NEED HELP: Two P8P67 EVOs & 2 2600Ks DEAD within 10 months. PLEASE READ CAREFULLY.

[MaxAwesome]

Hello guys, I'm at a loss for words as to what is happening to my hardware.

Back in January 2011, I got my 2600k and P8P67 Evo. Upgraded from Q9550 & P5Q-Pro

All was fine and dandy until February 11th, my computer simply shuts down during a CS 1.6 match, and never turned on again.

The symptoms were as follows:

- Computer shuts down out of the blue, does NOT power on.

- Green LED on mobo, no power on at all. Fans spin for a 1/4 of second, but nothing happens.

RESULT: Dead motherboard, AND as I found out on March 30th when I got my revision B3 Evo, also a dead 2600k CPU (there were no mobos to test it with during the recall, but it died together with the mobo).

Also, at the time of this first hardware failure, I had a mid-range 700w PSU, and I upgraded to a Corsair TX650 just to be on the safe side. It has been my PSU since March.


Bear in mind, my computer went dead during the P67 and H67 recall, due to the Intel bug thingy, so I was without a functional computer until March 30th 2011. Then, I got a B3 P67 Evo.

Flashforward to last sunday:

Same thing happens, not during CS 1.6, but during 3Dmark Vantage this time.

SAME SYMPTOMS:

Computer shuts down, no power on, fans spin for a fraction of a second. Turns out, dead mobo and CPU again :/

Hardware tested at my local store where I bought the stuff. Exactly the same problem. Both CPU and mobo die at the same time with no apparent reason.

When it happend for the fist time in February, I didn't think too much of it, I thought it was maybe just a bad mobo that died and took the CPU with it.

Now I'm worried, because it happened a second time, with the same symptoms.


For the record, my OC on the 2600k during both times was (IMO) an extremely harmless one:

- 2600k @ 4.2ghz with 1.24v under LOAD, with offset vcore set with MINUS 0.45v

- All other voltages, except RAM, were on AUTO

- RAM was running @ 1600mhz with stock timings 9-9-9-24 @ 1.55v

- C3 and C6 states were DISABLED, EIST and all other power options were ENABLED.

Rest of my system specs:

CPU Cooler: CM TX3 (the main reason my OC was so gentle). Besides, 4.2ghz is NOTHING for a 2600k.

PSU: Corsair TX650w

RAM: 2x2 + 2x2 Kingston HyperX 1600 CL9 1.65v (they run at stock speeds @ 1.55v)

GPU: Sapphire HD 6870


Bottom line:

So guys, I'm really baffled by all this. Am I doing something wrong? This is the second RMA I go through since January... I really need help.

You may laugh, but I'm actually SCARED of overclocking now, since my hardware has died on me, and it has never happened before.

I am an experienced OCer, and honestly I don't think I'm doing anything wrong. But since it has happened 2 times now, I can't just be coincidence, right? :/

Opinions? I really need help. I can't figure out what's killing my hardware.

Please don't forget that BOTH the CPU and the Mobo die at the same time. PC shuts down, never to turn on again.

Also, I've ruled out PSU causes, since the first time and this last one happened with 2 different PSUs. One with my previous PSU, and now with my trusty Corsair TX650.

[erocker]

You must have bad electricity or something. Get a quality UPS unit to plug your PSU into.

[Paulieg]

Originally Posted by erocker

You must have bad electricity or something. Get a quality UPS unit to plug your PSU into.

Damn it erocker, I was just going to say the same thing. How old is the wiring in your home? It seems the most likely cause.

[MaxAwesome]

My house is just 10 years old, so I think the electrical system is fairly modern.

All the sockets are properly grounded of course.

Besides, I also thought of power problems, but before this, I've had my P5Q-Pro system for 2 years, on the same sockets, in the exact same room. Even the same case.

Zero problems.

I'm baffled, and I really need help. Sorry if this is bothersome.

[erocker]

Originally Posted by MaxAwesome

My house is just 10 years old, so I think the electrical system is fairly modern.

All the sockets are properly grounded of course.

Besides, I also thought of power problems, but before this, I've had my P5Q-Pro system for 2 years, on the same sockets, in the exact same room. Even the same case.

Zero problems.

I'm baffled, and I really need help. Sorry if this is bothersome.

I honestly can't think of anything unless there's just a problem with that particular motherboard. Maybe you're attatching the CPU heatsink too tight? Really, no idea.

[MaxAwesome]

Originally Posted by erocker

I honestly can't think of anything unless there's just a problem with that particular motherboard. Maybe you're attatching the CPU heatsink too tight? Really, no idea.

Could it be that I have slightly inconsistent power delivery that my PSU won't filter out, and kills this new hardware?

Maybe the old LGA 775 stuff was more resilient?

