RedLine Death

[cowpuppy]

Runnining my redLine at 3.3v with my OCZ booster/K8N Neo2 cooled with a desk top fan. They were working fine, I powered down to go to work and apon returning home my PC wouldn't boot. Booth sticks bad.

[ben805]

test one stick at a time with memtest and see if you can find the bad one.

[cowpuppy]

Both sticks bad. PC wont boot at all. Slap some good old Bh5 in and wiz bang pc boots. I'll be getting ahold of Mushkin for RMA # monday.
just thought I would let some others know as most of the deaths seem to be coming from the DFI boards.

[jiff]

Thanks for that As I have said many times, I cant see it being a board issue, more ppl using NONE DFI boards with high volts would be good.
I suspect the Ram company are going to get stung on this. Life time warranty

[ben805]

if we see more and more issues like this where the UTT/CH chips keep getting fried on different mobo within the recommended vdimm, then we can guaranteed UTT chips is to blame.

[Dumo]

Make sure stickers and spreaders intact Not even rma testing will be done if those things peeled/took off.

[TuKo]

Just figured out tonight that my DFI has killed too one stick from my RedLine kit

[charlie]

I'm sure MUSHKIN will treat you guys right....

C

[Duonger]

no worries. we will take care of it. Just call us monday.

Duonger

[poiuy223]

its not mushkin's fault, it's just proof why these winbond UTT's suck

[HiJon89]

its not mushkin's fault, it's just proof why these winbond UTT's suck

I guess these chips just aren't capable of running high voltages for extended periods of time the way BH-5 were.

[sparkie34]

I guess these chips just aren't capable of running high voltages for extended periods of time the way BH-5 were.

its not mushkin's fault, it's just proof why these winbond UTT's suck

How about the booster?

[HiJon89]

How about the booster?

I dunno, a lot of people have run the booster for a long time without problems, but an overwhelming percentage of people with UTT based products have had sticks die on them.

[Disposibleteen]

i wonder if using a booster to get "cleaner" power to the ram would extend the life of these sticks, and then on the other hand maybe this is just a fluke, maybe there are only a few "bad" kits out there and this isnt typical.

[Duonger]

yes i am curious as to which is the cause also. we have had our pair runing 24x7 for 3 months now at 265 without a hickup. we have been comparing most of our testing to this particular set. i will talk to DFI monday and get some more boards in house as well as some ddr boosters to test and see. i will also have other utt chips tested too.

Duonger

[Reefa_Madness]

How about the booster?

Booster might explain the Neo2 situation, but not the multiple DFI occurrences. Its also happened to at least a couple of people that I know using TwinMOS Speed Premiums and DFI NF4 boards.

If you think about this logically, what is the one common component that is UnTesTed in all of this? As the old, but time-proven saying goes...a chain is only as strong as its weakest link. Its too early yet to lay the blame, but its not too early to have suspects.


edit:

Duonger,

In the case of the thread starter, as well as one of the cases that I am personally aware of the common thread was that in both cases the boards were shut down, then later restarted. You've just stated that you guys are running one 24/7 without issues.

Is there any chance that the problem could be caused by the sudden surge of high voltage when the boards are restarted?

Maybe as part of your testing you might want to consider having one rig that gets shut down and restarted, with timed intervals of run and down time simulating a non 24/7 useage pattern.

[Disposibleteen]

yes i am curious as to which is the cause also. we have had our pair runing 24x7 for 3 months now at 265 without a hickup. we have been comparing most of our testing to this particular set. i will talk to DFI monday and get some more boards in house as well as some ddr boosters to test and see. i will also have other utt chips tested too.

Duonger

sounds good to me, jeez you guys at mushkin dont mess around.

[jiff]

I had VX die on my DFI 3 I also know of 2 others that had them die on a NONE DFI 4 boards.
There is no way its voltage spike taking out RAM, if you remember when ppl where have Memory controller deaths on the shuttle boards, this was due to voltage spikes.
There was no/little cases of RAM deaths just on die controllers going down. So ppl are suggesting that the Spikes are large enough to take out the RAM but not the Memory controller

Its pretty obvious that the vast majority of deaths will be on DFI4 baords as its the only board that offers such volts out the box, it is also by far and away the board of choice for extreme users.

