Elpida Hyper Graveyard

**Bei Fei** · 06-26-2009, 01:21 PM

Originally Posted by bluehaze

It would seem so, Evga Classified overvolts .4-.5 volts on vdimm. I wouldn't think this would be enough to kill ram with good cooling but if it does, good luck getting them to replace your ram for you, it's hard enough to get them to rma thier own mobos lately

.4 to .5 volts?

Wow that is crazy excessive.

**kiwi** · 06-26-2009, 01:25 PM

He meant 0.04-0.05

**~~bluehaze~~** · 06-26-2009, 02:12 PM

Originally Posted by Bei Fei

.4 to .5 volts?

Wow that is crazy excessive.

Sorry .04, .05 lol my bad!

**drnip** · 06-26-2009, 02:14 PM

My Classified over volts .06 v. No more and no less. A setting in BIOS/Eleet of 1.59v is really 1.65v per multimeter reading.

**Mescalamba** · 06-26-2009, 02:15 PM

Originally Posted by bluehaze

It would seem so, Evga Classified overvolts .4-.5 volts on vdimm. I wouldn't think this would be enough to kill ram with good cooling but if it does, good luck getting them to replace your ram for you, it's hard enough to get them to rma thier own mobos lately

BloodRage undervolts DIMM little and Elpida dies still.. its not mobo.. definetly. Plus, all rams should have some "safe" range .. cause all mobos have some +/- on vDIMM.

Its Elpida fault, period.

**westsider** · 06-26-2009, 02:39 PM

i used my kingstons only for 2 days
after that they degraded somehow
they could only boot with cas10

**~~bluehaze~~** · 06-26-2009, 02:46 PM

Originally Posted by Mescalamba

BloodRage undervolts DIMM little and Elpida dies still.. its not mobo.. definetly. Plus, all rams should have some "safe" range .. cause all mobos have some +/- on vDIMM.

Its Elpida fault, period.

I agree, it's the first ram I've ever had to RMA over let's say at least the last 15 years and i've already had to RMA 2 sets in one month. I am surprised manufacturers are still using this stuff, it must be costing them a fortune in RMA's.

**drnip** · 06-26-2009, 06:06 PM

I've had to RMA two sets in two weeks. Different RAM manufactures as well as different mobos were used on both.

**Booj** · 06-26-2009, 06:23 PM

1 dead stick of Kingston 2000mhz C8 here. I put them into a DFI X58 Jr and after booting into windows I saw only 4gb recognized. This was at safe clocks & volts, something like 1333 7-7-7-18

I'll stay with 2 sticks for now unless another dies. I'll RMA in a few months to give them time to sort out issues.

**dinos22** · 06-26-2009, 06:48 PM

one of my Corsair sticks died from 2000GT CAS8 kit

i did however test up to 2.1vdimm with these (i always keep vtt within 0.05v and its usually closer)

i am really babying these new 2000GT CAS7s i have. no more than 1.7vdimm and most tests now 1.66vdimm

i just want to see if it's going to happen again with lower volts

i've tested on Gigabyte X58-UD5 and X58-Extreme
will also test on Classy
good to know that it overvolts so i know where to keep the volts roughly

**Raja@ASUS** · 06-26-2009, 11:16 PM

On my 760 100% load full load voltage is actually very close to what is set in BIOS. A few loops into LInx will show the real 100% load VDimm. THe controller specification has Voffset and Vdroop, which is why idle voltage is above the full load voltage in order to ensure adequate transient response.

**saaya** · 06-27-2009, 04:41 AM

Originally Posted by Tony

If the boards are killing the ram shouldn't the board manufacturers being paying for the ram RMA and not the memory manufacturers?

Hyper is not cheap to buy, if you RMA 1 kit thru death, get another and that kit dies also does that not point at the board as the issue?

Im not saying THIS is what should happen, I am asking a serious question here as I see memory dying that has not been massively overvolted and is within the specs laid down by the dram IC manufacturer.

if you can prove a board kills mem then the board maker will be ain a world of hurt

but from what ive heard so far i doubt thats actually the case...

Originally Posted by K404

As a general comment on a related problem...are Asus willing to ackowledge that they overvolt by towards 0.1v on some boards compared to the BIOS label?

