Stability issues w/ NF4 Ultra-D R.A02

[Liuke]

Was looking forward to this DFI board, but this is a nightmare! I'm having some major problems w/ the Winchester 3200+ and DFI Lanparty NF4 Ultra-D (R.A02) upgrades I just received. I've never had so many issues w/ doing a PC upgrade, and I'm all set to RMA something. I just don't know what the prob is or if there are any work arounds so any input is appreciated.

 

The system starts up fine, but it is massively unstable. Most of the time, it can't even complete POST before it reboots on its own. It often freezes up or reboots while I'm in BIOS! Installing Windows is impossible. I can't even get far enough to copy files from the CD to the HDD. Most of the time I get BSOD's running the setup CD.

 

I suspected a memory-related issue and this seems to be the case. Memtest86 gets massive memory errors. I've tried everything possible w/ memory tweaking and timings w/ no success. I've relaxed timings as high as I can go. I've set Command Rate to 2T. I tried increasing voltage, but apparently the DFI board is being stupid and doesn't let you change it. It's stuck at 2.60V right now. The only way I have gotten Memtest86 to successfully complete a pass of tests is to underclock my 3200+ to 9x200 w/ an HT multiplier of 4X (basically running as a 3000+). However, nothing I do seems to have any effect running at "stock". This is simply unacceptable in my book.

 

Oh, and I've tried it w/ only one DIMM installed. This doesn't improve the situation. In fact, I can only install the DIMM in slot 2 when running one and only in slot 2 and 4 w/ two. The system won't even start w/ DIMM's in slots 1 and 3. I doubt it's a memory stick prob because I used these same Crucial Ballistix PC3200 sticks in my P4 at a voltage of 2.5 w/ 2-2-2-5 timings. Easily more strenuous than the 2.5-4-4-8 and higher settings I have tried on this new setup.

 

At this point, I am thinking this is either a problem w/ the CPU or the mobo. It could be a CPU issue since the memory controller is built-in. It might also just be the DFI board playing nasty w/ my RAM. Anyone got any thoughts as to the issue? I'm ready to RMA one of these parts or possibly both.

 

Here are specs of what I'm trying to get running:

Winchester 3200+

DFI Lanparty NF4 Ultra-D

2x512MB Crucial Ballistix PC3200

BFG GeForce 6800 GT OC PCI-E

74GB Raptor, 160GB WD SE

BFG 550W PSU

[Angry_Games]

please check the rules and the Important Info threads stuck to the top of this section, paying close attention to the second post in teh 'critical' thread (about system specs in your sig).

 

this helps us tremendously to help you solve your issue if we can see what is in your system every time you post.

 

Be as thorough as possible please. Voltages, timings, settings, bios version. The more information you give us about the problem, and the more details about your hardware in your sig = more help we can give you.

[ISS]

You are the only one experiencing stability issues with the board. Out of 4 others I know who bought the board, 3 has the same problems as yours. RMA the board.

[pantonious]

if its possible, try another stick of RAM in the board...prefferably a different brand...i've had a few DFI boards that were VERY picky with RAM...i mean, specific stick of the same brand...(865PE and 925TX)

 

to get it to run stable to you HAVE to lower the HTT multi down to 4x??

have you tried it at 5x with the CPU underclocked??? and have you tried the CPU at stock and the HTT multi at 4x???

[cpuz]

You are the only one experiencing stability issues with the board. Out of 4 others I know who bought the board, 3 has the same problems as yours. RMA the board.

Your post makes no sense. Please read it again. Should it say "you are NOT the only one..."?

[Liuke]

Went ahead and added the specs to my signature though I think I got everything of relevance in my first post.

 

The one thing I would add is that the DFI board reports CPU voltage at 1.38V when set to "Auto". I went ahead and manually set it to 1.50V to see if that would help, but it doesn't.

 

I wanted to try some other sticks of RAM, but I didn't have any others at hand. I may try it again w/ some older 2x256MB Kingston HyperX PC3500 sticks I have, but I will have to run to another house to grab those. Once again, I know the Crucial sticks themselves are fine as they have run flawlessly on my P4 system for months overclocked. I hope the DFI does not have compatibility issues w/ them.

