DFI Admits 5Volt Vdimm Problem

[G H Z]

With the weirdness happening I wonder if the jumper change itself might not be the risk that takes out Ram if not done right. I know when I changed slots without changing jumper, the detection time was much longer. I wonder what would have happened if I had changed the jumper at the same time.

[situman]

Hey, NP, I totally understand. You're just lucky you got a board that worked for as long as it did. Me? I'm on my 6th board

u had 5 boards die on you and u didnt mention a word about it?

[s7e9h3n]

u had 5 boards die on you and u didnt mention a word about it?

To whom? If you mean in this thread, none have died from using the 5v Vdimm since I run tccd......

[dippyskoodlez]

u had 5 boards die on you and u didnt mention a word about it?

and I doubt its under normal 24/7 use... i.e. LN2.. dry ice... cascade...

[s7e9h3n]

and I doubt its under normal 24/7 use... i.e. LN2.. dry ice... cascade...

Well, it was, sorta....2 came doa, 1 I found a bulging cap, another killed my fx55 (Either that or I did, but exchanged the board anyways because I hated it for killing my favorite cpu ), and my last one just plain and simply just crapped out on me - would go through the entire boot sequence, just nothing was showing up on my monitor . This one seems to be running OK, but it's definately not as good a OC'r as my prior one...

[situman]

well there were plenty of "my dfi has problems..." plus many others, so it would be helpful in helping to assess failures

[s7e9h3n]

well there were plenty of "my dfi has problems..." plus many others, so it would be helpful in helping to assess failures

So I don't make that mistake again , could you please tell me how/what I should have posted about my boards that could have helped other members here?
Maybe started a thread stating I got a DOA board? Or maybe one titled "LOOK EVERYONE! I GOT A DFI WITH A BULGING CAPACITOR!", or maybe one like "I DON"T KNOW WHY NOTHING'S SHOWING ON MY SCREEN" Trust me, I've done my part here trying to help others and if you haven't noticed, I've been a little busy trying to share info about a few cpus which could be considered a *bit* more interesting than my dead dfi's.

[topboy]

im with you s7e9h3n

Let them find out for themselves

Like you .... i know what NOT to do....and i WILL NOT be running the 5V on a DFI thats if i run a DFI again

Good luck all

Im off to play with my new ABIT and X2 4800

[wtz54321]

im with you s7e9h3n

Let them find out for themselves

Like you .... i know what NOT to do....and i WILL NOT be running the 5V on a DFI thats if i run a DFI again

Good luck all

Im off to play with my new ABIT and X2 4800

You mean +5V jumper will kill the NF4 chipset in this board once you ever use it ? Where did you get this information ? I know you post some information about what you hear from DFI SJ that using +5V jumper may kill some ic on the board , but are you sure he means it's the NF4 chipset ? I have read many threads about the memory death on this board ... But you seems to be the only guys that kill your board even after you switch back to 3.3V ??? And it seems you are very angry about the board die even you switch back to 3.3V , are you sure the NF4 chipset is really killed by the 5V jumper once you ever use it ?

[hovo73]

Zero problems here, just to add my personal experience, I've been running on the 5v jump since my ultra-d was first installed, I moved to 5v before ever powering up the board for the first time so that I wouldn't have to screw with changing later on.

BTW, according to DFI you can't or you shouldn't do that. You have to boot with 3.3V line and change CPU VID from manual setting to AUTO in order to be able to use 3.2+ VDIMM. At least according to their manual.

I also, I also think the board is to blame, but topboy, what is with all the suspense?

[HARDCORECLOCKER]

yep and three are better than two

ill get a pic if i get a chance....nice job with the stock dfi chipset cooler too lol

THX mate - I hate it to have unused fans laying around........

@Juliette:

Was absent caused by holidays and then had a lot of work............, but wait till the FX-57 is released - think I will post then till my fingers are bleeding

[tictac]

Official statement from Oskar Wu (DFI Engineer)

Sorry , but I did not know anything like what you said ...

If you use 5V jumper to select the voltage higher than the DRAM or CPU can handle , either your CPU or DRAM will be damaged ...

If the high voltage cause the current over the MOSFET limit , the DRAM regulator may be damaged ...

But that number will be 53A continous when the MOSFET working under 100 degree Celcius , the peak current this MOSFET can handle is above 200A ...

The MOSFET we use in DRAM power regulator is PHILIPS PHD78NQ03LT ...

http://www.semiconductors.philips.co...78NQ03LT_4.pdf

Anyway , all these are expected to happened when the board released ...
We have put enough warning about all these thing ...

Please be understand that the NF4 chipset is not power by DRAM voltage ...

Also DFI or any other company will not gurantee or be responsible to anything damaged when you set voltage beyond the chip spec(cpu , dram , chipset , pci-e graphics or anything else ) ...

I don't know what you hear from DFI SJ , but I never say anything like this to anyone in DFI SJ ...

Source :
http://www.dfi-street.com/forum/show...&postcount=111

[stealth17]

Official statement from Oskar Wu (DFI Engineer)



Source :
http://www.dfi-street.com/forum/show...&postcount=111

hmmm....so whats that mean? there is are isnt issues?

[highoctane]

BTW, according to DFI you can't or you shouldn't do that. You have to boot with 3.3V line and change CPU VID from manual setting to AUTO in order to be able to use 3.2+ VDIMM. At least according to their manual.

I also, I also think the board is to blame, but topboy, what is with all the suspense?

Yeah Yeah I know I've read about all these things now, I didn't make it my priority to read through every detail when I built the system, not only that theres countless posts about every detail at this point.

Coulda shoulda woulda, its all too late to say that about the situation now when this was done months ago and the fact remains it works fine regardless of the sequence used to do it. The first thing I did with this board was move the jumper to put the memory on the 5v rail, it worked fine and my board has been running perfect since, knock on wood.

I'm guessing the reason they say run the vdimm up is so that the power mosfet doesn't have to step the voltage down so far possibly causing damage or undue stress under certain conditions.

Anyways I probably did a bunch a things that weren't by the book but I have no complaints about the outcome, I actually have more fears of changing something that might trigger a problem.

[highoctane]

Official statement from Oskar Wu (DFI Engineer)



Source :
http://www.dfi-street.com/forum/show...&postcount=111

Well that definitely helps clears some of the smoke/speculation surrounding the issue, at the very least we know the chipset should not be affected by using the 5v rail for vdimm.

[stealth17]

has anyone ever thought about just doing a vdimm volt mod? i know you are thinking "who needs more than 4v?" but thats not the purpose. i was thinking that a vdimm mod would possible reduce the risk of spiking as the pot is what controlling the voltage rather than a 3cent IC, and it might also fix or reduce the cold boot issue as the voltage it applied manually, so as soon as you turn the board on the dimms have the voltage....

just a thought....