I dont think board makers will take responsbility for someone elses component dying.

thats an important thing... and i hope the dram makers and the dram vendors can push asus to stop this... its really about time they stop those +1mhz default fsb and +.1v default voltage nonsense...

and no, board makers would nevber pay for dead components killed by their boards, but if they do kill hw it means they actually failed the fcc test or should fail it, and once that happens they are in trouble... casue from then on all their designs will be tripple checked and for some time they will lose loads of orders and contracts...

plus theyd have to recall all defective boards

actuall;y defaulting vdimm to +.1v is very close to violating the specs, so asus should really watch out...

and i think the same as eva, the more recent batches of hyper seem to be veeery flaky... its an elpida problem, im sure... i mean people buy mem and its doa... hows that possible? in this high numbers? elpida hyper is either mega ultra sensitive to ESD or the bga is so fk\laky it can break and get damaged if you touch the dimms slightly too strong... both cases is elpidas problem...

Originally Posted by illuminatiASUS

This seriously a concerting issues. It really is something that makes me unsure about investing into a higher speed memory kit. Really the saving grace has been at least these memory modules are supplied through companies we know have strong support and RMA policies. I am going to see if I can get any test reports on my end and see if can provide you guys with any additional information. Additionally has anyone had the chance to try the memory linked below or know what IC's are based on? I think it is also something we at ASUS have talked about in trying to work with memory vendors and testing these modules for a longer period of time to see if issues can be found before they are experienced. This does make wonder though if something is wrong or off with the binning process currently.

http://www.ocztechnology.com/product...triple_channel

the thread title mentions what memory chips this is about :P
and it would be good if you could give asus feedback to stop overvolting vdimm instead of pointing towards a memory vendor when there are reports of at least 5 memory vendors products being affected more or less equally.

Originally Posted by bluehaze

I agree, it's the first ram I've ever had to RMA over let's say at least the last 15 years and i've already had to RMA 2 sets in one month. I am surprised manufacturers are still using this stuff, it must be costing them a fortune in RMA's.

yes... whats the point in having superfast memory but it actually only lasts for a couple of months max...

i hope elpida is paying the mem makers for the rmas, if not, id drop elpida memory right away or only sell it rated at 1.5v... while there is still memory dieing at 1.5v it should be notably less than at 1.65v

Originally Posted by raju

On my 760 100% load full load voltage is actually very close to what is set in BIOS. A few loops into LInx will show the real 100% load VDimm. THe controller specification has Voffset and Vdroop, which is why idle voltage is above the full load voltage in order to ensure adequate transient response.

no need to defend the classified

we all know its not the boards causing this... like mescalamba said, there should be a big enough security headroom for the vdimm spec to make sure that the mem doesnt die if the board overvolts slightly.

i still think .1v on asus is way too much though, thats not an innaccuracy but done on purpose... same as on some gigabyte boards, i remember a gigabyte ddr2 board that overvolted by a whooping .27v which is insane...

so it seems most of the memory dies pretty fast...
which means if you got a stick of elpida and it went through a 1 week torture session it has a good chance of surviving for a few months at least... so thats at least something good to know...

and yes, i recommend everybody who wants to buy elpida highspeed ddr3 memory to make sure the shop they buy it from rmas quickly and without trouble or the vendor/brand will replace the memory quickly and reliably.

**Raja@ASUS** · 06-27-2009, 04:55 AM

Originally Posted by saaya

no need to defend the classified

That was an explanation not a defence - don't jump to conclusions

**tripgood** · 06-27-2009, 06:23 AM

I received my rma from Corsair for my dead 2000GT 7-8-7 3x2 kit last week. I will run these @1.65 vdimm and 1.41 vtt on my 760 Classified at the rated speed and wait and see. If this kit fails, they will go back as will any future kit that fails at rated specs. As always, the heat under the responsible party's butt will get hotter and hotter as rma costs climb.

Its obvious from the responses in this thread that the Elpida Hypers have problems. I would like to see an updated database of available higher speed triple channel ddr3 kits with the chips used.

**drnip** · 06-27-2009, 07:10 AM

Originally Posted by raju

On my 760 100% load full load voltage is actually very close to what is set in BIOS. A few loops into LInx will show the real 100% load VDimm. THe controller specification has Voffset and Vdroop, which is why idle voltage is above the full load voltage in order to ensure adequate transient response.

