E8400/8500 degradation myth possibly busted?

[CrazyNutz]

That's because you have safe vdroop and vdrop. On boards with vdroop and vdrop mods, it's easier to stabilize the CPU with low voltage, but it might be easier for spikes to appear as well. Double-edged blade, I guess...
But I'm going to test the vdroop/vdrop mod = spikes theory tomorrow.

I think that trying to catch one of these spikes (reliably) with a software app
is going to be difficult, obviously not impossible but difficult. Fist off the sampling rate of most software volt monitors is very low >1sec. these spikes
probably last only a few useconds (micro second = 1/1000000th of a second).
So a ton of these spikes can/will occur undetected. It would be best If you had access to a good oscilloscope (I had one, but it's gone now , long story). Anyhow I'll try my best to catch one as well.

[Leeghoofd]

Are these Vspikes only happening to Anus boards ????

No Spikes here, smooth as babyskin :

[CrazyNutz]

Well, I finally got it!!! It took awhile but here's 3 massive spikes.




EDIT: BTW this is with Load Line Calibration Off, so normal/intel spec vdroop/vdrop is in effect.


UPDATE:
My 12v dropped at the same exact time the first vcore spike occurred. Indication of my psu failing
and causing these spikes? Maybe! Could be if the psu 12v rail is failing and the VRM may be overcompensating the
vcore causing the spikes.

[RunawayPrisoner]

It did not need to take a while? And now you're scaring me off "trying" because of those spikes. 1.65v+ SPIKE? I'd be lucky if my chip could still run at stock then. D:

[mrcape]

I don't get the stats at the top of the graph. What is your clock when getting these spikes?

[CrazyNutz]

I don't get the stats at the top of the graph. What is your clock when getting these spikes?

OCCT is reading it wrong because I have C1E on, it's actually 3870mhz

[Eldonko]

Yeah the 2.1v spike I saw was over many days of running speedfan 24/7. Scary thing is it happened when I was at work so it was 100% random and had nothing to do with anything I was running. Ahh good ol asus.

[RunawayPrisoner]

Did you mod your motherboard in any way... by any chance?

[CrazyNutz]

Did you mod your motherboard in any way... by any chance?

Nope, It's 100% stock. 0 mods

[CrazyNutz]

Yeah the 2.1v spike I saw was over many days of running speedfan 24/7. Scary thing is it happened when I was at work so it was 100% random and had nothing to do with anything I was running. Ahh good ol asus.

Are you still using the asus board, or do you use something else now?

Man I really want to get this board out of my system now!!! Although
I also supecting my PSU.

[RunawayPrisoner]

Nope, It's 100% stock. 0 mods

That is hella weird. And I am having another theory:

It might not be that the spikes are degrading the chip... it might be that... the spikes are causing unstable moments, thus making it look like your CPU is failing, whereas it's actually stable! From my own experience, smooth graphs would be very stable, and those with slight distortions are more subject to instablity... even when the vCore is enough, a slight movement in 12V rail would cause the test to fail immediately.

[CrazyNutz]

That is hella weird. And I am having another theory:

It might not be that the spikes are degrading the chip... it might be that... the spikes are causing unstable moments, thus making it look like your CPU is failing, whereas it's actually stable! From my own experience, smooth graphs would be very stable, and those with slight distortions are more subject to instablity... even when the vCore is enough, a slight movement in 12V rail would cause the test to fail immediately.

Could be, however even with that huge spike/s, OCCT still ran a "Stable" test.

I would bet these spikes are occurring a lot more than what the graphs are showing. I'm going to try to barrow a oscilloscope so I can see the actual
frequency they are happening.

[RunawayPrisoner]

Sounds good. Thanks for your efforts, and I am awaiting further results from you.

[Eldonko]

Are you still using the asus board, or do you use something else now?

Im done with ASUS, I went through 4 P5K and moved on to DFI T2R. I find the prob with asus is you have to buy the $300+ boards to get decent components and quality. Im just tired of the vreg issues and cheap transistors in the affordable boards.

[MosIncredible]

Is it possible for one core to degrade? My 2nd core is horrible. Whenever I prime, it always fails at least 5 hours before my 1st core, and of course, its ALWAYS the one that causes Orthos to fail.

[sfdmalex]

On the subject of spikes and dips.

I changed my power supply, an older noisetaker enermax 495W to the Corsair HX Series 620W and guess what, not one spike or dip on OCCT. And Vcore 100% stable.

[dnottis]

On the subject of spikes and dips.

I changed my power supply, an older noisetaker enermax 495W to the Corsair HX Series 620W and guess what, not one spike or dip on OCCT. And Vcore 100% stable.

Yea my TT 1000 watt doesnt flux at all.

[alpha0ne]

Im done with ASUS, I went through 4 P5K and moved on to DFI T2R. I find the prob with asus is you have to buy the $300+ boards to get decent components and quality. Im just tired of the vreg issues and cheap transistors in the affordable boards.