For the life of me, I can't figure it out... and the store technician was just as baffled as I am when the same thing happened. He also handled the previous RMA.

He asked if I overclocked, I mean... I have a 2600k... is that even a question? It's meant to be overclocked.

He said my OC was supposedly 100% safe, something I already knew.

[phanbuey]

PSU is killing your stuff.

[Sasqui]

I KILLED one motherboard and seriously wounded another with a faulty Hiper TypeR power supply. I swapped EVERYTHING out, before trying a different PSU and finally finding out that was the problem. So... don't discount it.

It was the most unpleasant computer issue I've ever encountered.

[MaxAwesome]

Originally Posted by phanbuey

PSU is killing your stuff.

What are the chances of 2 different PSUs (one mediocre, on quality) causing the same thing, on the exact same hardware (both CPU and mobo) and nothing else?

I'm not saying you're wrong, I'm genuinely interested in knowing :/

First time: Nox Urano 700w (probably unknown in the US, widely used in Europe as a mid range PSU.)

Second time: Corsair TX650.

Two different PSUs, in 2 totally different leagues. What are the chances?

I still think it's something else

[Crap Daddy]

I have the same TX650. I stand by it's quality. I have frequent power cuts in my area, sometimes short bursts on/off and no problem. Once even all the electronical/electrical equipment in the house went out at a power drop while the machine powered by the TX650 didn't felt a thing.

It's bad luck maybe? Try another mobo manufacturer.

[Sasqui]

Originally Posted by MaxAwesome

What are the chances of 2 different PSUs (one mediocre, on quality) causing the same thing, on the exact same hardware (both CPU and mobo) and nothing else?

I'm not saying you're wrong, I'm genuinely interested in knowing :/

First time: Nox Urano 700w (probably unknown in the US, widely used in Europe as a mid range PSU.)

Second time: Corsair TX650.

Two different PSUs, in 2 totally different leagues. What are the chances?

I still think it's something else

I tried to read throughly and missed that myself. It's is unlikely the PSU in that case.

Do you have a UPS you can try with?

Originally Posted by Crap Daddy

It's bad luck maybe? Try another mobo manufacturer.

Shit indeed does happen.

[TRIPTEX_CAN]

This is a long shot.. but did you make sure nothing is behind the mobo in the case causing a short?

[erixx]

Maybe a short between case and mobo-base, rare but...

If not, I would just say: FUCKING BAD LUCK, just RMA and keep on!!!!!

PD: I had 2 ASUS mobos die in front of me. Really nobody knows. The best is quickly RMA board and cpu without saying much hoping you get new parts!

[phanbuey]

Originally Posted by MaxAwesome

What are the chances of 2 different PSUs (one mediocre, on quality) causing the same thing, on the exact same hardware (both CPU and mobo) and nothing else?

I'm not saying you're wrong, I'm genuinely interested in knowing :/

First time: Nox Urano 700w (probably unknown in the US, widely used in Europe as a mid range PSU.)

Second time: Corsair TX650.

Two different PSUs, in 2 totally different leagues. What are the chances?

I still think it's something else

Yeah not high actually, I was under the assumption that it was the same PSU (didnt read srry).

In which case it might be that model of Mobo... i.e. like if you overvolt a 590/570, it will die.

I think the first two posts were right... the power in your house is could be unhealthy (old circuits, etc) and this model of board may be more sensitive to it than your old one... i dont think it is a short - shorts usually don't last months before appearing, they are a bit more instantaneous.

I would swap the mobo out for a different model/make (if possible). That is crappy luck tho.

[Crap Daddy]

On the other hand I use my 2500K at 4.2 turbo, all on auto. Works like a charm. Don't have to tweak anything for that. Was thinking that maybe your RAM voltage is out of spec. - max 1.5V on Sandy - but I used mine for some time at 1.65V nothing happened and other still go as high.

[cadaveca]

Originally Posted by MaxAwesome

RAM: 2x2 2x2 Kingston HyperX 1600 CL9 1.65v (they run at stock speeds @ 1.55v)

That's enough, right there, methinks. I mentioned maybe last week in another thread, that it seems Intel is now denying RMAs based on using XMP-rated memory on SB.

Nevermind, Intel does have a list of XMP-compatible motherboards and memory, and neither of yours is on that list, which can be found here:

http://www.intel.com/content/www/us/...datasheet.html

Until you use "approved" parts, we can simply blame your overclocking for the premature death.

Could be the power in your house too, for sure. Could be many things, but until you stop overclocking, you'll never know. There's a reason overclocking voids your warranty, after all.

[phanbuey]

^^I think it would be MUCH more widespread if that was the case. So widespread that there would be "RAM: Can kill your Sandy Bridge!!!" articles and posts all over the web.

Again, totally possible but just my 2c.

