Core i7 Vdimm/VTT Graveyard :D

[Jor3lBR]

3835A730 i7 940, 3rd memory channel died last week. Was running only 1,7V Vdimm 24/7, around 1,45-1,5V VTT. Used 2V for like 5 minutes when I first got it, but its IMC never even could handle 1600Mhz on the rams stable. So I am guessing its IMC was flawed to begin with... on phase change.

WoW are they so frail like that?

[jcool]

No idea, but like I said, I figure its IMC has always been broken to begin with. Got it brand new off my distributor so it can't be
a previous owner screwing with it.

[ColonelCain]

Hope you get a good one, Intel usually gives out some gems in RMA

Really?

Cause when I RMA'd my E8400 a month and a half ago (purchased originally last January), they sent me a C0 chip (didn't think that they even made them anymore). And, even more weird, was that the Batch # was VERY close to the one I RMA'd.

[TheKarmakazi]

Really?

Cause when I RMA'd my E8400 a month and a half ago (purchased originally last January), they sent me a C0 chip (didn't think that they even made them anymore). And, even more weird, was that the Batch # was VERY close to the one I RMA'd.

Just speaking from my experience Got a really nice QX6850 from RMA, and back in the day a Pentium D 805.

[saaya]

T_Flight, i still think its the bios setting vtt not vdimm, still havent heard anything from intel regarding xmp 1.2 btw

and so far there have been barely reports of dead i7 cpus, only a handful and those were tortured pretty badly or died in freak accidents it seems, so i think the vdimm sensitivity isnt all that bad

zads, ive never seen that intel certified xmp logo... :/
really... and we even have that ampx mem here but not one with the certification... it seems most xmp profiles are unoficial then, and not tested... we usually test xmp profiles that dont work properly on our intel reference boards first, and if they dont work there we contact the module maker before trying to fix a broken profile on our end...

little bump for the topic

used this voltages for testing my retail 965XE in P6T Dlx aircooled all of them over 10 hours for SPI 32M, still nothing goes wrong, nothing died, no degrades.

vcore: 1.40-1.58~
vqpi: 1.60-1.65~
vdimm: 1.80-1.96~

well thats only half a day
i think week or month numbers are more interesting

That is exactly what I have kept all my testing voltages at and with that range. Most of the testing was done under lN2(CB-120) and now with all 8 cores enabled CPU fails OCCT instantly at stock settings. If HT is turned of then it is fine.

965 retail 3836A767

going to rma this week.

btw, this is the second retail CPU I know of that this has hapened to that was seeing the voltage with in the limits you have posted.

hmmm that sounds like a vcore damage rather than vdimm/vtt though...
only 1.58 vcore? and it damaged the cpu? damn... how much time did you have it running at 1.58v?

3835A730 i7 940, 3rd memory channel died last week. Was running only 1,7V Vdimm 24/7, around 1,45-1,5V VTT. Used 2V for like 5 minutes when I first got it, but its IMC never even could handle 1600Mhz on the rams stable. So I am guessing its IMC was flawed to begin with... on phase change.

hmmm mark, could you check the cpu socket and make sure no pins are slightly bent or corroded? that could cause channels to stop working as well

how long did you run it with those volts?
only 1.7v vdimm with 1.45v vtt all the time?
thats really close vdimm and vtt... if that really damaged the imc thats bad news... hmmm

[Achill3uS]

well thats only half a day
i think week or month numbers are more interesting

run your chips a week/month long with ~1.58vcore/1.65qpi/1.96vdimm on air bro

[jcool]

hmmm mark, could you check the cpu socket and make sure no pins are slightly bent or corroded? that could cause channels to stop working as well

how long did you run it with those volts?
only 1.7v vdimm with 1.45v vtt all the time?
thats really close vdimm and vtt... if that really damaged the imc thats bad news... hmmm

Who the hell is mark?
Anyway I did run up to 2V Vdimm and 1,55V VTT for short periods of time (like 10 mins) but that shouldn't have caused it to die... as the temps were subzero. I couldn't detect any signs of corrosion on the CPU pins either.

[NBF]

I have been running 1.33Vcore/1.25qpi/1.7Vdimm 24/7 for almost 2 months. Not extreme but so far so good.
Also have benched 1.55Vcore/1.60qpi/2.00Vdimm on air several times, but never for more than a couple of hours.

[eva2000]

no probs with either of my 2x i7 920 3836A756 and vdimm up to 2.09v for benching with max CPU VTT around 1.34-1.65v and vcore up to 1.616v for benching all air cooled with TRUE120 (see sig links)

[Rammsteiner]

Reading thus far results I guess 1.4VTT ain't really an issue (regarding Intel's max 1.35VTT).

BTW, Im wondering whether this max 1.35VTT is based on something like that Vdimm should be within a certain VTT range? And I mean not like =< 0.5V, but like for example minimal 0.2 and maximal 0.5V? Ive no idea, it's basicly just a thought.

[BenchZowner]

At least 100 hours with 1.75Vqpi & 2.25Vddr on both my i7 920's that I use for the reviews.
No degradation or issues at all... at least not yet

[Prodigys]

So what are the good value's now ?

vQPI not over 1.35V (according to Intel)
and vDIMM not over 0.5V in range of vQPI (so not more then 1.85 ?) ...

