar, he also has the exact
same mobo/bios version as me. But I do agree if you are "never" able to make it stable at the Initial settings then It would likely be degraded.

Like situman suggests a poll would be good at this point, however we need to clearly define what criteria must be met to clam degradation.

No, you still didn't read my posts. I never made a flat statement like "it's stable again". I never had a problem getting stable.

It's stable if I add voltage, and that's exactly what a chip needs more of if it degrades.

I never had anything similar to what you're experiencing. Just had to reiterate that.

I think you're right it's poll time. Can you add it to this thread or does it need a new one? It's going to be hard to get comprehensive data from a poll, but I guess if you frame it correctly...

[CrazyNutz]

No, you still didn't read my posts. I never made a flat statement like "it's stable again". I never had a problem getting stable.

It's stable if I add voltage, and that's exactly what a chip needs more of if it degrades.

I never had anything similar to what you're experiencing.

Ok my bad. I think the poll can be added, I'll check

EDIT: yes the poll can be added, we need to come up with the poll questions.
Like:
Do you think/Is your 45nm CPU degraded?
At what voltage
At what Freq
At what FSB
etc, etc..

[mrcape]

Yeah I think those are the crucial questions and the answers could be ranges.

[tranceaddict]

What is the batch# of this CPU?

Batch #: Q740A553T
Ship/Packaging Date: 12/20/07

[CrazyNutz]

Yeah I think those are the crucial questions and the answers could be ranges.

Ok, I'll have it up monday.

[ninjagordy]

Well lads,

After Reading all the degredation threads etc all i can say is im glad i never sold my Q6600 to get one of these.

it runs sweet and im always pushing for more so im sure i would have dusted a couple of these....

sorry to throw a dampner into the wolfdale party but im sure theres others out there like me reading all these and saying "phew! close one!"

i will keep watching with interest to see how the new line of quads fair before i take the plunge.....

good thread though, well worht the read!

[kathman]

Hi everyone =D. Would a small overclock to 3.6 and a small increase in voltage cause much/any degradation?

[SoulsCollective]

No. Degradation appears to only kick in in a serious way at above 1.4V. Higher clocks won't cause degradation as long as you're adequately cooling your CPU.

[kathman]

Oh phewf. Sweet, my cooling is good, and my voltage is way below that =D. Thanks for the reply

[kiwi]

Interesting I have noticed something similar

Here is what happened to one of my cpus (8400) and commando:

1. new cpu, stable 1.33V 4.1Ghz
2. 6-7hours bench DI (1.85V, pll default)
3. cpu not stable anymore as it was in #1, now needs 1.35V for 4ghz, 4.1 fails with 1.35
4. 6-7hours bench ln2 (1.85V,pll default)
5. Now 4.1 ok with 1.35V but 1.33V fails
6. After 2-3 weeks without any benching, system on 24/7 now 1.34V (will try 1.33V now since testing is still in progress) and 4.1Ghz OK


Go figure

[TheThreeDegrees]

I'm glad I sold my Q6600 for one of these.
Tell you why:
Temps dont bother me anymore.
At stock speed it knocks the socks off any 65NM proc and most current 45NM.
It overclocks like a mo fo.plenty of verified clocks on this forum.
The so called "degradation myth" has not been proven by anyone.

I think I know why they are failing certain Prime / Orthos checks.
They are using 30% less power than your previous generation procs and the Motherboards out there now are not fully compatible with them and thus are allocating the incorrect power.

[aoch88]

I had the same problem today. My E8200 @ 4Ghz turned unstable with my normal settings. These 45nm sucks!

[Leeghoofd]

I had the same problem today. My E8200 @ 4Ghz turned unstable with my normal settings. These 45nm sucks!

Loooooooool 4ghz with 8 x 500FSB things can give in.
Are you sure it's the CPU ?

[mrcape]

As soon as I have the scratch I'm getting a third wolffdale or xeon, or maybe e7300. I think they rock! Although mine have shown signs of damage, I have found safe zone of 4.25ghz with 1.32vcore, and at that clock no further damage for two weeks now. It's pretty damn good for $200.

I still want to see what the poll says, what the damage points are etc.. I plan on getting more of these and can't wait til I do so I can go back and bench the crap out of the 745.

[jonny_ftm]

As soon as I have the scratch I'm getting a third wolffdale or xeon, or maybe e7300. I think they rock! Although mine have shown signs of damage, I have found safe zone of 4.25ghz with 1.32vcore, and at that clock no further damage for two weeks now. It's pretty damn good for $200.

I still want to see what the poll says, what the damage points are etc.. I plan on getting more of these and can't wait til I do so I can go back and bench the crap out of the 745.

If we believe what was said on this forum in many topics and on other numerous forums if you read around, +1.40v with high frequency is most of the times deadly, +1.35v with high overclocks would degrade over longer periods for many, no degradation reports I ever seen reported at 1.33v maximum for any speed. Probably, vdroop desactivation is playing a big role here with deadly voltage spikes. People should try to overclock at a maximum of 5% within intel specs: 1.36+5% = 1.43v as a max bios set without any vdroop disabled. For most boards, it'll give you a 1.33-1.35 real vcore after vdroop (my little finger is saying this 1.33-1.35v is not by case as it corresponds to most degradations feedbacks). The problem is that setting +1.35v with vdroop disabled, you'll exceed the 1.43v spikes at about thouthand times/sec

Unless we have more feedback or new steppings, that's what I'll do with my chip. If you have money to throw away, than you can of course bump 1.5v into them. Now, if people like living with some dreams that their CPU is not degraded while it needs a heat up period, they are free. But hey, a CPU needing heat up and think it's OS issue is really blindly denying the truth

Finally, the poll is a nice idea if many contribute with real objectiveness