Ditto

If you need to send an Anus board back RMA forget it, they would do ppl a huge favour if they just told ppl the truth.................that is don't send broken/faulty boards back coz we here @ Anus don't give a toss

[Leeghoofd]

Ditto

If you need to send an Anus board back RMA forget it, they would do ppl a huge favour if they just told ppl the truth.................that is don't send broken/faulty boards back coz we here @ Anus don't give a toss

Well that differs from country to country it seems, I always had my RMA boards back in proper working condition.( Asus , DFI and Abit... got one RMA for Gigabyte now see how that one works out ) k some took 5 weeks but i got them back, and for the last my striker extreme I got a very nice refund... no need to generalize. Everybody will have his bad things with each brand... no need to post it all over the place... over and over again...

[dnottis]

Asus boards have always been the best overclockers for me. Never had any real issues with an Asus board. My GB boards were garbage, Abit usually are too. Maybe the spikes have nothing to do with the board but are PSU related. My TT toughpower 1000 doesnt flux at all during OCCT. The other poster changed PSU and notice no spikes either.

[CrazyNutz]

On the subject of spikes and dips.

I changed my power supply, an older noisetaker enermax 495W to the Corsair HX Series 620W and guess what, not one spike or dip on OCCT. And Vcore 100% stable.

I do suspect the PSU, and I have a 550w in a server I am going to pull and try
out today. I'll post the results when I do. Also I did not score an oscilloscope
since my buddy who had one no longer has it. oh well!

I also left the speedfan vcore, and +12v monitor graphs running over night,
and no spikes were recorded.

[dnottis]

Back on topic - degradation.

Alrigh I've now played with an E8400 and Xeon E3110 and have a few things to add about degredation.

- After running higher volts then coming back to low volts it will fail prime at a previously stable OC. After a few days this seems to work itself out however and I can prime the lower volts I was before. (Im not even talking high volts, I'm talking 1.4v). I actually found that in a few cases resetting the CMOS and starting over fixed this.

- I had issues where orthos was actually crashing. I'd get Orthos has stopped responding errors but the log showed it running then it just stopped. The chip never gave an error the app just quit. I was pushing 4.3Ghz, so I went back to my previously orthos stable OC 4.1Ghz and same crashes. I scrapped my orthos folder, re-downloaded it, kept doing it. To fix this issue I reset the CMOS and started over. Problem solved.

Like most people I start low and start working up. Once I get a failure with prime things get inconsistent. Lately I just reset the CMOS, go back in and reload the profile that was previously stable. This has been working quite well and even after failing orthos on an overclock it will be fine at the previous setup again.

I don't think there is degradation, just some kind of goofiness that starts once you fail pushing an overclock. My suggestion would be to reset the CMOS and load a previously stable OC - if this still doesn't work, give it a day or two and Im sure you will be able to prime a lower volt OC like before.

[mrcape]

Just an update...

re: spikes - I ran occt last night at 500fsb+ / 4.3ghz and saw no spikes on the reports. I'm on the DFI in my sig.

About clearing Cmos, I've tried this many times. I've also gone as far as re-flashing the bios, fresh windows installs, fresh ram kit, etc.. I still can't get to my original clocks/low volts (-.05v lower). I've exhausted every method to try and get them back. For the past weeks I've moved on to testing to see if the chip is sliding further. Good news -No further sliding downhill at all after a month of re-testing. More good news - I did see a tiny hint of recovery (.01v overall).

I'm also running my raptors as totally separate system drives, keeping them very clean to get accurate tests and comparing them. I re-install each probably twice a week. I find that after beating the box for a while back and forth, software/OS/bios condition definitely makes a difference.

[Gr1mR34p3r]

No degradation with my E8400 Q745A822, been running 1.55V set in BIOS, CPU-Z load around 1.5V, and even to 90°C+ with stock cooler.
It still passes Orthos 445x9 1.375V, just like out of the box.
Of course I will be more safe in the future, max 1.40V.

[RunawayPrisoner]

Back on topic - degradation.

Alrigh I've now played with an E8400 and Xeon E3110 and have a few things to add about degredation.

- After running higher volts then coming back to low volts it will fail prime at a previously stable OC. After a few days this seems to work itself out however and I can prime the lower volts I was before. (Im not even talking high volts, I'm talking 1.4v). I actually found that in a few cases resetting the CMOS and starting over fixed this.

- I had issues where orthos was actually crashing. I'd get Orthos has stopped responding errors but the log showed it running then it just stopped. The chip never gave an error the app just quit. I was pushing 4.3Ghz, so I went back to my previously orthos stable OC 4.1Ghz and same crashes. I scrapped my orthos folder, re-downloaded it, kept doing it. To fix this issue I reset the CMOS and started over. Problem solved.

Like most people I start low and start working up. Once I get a failure with prime things get inconsistent. Lately I just reset the CMOS, go back in and reload the profile that was previously stable. This has been working quite well and even after failing orthos on an overclock it will be fine at the previous setup again.

I don't think there is degradation, just some kind of goofiness that starts once you fail pushing an overclock. My suggestion would be to reset the CMOS and load a previously stable OC - if this still doesn't work, give it a day or two and Im sure you will be able to prime a lower volt OC like before.

Sounds to me like it's the mobo's fault now... or maybe a bug in premature 45nm BIOSes. Which reminds me, my mainboard also gives different results with the same BIOS settings... after each CMOS reset. I've found some consistent or at least near perfect settings that wouldn't give much of a difference whatsoever, but +-.01v away from any of the voltage options, and I'd see the system acting funky.