Dead retail 980X? - please post here

[Rosty]

my 980x just dead here last week

[Falkentyne]

And this is useless without details....
I mean you could have static shocked it or drowned it in acid for all we know....

Details? Conditions? vcore? cooling?

[bratboy]

I haven't gotten around to installing mine yet, having a problem with my mem right now but I did look on the box and mine says Build 3003B396. Is that a good build as I seem to have seen a few with an F in their build having reported bad luck. Not found a good reference for comparisons of different builds but I may be searching wrong. Also, other than by installing the CPU is there any way to determine the stepping?

[BenchZowner]

This is a thread for Retail 980X death reports, not a batch genie or the place to talk about the 980X in general.

[BobyTT]

So any dead retail 980X already?

[=[PULSAR]=]

Dead after 5 days. Was running WCG @ 100%

Retail B1

frequency - 4.4ghz
motherboard - evga E760
Batch # - 3003A885
max Vcore - 1.4v
max VTT - 1.2
max loaded Vcore - 1.4
max loaded VTT - 1.2
operating temprature - 70-75C

I'm guessing the heat killed it

[Sam_oslo]

Dead after 5 days. Was running WCG @ 100%

Retail B1

frequency - 4.4ghz
motherboard - evga E760
Batch # - 3003A885
max Vcore - 1.4v
max VTT - 1.2
max loaded Vcore - 1.4
max loaded VTT - 1.2
operating temprature - 70-75C

I'm guessing the heat killed it

Or maybe the vCore did it? Intel has reduced the max vCore to 1.4v for 32nm, maybe with a good reason? it seams like it may not be good for 5 days at 100%. Disappointing anyways.

[eva2000]

Dead after 5 days. Was running WCG @ 100%

Retail B1

frequency - 4.4ghz
motherboard - evga E760
Batch # - 3003A885
max Vcore - 1.4v
max VTT - 1.2
max loaded Vcore - 1.4
max loaded VTT - 1.2
operating temprature - 70-75C

I'm guessing the heat killed it

ouch, what vdimm used ? intel vcore droop enabled or disabled ?

[=[PULSAR]=]

Or maybe the vCore did it? Intel has reduced the max vCore to 1.4v for 32nm, maybe with a good reason? it seams like it may not be good for 5 days at 100%. Disappointing anyways.

Plenty of people run them @ 4.5ghz for WCG w/ 1.45vcore on water so thats what made me think temps killed it. Temp differences were pretty drastic also 8C difference between cores after multiple remounts, may have been a badly placed IHS.

ouch, what vdimm used ? intel vcore droop enabled or disabled ?

1.65V in Bios 1.67V via DMM, vcore droop was set to disabled.

[kill_a_wat]

Plenty of people run them @ 4.5ghz for WCG w/ 1.45vcore on water so thats what made me think temps killed it. Temp differences were pretty drastic also 8C difference between cores after multiple remounts, may have been a badly placed IHS.



1.65V in Bios 1.67V via DMM, vcore droop was set to disabled.

wow interesting. the volts you used dont seem to bad either...

[kitfit1]

vcore droop was set to disabled.

The answer to the early death probably lays there. Disabling Vcore Droop is ok for benching in the short stints that are the norm. But running a cpu flat out for 5 days with Vdroop disabled is a disaster waiting to happen.

[rge]

The issue is not disabling vdroop, it is just you have to factor in the equivalent voltage with it off. Intel max specs refers to bios setting of 1.4 (bios = max overshoot vcore), and since overshoot will be ~0.1 on these (granted vary board to board etc., using round number for ease), intel specs load is probably ~1.3 max.

If you disable vdroop, and set to 1.4, overshoots will still be .1, so now to 1.5. But with vdroop disabled, your cpu sees less resting vcore.

You could argue 1.3 with vdroop enabled is better than 1.4 with vdroop off, since overshoots same to 1.4, but with vdroop on means idles lower at 1.3 vs 1.4 with vdroop off. But when increasing to 1.4 with vdroop off, now your out of specs, intel was assuming you would load at 1.3vcore by placing 1.4 bios limits.

When 45nm first came out, there were several research papers saying biggest issue was interconnect variability and hence varying quality in early process, and nearly 1 in 4 ocers that posted in links were claiming degradation. Wonder if 32nm will go through same issues early on, only to be sorted out later as process matures, like what happened with i7s.

[BlindFreddie]

@ PULSAR: You said in It's crunch time!, that you were running 1.35 Vcore:

Yep...soo my 980x just bit the dust? WTF!!! Was running it 1.35vcore 1.2VTT @ 4.3ghz for a week...the voltage was well within intels specs.

1.35 or 1.40?

Retailers with 980X's in stock are like hens' teeth here in Oz, and those that say they have them quote inflated prices. Do you think Intel is holding back production until they sort out the dying chip problems, or is their max production capacity still rather limited?

[=[PULSAR]=]

@ PULSAR: You said in It's crunch time!, that you were running 1.35 Vcore:

Yep...soo my 980x just bit the dust? WTF!!! Was running it 1.35vcore 1.2VTT @ 4.3ghz for a week...the voltage was well within intels specs.

1.35 or 1.40?

Retailers with 980X's in stock are like hens' teeth here in Oz, and those that say they have them quote inflated prices. Do you think Intel is holding back production until they sort out the dying chip problems, or is their max production capacity still rather limited?

I was running it at 1.35V and 4.3ghz for 5 days and after a couple hours of 1.4V 4.4ghz it died. So the majority of its life was at 1.35V.

[miahallen]

Thanks Z....please redirect your results to the thread linked, and we'll let this one die

Thanks

Dead after 5 days. Was running WCG @ 100%

Retail B1

frequency - 4.4ghz
motherboard - evga E760
Batch # - 3003A885
max Vcore - 1.4v
max VTT - 1.2
max loaded Vcore - 1.4
max loaded VTT - 1.2
operating temprature - 70-75C

I'm guessing the heat killed it

Ouch dude....that sucks