KILL your i7 980X AT ONCE (and RMA it)...

[xpatar]

tryed 2c2t and 1c1t for spi 6.2ghz 1.875v on my a0 with RIIIE still works daring arent I?

[hipro5]

I finally lost my mobo too during doing so.....
I thought that it might be the bios chip corrupted but after placing another bios chip on, it didn't boot too......FF.....

[eva2000]

George could it of been motherboard taking out the cpu ??

[[XC] Oj101]

His friend's one as well?

[Sam_oslo]

Even though at first I wouldn't even try it, I decided to give it a try.

What the ........ If anything goes wrong I'm going to rma

Currently on air and everything @ stock speed.

Bios v. 1003, batch number 3003B320.

Bios settings: 1 core HT off,

http://i286.photobucket.com/albums/ll104/mpamphs70/INTEL%20980X/1COREHTOFF-32M.jpg[/IMG]

CPU still alive

...

Good news. When you get time, can you try this 1C/1T with high-BCLK, BCLK210+ ? Then it can actually disprove my theory about the big voltage bump that is caused by 2-speed-bins and high-BCLK combination.

[xpatar]

Without oc you do get with 1 core or 2 cores HT off, a 2 speed bins (266 MHz) boost, Vcore keeps steady @ 1.24v at load.
With higher BCLK I'll have to check it out.

[hipro5]

George could it of been motherboard taking out the cpu ??

One died instantly getting mobo with it on an EVGA 760.
One after one seesion with 2 cores without HT on GIGABYTE UD7
...and my friend's instantly on a UD7 one too with 1 Core and no HT...
I don't know if all these are coincidence!.....

[Movieman]

One died instantly getting mobo with it on an EVGA 760.
One after one seesion with 2 cores without HT on GIGABYTE UD7
...and my friend's instantly on a UD7 one too with 1 Core and no HT...
I don't know if all these are coincidence!.....

I'm staying on the dualie boards with the xeons where it's safe!

[eva2000]

One died instantly getting mobo with it on an EVGA 760.
One after one seesion with 2 cores without HT on GIGABYTE UD7
...and my friend's instantly on a UD7 one too with 1 Core and no HT...
I don't know if all these are coincidence!.....

hard to say maybe too much LN2 loving for the cpus ?

[damha]

As if anyone who knows george in person wouldn't have touched any of their BIOS settings!!

I am willing to assume there is another factor with another BIOS setting because of the reports that the CPU worked for some after this.

It could also be a flaw in gigabyte/evga boards.

[ci2kla]

yup I have killed a 980x before...just by undervolting...it will POST but never get into BIOS...have to say they die quite easily

[principino1984]

same thing here... just playin with my 980x with no HT enabled and one core...now it enters windows only with 6cores and no HT, or 1 or 2 cores and HT enabled...

damn it!!!

Marco

[Falkentyne]

It would help if you gave us the batch # of the chip and the motherboard in use, and not just saying your 980 died.

[principino1984]

sorry..my bad...

3951a763

killed on Ramp III Extreme

Marco

[Falkentyne]

Ok that's not good...
Considering the R3E natively supports that processor...
So there is a proven hardware flaw in these processors, then...

What's the highest voltage (vcore) you put , and VTT/QPI, etc....through the processor (even momentarily?)
And what's the highest speed it was run at?
Was it ever on subzero?

[dinos22]

strange thing George

I've tested a couple of chips on my UD7 but they are ok (ES chips though)

UD9 too but also ES chips too .... as mentioned it might be a retail chip thing but how come xpatar didnt kill his one....odd

[radaja]

Miravo,had his die after 15m of using?
Official Intel Gulftown (Westmere) Overclocking and Batch Tracking Thread post 108

[hipro5]

Yes these CPUs are crazy.....They degrade VERY FAST above 1.75VCore (thus ~1.9VCore to Yorkfields)......It reminds me very much the Yorkfields reactions above 1.85VCore.....

dinos, the first ES I touched in my hands, 2.4GHz A0, I pushed it a bit hard and I didn't noticed "strange reactions" until I sent it back......
The second one though which was a 3.33GHz one, after a session, it was hard to bench it again.....A problem to it's memory controler happent....

[xpatar]

The last days I've playing around with 1 core HT off @ stock.
Playing games, running prime 95, 3d mark vantage, 3d mark 2006, furmark, still haven't notice any degradation or a sign of killing my cpu.



Very strange some people faced issues
Has anyone read any official report from Intel?

[hipro5]

You are lucky.....

To be honest, I will NEVER again disable any cores unless I want to gumble and kill my cpu once more....

I'll let it degrade as is and not pushing it with 1 or 2 cores.....I'll use all 6 cores of it and I will only disable/enable the HT and that's it...

[monteboy]

I can confirm all that hipro said.

2 cores active , HT off , Gigabyte UD7 (Bios F7m)
My Gulftown Batch : 3003B175 , run 5.8GHz Vantage easy !
Last day's i did some 2d run's , MaxxMem for GOOC and spi1m (2 active cores)

Yesterday on Ln2 all cores active,HT on (Wprime32) , 1,6V 5.5 GHz -> BSOD
Yesterday on Ln2 all cores active,HT on (Wprime32) , 1,6V 5.4 GHz -> BSOD
Yesterday on Ln2 all cores active,HT on (Wprime32) , 1,6V 5.1 GHz -> BSOD

today .... MB did'nt POST -> CPU dead

Epic fail !

[hipro5]

Fluck!.....

I'll tell you what!.

UD7 doesn't like 2 cores and RIII and EVGA 760 don't like 1 core or it's dead.....

Those CPUs are really crazy, I don't know what else to say...

[ci2kla]

Fluck!.....

I'll tell you what!.

UD7 doesn't like 2 cores and RIII and EVGA 760 don't like 1 core or it's dead.....

Those CPUs are really crazy, I don't know what else to say...

What the hell?? you've killed more 980x?..........
In my case, I lowered all voltage out of range that MB can set, save profile, reboot = dead...

[mbreslin]

3003B396: 1c/1t linx 20 pass
x58a-ud7 f7p

Also on the high end I've done 6c/12t 6ghz 1.62 vcore 1.45 vtt 5ghz+ temp 90+ (unchilled water)

CPU fine after both. I was scared to try the 1c thing but if it died I would have an excuse to get an sr-2.

[dinos22]

this is really interesting (terrible to hear of course guys)

what did intel change i wonder in retails