 

Setting my system to "stock" at 10x200 w/ an HT of 5X causes major probs w/ every combination of memory timings I can think of. Even if I clock the RAM down to PC2700 speeds. The system seems to pass Memtest86 fine if I set it to 9x200 w/ an HT of 4 or 5X and really conservative timings of 2.5-4-4-8 w/ a Command Rate of 2T. The system won't even start w/ a CPU multiplier of 9.5X so it looks like that is "broken".

 

It's hard to say much more as the only thing I can get the system to run right now is Memtest86 (sometimes it crashes or the screen corrupts). I've pretty much spent 20 hrs dilly-dallying on the BIOS and POST screens. I'm all set to RMA this thing, but I don't want to give up yet if this can be fixed here and now. If you guys have any other advice please let me know. Thanks!

 

I'd be happy right now if I could even get Windows on the PC running at "stock".

[ISS]

Your post makes no sense. Please read it again. Should it say "you are NOT the only one..."?

 

Yup. Thanks for the correction. I meant to say you aren't the only one...

[PGBK]

I have the same problem for bios 25/01/2005 for vdimm can't change but u can work around by using page up/down and save it w F10 at the vdimm area. The vdimm would stay.

 

for the bios 25/01/2005, I need to pump my vdimm up to 3v even for TCCD. The original vdimm display must be out since my tccd is running cool even w 3v.

 

it is solve after flashing to 0201 bios

[pantonious]

do you have any friends with a Socket-939 computer???

 

the only way to know exactly what is at fault is to either try a different CPU in that motherboard or to try your CPU in a different motherboard...

[PGBK]

Its the memory issue I went through exactly the same in the first place. The vdimm shown is way out so I test it out in memtest. It can pass all but #8 if voltage is not enough.

 

This problem is solve on 0201 bios though......but i still can't oc high w it yet...

[Liuke]

Gosh, reading through some of these threads is starting to scare me about issues w/ this Ultra-D board. Unfortunately, I'm like the "new kid on the block" and no one has a S939 system I can test with.

 

I wonder if it's worth trying that beta BIOS to see if it fixes this issue? Not even sure if applying beta BIOS'es would void the warranty. I tried using Page Up/Down to change the mem voltage, and it still doesn't stick even after doing that. The number changes right back to 2.60V as soon as I highlight another field.

 

I should probably also mention that I have been pulling out cards and drives for the better part of the day one by one until I was down to the bare minimum necessary for the PC to boot and and still see the display. I got the same result. This seems like a good enough test to rule out the PSU as the problem in my book.

 

The only thing I can't really distinguish are the CPU and mobo. Having the memory controller integrated on A64's seems to make it harder to tell apart. The symptoms and other issues I saw here are making me lean toward the mobo: freezes and reboots while in BIOS, reboots at random points during POST, and even screen corruption a few times I was trying to get into BIOS or POST.

 

Once again, any pointers in any direction are much appreciated. In the meantime, I'm taking a break from this. There are some other things I need to do and this has consumed about two days of being in nothing but POST and BIOS! :eek:

[PGBK]

Gosh, reading through some of these threads is starting to scare me about issues w/ this Ultra-D board. Unfortunately, I'm like the "new kid on the block" and no one has a S939 system I can test with.

 

I wonder if it's worth trying that beta BIOS to see if it fixes this issue? Not even sure if applying beta BIOS'es would void the warranty. I tried using Page Up/Down to change the mem voltage, and it still doesn't stick even after doing that. The number changes right back to 2.60V as soon as I highlight another field.

 

I should probably also mention that I have been pulling out cards and drives for the better part of the day one by one until I was down to the bare minimum necessary for the PC to boot and and still see the display. I got the same result. This seems like a good enough test to rule out the PSU as the problem in my book.

 

The only thing I can't really distinguish are the CPU and mobo. Having the memory controller integrated on A64's seems to make it harder to tell apart. The symptoms and other issues I saw here are making me lean toward the mobo: freezes and reboots while in BIOS, reboots at random points during POST, and even screen corruption a few times I was trying to get into BIOS or POST.

 

Once again, any pointers in any direction are much appreciated. In the meantime, I'm taking a break from this. There are some other things I need to do and this has consumed about two days of being in nothing but POST and BIOS! :eek:

 

my mistake, don't go other field using this method. after selection just F10 and quit. It would save permanently there.

 

the vdimm should be purely a scare only....don't quote me on this....I pump 3.0v just to get memtest pass without errors and there is no heat on my tccd so i say is just a scare....