So, if I would setup my rig, this is good:

vQPI on 1.35V
vDIMM on 1.85V
Vcore on 1.45V

???

[Hornet331]

QPI at ~1.4V is ok. Especial if you run more ram.

I run 12gb ram and need 1.375 to be 24/7 stable.

[ReverendMaynard]

At least 100 hours with 1.75Vqpi & 2.25Vddr on both my i7 920's that I use for the reviews.
No degradation or issues at all... at least not yet

haha, as usual.

[ANTC]

anyone know what gskill sets can normally take before crapping out

[Mescalamba]

anyone know what gskill sets can normally take before crapping out

I guess that 1.8V should be ok. Depends on chips used, really.. (remove heat-spreader, its for look anyway and take a look + hope you can put it back and RMA). 2V is usually max.. and some chips doesnt scale well with voltage, so as I said.. you need to know chips.

[Mescalamba]

..that those voltage problems and dead i7's is matter of motherboard used? Just idea.. cant test it. My guess is that DFi UT and Bloodrage users have like zero amount of dead i7's (at least not cause of high Vdimm/VTT).

Cause it looks like ppl sometimes have CPUs from like almost same wafer, but one dies and other one not and difference is mobo..

[liftedcj7on44s]

I ran my i7 920 on a MSI Eclipse X58 for 1 month and a EVGA X58 for about 2 weeks. On both setups i ran Vdimm at 1.9-2.0v and the CPU didnt care. The MSI board blew out one of the DDR dimms though.

[sakelover]

3835A730 i7 940, 3rd memory channel died last week. Was running only 1,7V Vdimm 24/7, around 1,45-1,5V VTT. Used 2V for like 5 minutes when I first got it, but its IMC never even could handle 1600Mhz on the rams stable. So I am guessing its IMC was flawed to begin with... on phase change.

I have been running over 1 month now 1.5Vcore and 1.5VTT 24/7 no problems whatsoever (Vdimm is only 1.6v though). I have 1 pig of an i7 920 that NEEDS 1.5v Vcore and 1.5v VTT to get 4.2Ghz stable, if I lower my voltages even 1 notch, P95 BSOD in less than 30min, even if I play with the other voltages or keep them on AUTO, still needs same Vcore/VTT! I got a bummer of a chip!

i7 920
Gigabyte X58 UD3R
6GB G.Skill 1600Mhz
9800 GX2
CoolerMaster V10
(V10: P95 loads to 80c at these volts (1.5v). Stock Clocks: V10 hardly breaks 50c or then TEC kicks in and drops below 50c, and then back up, keeps fluctuating 45-55c running P95)

btw
my chip needs such high volts, I am kinda hoping it will just go ahead and die!!!

[Chicken Patty]

I have been running over 1 month now 1.5Vcore and 1.5VTT 24/7 no problems whatsoever (Vdimm is only 1.6v though). I have 1 pig of an i7 920 that NEEDS 1.5v Vcore and 1.5v VTT to get 4.2Ghz stable, if I lower my voltages even 1 notch, P95 BSOD in less than 30min, even if I play with the other voltages or keep them on AUTO, still needs same Vcore/VTT! I got a bummer of a chip!

i7 920
Gigabyte X58 UD3R
6GB G.Skill 1600Mhz
9800 GX2
CoolerMaster V10
(V10: P95 loads to 80c at these volts (1.5v). Stock Clocks: V10 hardly breaks 50c or then TEC kicks in and drops below 50c, and then back up, keeps fluctuating 45-55c running P95)

btw
my chip needs such high volts, I am kinda hoping it will just go ahead and die!!!

is 1.5v full load or idle?

[sakelover]

is 1.5v full load or idle?

well 1.5v Vcore and VTT set in Bios, so slightly less at desktop/under load

[Chicken Patty]

well 1.5v Vcore and VTT set in Bios, so slightly less at desktop/under load

it takes me idle 1.405v for 4.2 GHz HT on, and it goes up to about 1.47v under 100% load. So I think you are not too bad.

[sakelover]

it takes me idle 1.405v for 4.2 GHz HT on, and it goes up to about 1.47v under 100% load. So I think you are not too bad.

thanks,
(this is without HT on btw)
anyway still not all that impressed with my chip
but oh well, it is what it is
I will use it till it dies or a newer 1366 gen chip comes out
(wanna buy a i7 920? lol)

[sakelover]

it takes me idle 1.405v for 4.2 GHz HT on, and it goes up to about 1.47v under 100% load. So I think you are not too bad.

what are u talking about?
are u saying EVGA X58 mobo go up in voltage when loaded?
every mobo I've own (evga inclueded) go way down in voltage when loaded
meaning I just cheaked and when I set 1.5VCORE and VTT in BIOS
I see 1.47v in windows
and
1.42v Vcore when loaded.

since when did mobo's start upping voltage under load?

btw
this is with VDROOP control enabled.

[zlojack]

what are u talking about?
are u saying EVGA X58 mobo go up in voltage when loaded?
every mobo I've own (evga inclueded) go way down in voltage when loaded
meaning I just cheaked and when I set 1.5VCORE and VTT in BIOS
I see 1.47v in windows
and
1.42v Vcore when loaded.

since when did mobo's start upping voltage under load?

btw
this is with VDROOP control enabled.

The EVGA X58 does raise the voltages under load.