So saying this a setting of 1.59v in BIOS/Eleet nets me 1.65v multimeter reading in Windows and Memtest. LinX prolly stresses a lil more so in reality I would be gettin closer to the 1.59v setting in BIOS/eleet? Should I just use a setting of 1.65v BIOS/Eleet even though multimeter would show a reading of 1.71v?

**saaya** · 06-27-2009, 07:16 AM

you mean youd like to see some infos on what highspeed ddr3 kits are NOT based on elpida so you can get a kit and stop rmaing every few weeks? ^^

**drnip** · 06-27-2009, 07:19 AM

I would like to see info on anything as it seems the Elpida's or at least Hyper are too damn sensitive.

**tripgood** · 06-27-2009, 07:58 AM

Originally Posted by saaya

you mean youd like to see some infos on what highspeed ddr3 kits are NOT based on elpida so you can get a kit and stop rmaing every few weeks? ^^

That would be helpful if there is such a thing.

**kiwi** · 06-27-2009, 10:31 AM

Asus P6T dlx overvolts by 0.03V only

**saaya** · 06-29-2009, 04:45 AM

at what vdimm kiwi? it makes a big diference... on my P5E3 vdimm was fine at low volts and then overvolted more and more the high i got.
and historically rog boards are overvolting, asus has been ok since they started to push enthusiast stuff to rog...

about memory... as far as i know everything thats rated at more than 1.65v is not elpida...
the best is probably to ask your favourite mem vendor what kits are NOT based on elpida... im sure they will give you an honest reply as its in their interest as well to have less elpida in the market that will most likely come back to them again... and again... and again...

**kiwi** · 06-29-2009, 06:29 AM

From 1.6 up to 2V

However, maybe for some elpidas 1.65V is an absolute max. And since some boards overvolt by say ~0.05V then setting 1.65V in bios would yield ~1.7V which might be already deadly

**Leeghoofd** · 06-30-2009, 02:16 AM

You guys ever though about heat ? These Dimms get pretty hot even at just 1.6Vdimm, plus bundle that with eg the X58 IOH which pumps out a lots too. Currently testing the Kingston 2000C8 rams, I always start at 1.6Vdimm and work my way up slowly but steadily... one thing I hope is that my OCZ blades will never ever fail on me, they are soo nice...

You can have bad luck with eg one kit, but not with 3-4 like some users here, there must be something else that's wrong , don't always blame the ram. If so you would get the BAD KARMA award of 2009 for sure...

**Raja@ASUS** · 06-30-2009, 08:46 AM

Originally Posted by Leeghoofd

You guys ever though about heat ? These Dimms get pretty hot even at just 1.6Vdimm, plus bundle that with eg the X58 IOH which pumps out a lots too. Currently testing the Kingston 2000C8 rams, I always start at 1.6Vdimm and work my way up slowly but steadily... one thing I hope is that my OCZ blades will never ever fail on me, they are soo nice...

You can have bad luck with eg one kit, but not with 3-4 like some users here, there must be something else that's wrong , don't always blame the ram. If so you would get the BAD KARMA award of 2009 for sure...

Lol, I'd never have thought of putting a fan over my memory if you'd not have said so!

C'mon man what do you take us for?

There might be 1 or 2 people without some form of cooling over the modules, but I'd hedge bets most are cooling the modules actively.

**KILLER_K** · 06-30-2009, 02:09 PM

Well seems i got it bit by it also today. Cut the pc on to check some email and look at the forum's and BAAAM! blue screen.

So i checked the ram settings to make sure, into the boot it blue screened again, i though not me too.

Again moved everything to default 1066, and booted again blue screen again. So after about 6 of them in a row, i broke out the memtest86+ 1 stick has already failed 1 minute into the test working on the second stick now.

I'm scared to even rma this set as i'm sure i will get another set of hyper's back and it will only be a matter of time before that set dies also.

Also not i haven't over volted or went past specs on this ram, it just up and died today for no reason.

Has anyone rma's then lately and what ic's did they get back, was it another version number? I mean Kingston has got to have figured this issue out by now that it is the ic's random failing.

Just wonder what they are going to do about it now......

**hipro5** · 06-30-2009, 02:16 PM

Originally Posted by Bei Fei

.4 to .5 volts?

Wow that is crazy excessive.

1.66Volts in bios, it's 1.74Volts REAL...

Thread: Elpida Hyper Graveyard

Thread Tools

Search Thread

Rate This Thread

Display

Bookmarks

Bookmarks

Posting Permissions