[MaxAwesome]

Originally Posted by cadaveca

That's enough, right there, methinks. I mentioned maybe last week in another thread, that it seems Intel is now denying RMAs based on using XMP-rated memory on SB.

Nevermind, Intel does have a list of XMP-compatible motherboards and memory, and neither of yours is on that list, which can be found here:

http://www.intel.com/content/www/us/...datasheet.html

Until you use "approved" parts, we can simply blame your overclocking for the premature death.

Could be the power in your house too, for sure. Could be many things, but until you stop overclocking, you'll never know. There's a reason overclocking voids your warranty, after all.

Well, holy crap. I never knew about this. Could my RAM be killing my stuff?

Sorry for the noobness, but what is XMP-based RAM?

[cadaveca]

Originally Posted by phanbuey

^^I think it would be MUCH more widespread if that was the case. So widespread that there would be "RAM: Can kill your Sandy Bridge!!!" articles and posts all over the web.

There has been a greater level of prevelence of such posts on various forums, including Intel's own.


If RMAs are being denied simply for use of XMP ram, I'm sure it's coming, and I'm just a bit ahead of the game. No different than earlier 1366 chips. Some chips can handle it, some cannot, and there are caveats as to what will cause the death, and what won't.

I mean really now, these chips have been out for like 10 months only. I think it's far too early to simply state it's not a possible issue. There's gotta be good reason for RAM OEMs releasing a bunch of 1.5v sticks for SB, and this wouldn't be the first time with this platform tht voltage has led to problems.

[MaxAwesome]

Originally Posted by phanbuey

^^I think it would be MUCH more widespread if that was the case. So widespread that there would be "RAM: Can kill your Sandy Bridge!!!" articles and posts all over the web.

Again, totally possible but just my 2c.

Well, yeah... I tend to agree with you, though.

Besides, one thing I noticed is that my RAM is rated @ 1.65 on the box, but the JEDEC specs on CPU-Z say 1.5.

What gives? Misreading? or Mislabeling?

[cadaveca]

Originally Posted by MaxAwesome

Sorry for the noobness, but what is XMP-based RAM?

Any memory for the Intel platform that is NOT 1333 MHz or less, and even some of those sticks, are rated as "XMP-enabled", meaning they have a specific profile within thier SPD ROMs for higher speeds, and sometimes, voltage.

Your memory is XMP-rated.

The profile that lists 9-9-9-24 and 1.5v @ 1333 MHz, which most people refer to as "JEDEC" is STOCK for SB chips, and anything higher than 1333 MHz is an overclock on the CPU that can, and has, in some instances, been reason for RMA being denied.

If you check out INtel's XMP website, they wil ltell you in a disclaimer at the bootom that Intel does not support XMP, and it's use will invalidate your warranty.

[phanbuey]

Originally Posted by MaxAwesome

Well, yeah... I tend to agree with you, though.

Besides, one thing I noticed is that my RAM is rated @ 1.65 on the box, but the JEDEC specs on CPU-Z say 1.5.

What gives? Misreading? or Mislabeling?

those JEDEC specs are common specs, alot of 2000 MHz 1.65 sticks show up at 1600 cl9 with 1.5v rating. I wouldn't worry about that.

Although to cadaveca's point, ram is pretty cheap these days, might want to swap it out for good measure as it was a common component across both builds.

[Crap Daddy]

I have a pair of HyperX 1600 at 1.65V. Have run XMP profile with 1.65V and 1.6V. Nothing appeared to be wrong. I backed down a few months ago and now I'm at 1333, 1.5V. Just to be on the safe side. The difference in real life usage is zero.

[cadaveca]

Originally Posted by phanbuey

alot of 2000 MHz 1.65 sticks show up at 1600 cl9 with 1.5v rating.

Exactly, and that's why Intel is denying RMAs with some XMP-rated sticks, as even the JEDEC profile that is active is an "overclock", and use of such sticks indicates that the CPU was NEVER run @ stock.

Originally Posted by Crap Daddy

The difference in real life usage is zero.

Which is why I have no problem telling people to run 1333 MHz. Benchamrks are one thing, daily usage is another. If ya wanna clock up your ram to run benchmarks, go right ahead! If something dies, you have a good reason as to why. I don't expect such things to kill a chip, but there seems to be a risk, and for the minimal gains, I don't see much point in 24/7 memory OCing.

[phanbuey]

Originally Posted by cadaveca

Exactly, and that's why Intel is denying RMAs with some XMP-rated sticks, as even the JEDEC profile that is active is an "overclock", and use of such sticks indicates that the CPU was NEVER run @ stock.

I was always under the impression that the MB would override JEDEC regardless. But then again I do not have an SB and am still on an old school BIOS.

On my setup, at least, JEDEC is a bit like a speed limit sign on a Miami highway - to be ignored immediately by the other components.