Printable View
Hi Andy.
I know it is not recommended (if it even works on this board), but dont you want to try the OCZ booster on the board :) I don't want to break mine :D Then we can know if it is a 5v jumper problem or not. I know it's 20 pin device, well, your call....
I also have the cold boot problem, waiting to see if it gets resolved.
Hehe
I'd also like to know about this. Also thanks alot Andy for the help regarding this issue. :DQuote:
Originally Posted by OzSnoal
You are seriously tempting me........;)Quote:
Originally Posted by ADM
Nope, look up a bit........Quote:
Originally Posted by arctic-k20
I have the 3 leds issue with 2x256 XMS3500(BH-5) AND my Gskill LE sticks.
It is NOT my ram, or my cpu.
Let's do as AndyOCZ suggested, we will all benefit... there has to be something common in all of this. Unless it's just VX and DFI NF4 series.
AndyOCZ's wise words were
Yeah it's all in my signature, but if we list it, it's easier to read and compare.Quote:
If I were looking to solve this I would be asking about boards revisions, CPU steppings and week codes and the like to find some that iyou all have in common, not your memory, as we can see it's not a certain type of memory or a certain brand.
What do you think?
How about posting your DFI mobo type (UT unmodded, UT modded to SLI, SLI) revision (located on the sticker on the bottom side of the last PCI slot, like AA0 or A02), PSU model and size, and whatever els you can come up with to find something in common with all of you.
Please keep the format as below, if you think we should add something just say so.
MBoard: DFI NF4 Ultra
MBoard Rev.: R.AA0
CPU: Winchester 3500+ @1.4V
CPU REV.: CBBHD VPFW (You can only find this on the CPU, before heatsink mount)
FSB: 250
CPU Mul. x10
MEM: 2x512, OCZ VX PC4000 FSB Ratio 1:1, 2-2-2-5
VDimm: 3.5V
SLI Mod: No
PSU: Antec PP-412XF 20 pin 400W : 12V(18A), 5V(40A), 3,3V(28A)
PCIe: Leadtek 6600GT Extreme 128MB (Stock speeds)
Other: Audigy2, Maxtor EIDE 6Y120P0, Toshiba 1502 DVD, BenQ DVD-RW DW1620, 1.44 Stiffy
i heard that many people have problems booting if their vdimm is set too high, if you have a booster lowering vdimm and booting with it and then raising it in windows helps from what ive heard...
what vdimm are you guys running?
well my board is A00 - and have not had the issue with any of my tccd (3 sets) but have had the problem with all of my vx sets...
i did not have the problem using the 5v line for the tccd either although i am using the 3.3 line exclusively now...
Doesn't matter whether I am running 3.7 or 2.7 ( I use both).Quote:
Originally Posted by saaya
Even with my BH-5 at stock speeds and 2.8v it gives me grief.
No matter what volts I try and boot at.
List:
2x256 XMS3500/BH-5 , good sticks- no issues with them, USUALLY run 3.7v off 5v rail
EDIT: OR
2x256 Gskill LE "431" chips, decent clocking, np sticks. usually 2.7-2.8v
Powerstream 420, readings are good with MM, during/after boot and when I finally get it running.
1-DVD drive
1 stock HSF, off header/ tried direct to psu- no difference
1 X800l- stock
1 HDD/ PATA
Winny 3000+, 0441/SPMW, works fine, but will not boot correctly( without playing games with cmos or swapping/reseating.... sometimes for a good while) regardless of what voltage I try to boot with.
No extra fans or lights, extra cooling(when needed at high vdimm) is provided by a floor standing fan ( laying down of course).
Board is on the foam pad on top of a phonebook, which makes a very stable platform.
All 4 power connectors hooked up, tried less, no difference.
3 leds everytime I boot now, insert works sometimes, but requires a reboot(sometimes on it's own).
That pretty much covers it.
So far:
- Occurs with either EIDE or SATA drives
- Is not hard drive specific, seen Samsung, Maxtor
- Occurs with Nvidia and ATI graphics cards
- Occurs on DFI NF4 Revision A00, A02, and AA0 (EDIT: Reworded, added A02)
- Is not related to using the 5V line for Vdimm. (At least for memory other than VX.)
- Is not PSU related, occurs with 20 pin supplies and 600W 24pin 'approved' supplies.
Questions
- Has anyone got one of those 'Startup monitoring/debugging boards'? An error code?
- The presence of 5VSB, when the PC is off, but not unplugged from the wall socket.
The presence of 5VSB allows for VX boot up, yet the lack thereof prevents VX bootup. Why? If we answer this, the problem would be solved or at least solvable.
(NOTE: 5VSB/ 5V StandBy, is used to power devices when the PC is turned off, like network card Wake up on Lan, Power on by keyboard etc.)
Troubleshooting for the powers that be
So far it all points to the VX/DFI NF4 combination. What I wonder is, if DFI and OCZ are working together to investigate why this is occuring. I think the problem lies in both company's hands;
1) DFI, because it requires a bios update to make the DFI boards more VX friendly
2) OCZ, because the VX ram appears to 'settle in' to a state where cold boot becomes a problem.
OR, perhaps it is not a good enough quality control/testing on the OCZ VX series.
Friendly Rant
Don't take this in the wrong way, I like that DFI is continously bringing out new beta bioses, but I do find it questionable the way DFI has handled their bioses. Why RAM has to be in certain slots, then later 'tweaks' are applied, then you can use the other RAM slots. Then there is the guessing game of which slot to put the RAM in. It's all too much.
A board should not behave in this way - maybe it's a DFI feature to compliment NF4 series DFI's motto:
'UV on, Tweak fun!!!'. ;) Please stop, it's not fun anymore.