[Disposibleteen]

maybe DFI should examine their voltage delivery systems, would a line filtering UPS help?

[bachus_anonym]

Its pretty obvious that the vast majority of deaths will be on DFI4 baords as its the only board that offers such volts out the box, it is also by far and away the board of choice for extreme users.

That is XTREMELY good point
Everybody thinks that DFI kills mems. But few seem to stand aside, look at the situation and realize that 95% of folks we hear about here with UTT are on DFI NF4. So no wonder, they jump to conclusion that DFI kills as nobody else runs anything else that DFI NF4. It's great that we have more feedback from different boards (my 2 sets of UTT died on Neo2, too). More and more information like that will open other's eyes a bit wider

[poiuy223]

while these UTT memory do die in DFI boards, i highly doubt its a DFI problem. it's not a memory maker problem either. it's the IC's. there's a reason why they're so cheap. it's untested so actual quality is unknown. if they do run for a long period of time, be happy. if they die, dont be surprised or angry.

overclockers love it because its cheap and does what we want. but if you want stability and reliability, UTT is at the bottom of the list. i personally stick to micron or samsung based memory. they overclock and they actually live

[deception``]

These UTT deaths are extremely perplexing. I suppose that it could be due to a number of factors, but it is in my honest oppinion that these untested Winbond dies simply are not up to par with the older Ch-5/Bh-5 sticks. Perhaps I could be way off, but I certainly think that there was a very good reason that low latency was hard to find for some period.

deception``

[Duonger]

Duonger,

In the case of the thread starter, as well as one of the cases that I am personally aware of the common thread was that in both cases the boards were shut down, then later restarted. You've just stated that you guys are running one 24/7 without issues.

Is there any chance that the problem could be caused by the sudden surge of high voltage when the boards are restarted?

Maybe as part of your testing you might want to consider having one rig that gets shut down and restarted, with timed intervals of run and down time simulating a non 24/7 useage pattern.

Thanks Reefa. I will have this simulated also and see what happens.

Duonger

[hovo73]

You guys crack me up, there has been plenty of people with "old" Winbond and even TCCD going bad on DFI board, yet everytime this issue comes up there is always someone claiming that this only effects UTT. Go check 2 threads on this issue in AMD sections if you care.

PS. Does anyone know how which rail does Abit AN8 Fatality SLI board is using to power DIMMs?

[felinusz]

I'm also of the strong opinion that the UTT memory IC is to blame here.

As bachus has said, the UTT memory ICs do not undergo testing for any sort of integrity/performance standard after fabrication. BH-5 did go through such testing immediately after fabrication, which would explain why it seems to be 'hardier'. The "bad" BH-5 ICs were tossed into the trash, and never hit the market.

Basically, the company fabricating UTT is completely defeating the yield problem, by removing any standard of integrity or operation from their product. By doing this, every single UTT memory IC makes its way into the market, yields be damned. In times past, yields could cripple a company's profits, when a percentage of the part being fabricated was discarded, and not sold.

A similar trend has occured with Samsung, and their TCC5 part. Instead of testing all of their TCC5 memory ICs to a DD500 standard of operation and integrity, Samsung now only tests them to a DDR466 standard. TCC5 is identical to TCCd, except for the different standard of operation/integrity after fabrication. Thus, with TCC5, more of the memory ICs pass testing, and make it to the market than with TCCD, due to laxer standards.

This recent trend in the memory industry of removing or compromising standards of integrity/operation in order to sell a higher percentage of the memory ICs is somewhat troubling.


While the UTT that we buy in 'enthusiast' RAM sticks has been speed binned to operate at high speeds, it hasn't neccessarily been tested for long-term integrity at high voltages, as BH-5 likely was.


jiff, I agree, the DFI nForce4 board is likely not the issue here - almost everyone who is running UTT in a high voltage environment in the first place, uses a DFI nForce4 motherboard, this heavily skews our sample-base of UTT memory death.

I made a thread on this topic on OCF, some of you mught find it interesting: http://www.ocforums.com/showthread.php?t=389451