Sometimes if I reset the CMOS and change the RAM voltage to 3.2v I can get both sticks to boot. Yet this has only been a rare thing - and usually always spontaneous. :(
The DFI board has more revisions than what you state. I have an A02 and an AA0. I hear there is an AA2 now also. What was changed and why?Quote:
Originally Posted by OzSnoal
If others with BH5 form our competitors are having this issue, how can you say our quality control is the possible problem? This is like accusing the roads for causing Firstone tires to explode on Ford's a few years ago.
What if your CPU is developing a problem? Have you swapped in another? What if it's the board? Have you tried another? Why have I not had the same issue with the 10 kits of VX that I have had here in my lab over the last 3 months? Some of them ran high volts for weeks of hard testing.
We want to help, but need your help. No hard feelings, but lets try and figure out what the problem actually is.
If someone in the USA wants to send me their VX from a system with this cold boot problem, I can test on an A02 and AA0 DFI NF4 boards, 2 Winnies, 1 Clawhammer and 1 Venice here in San Diego, CA. I will let you know what I find.
Just had another thought. If someone wants to send me their CPU, RAM and mobo I could replace each component until it boots properly.
Thanks
awesome Customer Service andy:)
I had same problem with DFI NF3, so its not NF4 only.
Its not VX only either.
Its the chip on ram i think and seems to be only UTT that has it.
Has anyone tried to cool the ram or warm them up before trying to boot ?
Ok, will reword and update it, I found a user Paudi, that uses A02 and has the problem (http://www.bleedinedge.com/forum/sho...?t=9573&page=5)Quote:
The DFI board has more revisions than what you state. I have an A02 and an AA0. I hear there is an AA2 now also. What was changed and why?
The word was perhaps, and I was just throwing things in the air here. Anythings possible, right?, until we know better. You're in a better position to know what goes on. Like you and all of us here, would like to know what it is, no harm intended.Quote:
how can you say our quality control is the possible problem?
How do you explain Dsunamik case? Who RMAd his ram, got a replacement, worked great for 2 days and the problem slowly crept back to to eventually end up back where he started.
With the new sticks, Dsunamik did the following:
1) i switched the sticks around and now it is cured...this is an awfully familliar pattern.
2) can shut down, turn off do whatever i want...
3) (2 days later) Reboot problem returned. Could not post for about 3 hours. Left it overnight and it magically started up this morning no problem.
I want to help out too and get to the bottom of this.
no problem with Hynix BT-D43 RAM.
i dont think it is cpu related - it might be but i have a 3500 newcaslte a 3000 winny and a 3200 winny and all did it with 3 different sets of vx(utt). the only common thread in my experience was the board - the board has been the same all the way through.........but tccd has never failed - ever - and i have 3 sets of tccd to try.
uberman: that is interesting that you had issue with nf3 as well.......definately makes one believe that is limited to ram only (not just ocz but the actual chips themselves)
How do you find the boards revision number? Is it located somewhere on the board? :rolleyes:
last pci slot there is a sticker on the underside
Okay my boards revision is A02.
EDIT: Added 4x4N Board revision
Summary of CPUs used and NF4 revisions:
(Need to add something, go ahead)
arctic-k20:
Winchester 3200+ CBBFD 0441
Newcastle 3500+
Winchester 3000
Rev. A00 DFI NF4 SLI-DR
ADM:
AMD San Deigo 3700+
Winchester 3200+ (Added)
Rev. A02 NF4 DFI Ultra-D
Paudi (http://www.bleedinedge.com/forum/sho...?t=9573&page=5)
Winchester 3200+ @2.4Ghz
Rev. A02 DFI NF4 Ultra-D
4X4N (NOTE: problem with TWINMOSS TMSP400)
3200+ CBBID
Rev. A02 NF4 DFI Ultra-D
dsumanik:
Winchester 3000+ @ 2322 mhz
Rev. AA0 DFI NF4 SLI-DR
lowdog:
Venice 3500
Rev. AA0 DFI NF4-SLI
(Geil UltraX 4400 x 2 512meg = coldboot probs)
(Geil ULD pc3500 x 2512meg = coldboot probs)
(Cosair Twinx1024 3200LL BH5 = coldboot probs)
OzSnoal
Winchester 3500+ CBBHD 0450 VPFW
Rev. AA0 DFI NF4 Ultra-D
chiyau (http://www.bleedinedge.com/forum/showthread.php?t=10069)
3700+ ??Details??
??Rev.?? NF4 DFI Ultra-D
maxoverclock
4000+ @ 2977mhz 1.65V
??Rev.?? DFI NF4 SLI-DR
turtle
Winchester 3500+ @ 2600 (2720 max)
??Rev.?? DFI NF4 Ultra-D
I get the exact same problem with my Winchester 3200+ aswell.
Are all bad rams on brainpower ?
http://www.insanetek.com/index.php?p...goldpc4000vx_2
Quote from the article titled 'OCZ Gold PC4000 VX 1GB DC Kit':
Quote:
I also noticed that OCZ used the B6U808 Brainpower PCB
I've been trying to find other people with the same problem on google, if I find them I'll link them to this thread. Hopefully this problem will be fixed soon. :) Also I tried my RAM in another mobo today (Socket A) and it boots with the bad stick. Yet once in BIOS it reports that stick to have 256MB of RAM instead of 512. Yet the good stick gets reported correctly at 512MB.. strange.
Moste defanitly cold boot issues. What a pain in the arse.
This is my 3rd set of memory. I had GEIL PC4400, same issues. I was hopeing this OCZ stuff would be the cure. But NOT!