DFI NF4 the UTT/VX/BH5 KILLER READ IF YOU HAVE DFI

[SET]

Single as a Information.

I also used the rail of 5V and bios 414-3 when the ram died. Also I control my tccd now and I look for a solution to replace my UTT.

This has happened in some SLI or all the problems have been with Ultra-D?

Greetings

[FiNcH64]

well, with all the knowledge that i have, and some other guys that are specialist in hardware, that includes ram memory.. i can pressume that if you want a high overclock for 24/7 use, you can't never reach your mems to more than 3.0 volts, cause even at 2.9v the ram its hitting almost the 50 centrigrades wall, that for what i know its like the temperature risk start that can cause fails and deaths on your sistem components, you named, cpu, chipset,ram.. etc

so getting a samsung tccd ram chipset..can give you an edge of a high overclock with low voltages, so i guesss that these ones are the real kings of overclock , cause al winbond chipset are made for overclock benchmarks and that's all..

so i agree that DFI is free of any responsability of memory deaths.. just because their line of nforce 4 mobos have an option of even 4.0 volts on ram,
its all on you and you are moving the rail at your risk.. thats why DFI block the voltage increase on 3.2 v ... to avoid this things guys

don't you think so??

[hovo73]

so i agree that DFI is free of any responsability of memory deaths.. just because their line of nforce 4 mobos have an option of even 4.0 volts on ram,
its all on you and you are moving the rail at your risk.. thats why DFI block the voltage increase on 3.2 v ... to avoid this things guys

don't you think so??

I disagree. It is not a hack or a mod, it is a feature of the board and is advertised as such and I am sure a lot of people considered that when they bought the board. As such, if it doesn't work correctly, if for example the voltage fluctuates, or spikes to 5V (I would like to stress that we don't know that for sure yet, but if it does) I is a malfunction and I think DFI is liable.

My $0.02

[ADM]

I disagree. It is not a hack or a mod, it is a feature of the board and is advertised as such and I am sure a lot of people considered that when they bought the board. As such, if it doesn't work correctly, if for example the voltage fluctuates, or spikes to 5V (I would like to stress that we don't know that for sure yet, but if it does) I is a malfunction and I think DFI is liable.

My $0.02

Yeah I agree with this, although DFI gave minor warnings about using it, at the end of the day it's one of the boards selling points.

[Eversor]

any short/long term side effects of having the 3.3v line set to 3.6-3.7v on the psu?

My PSU doesn't support adjustment to the rails, it's a 520w bequiet.
But running for 3 months on the 3.3v rail and had no problems. I've put a switch to turn the mod off, when i need to test with the 5v rail.
Don't think it will be a problem.
Maybe i'll make a mod to mine, cus my ram really likes volts, now its 255 with 3.3v, but with 3.5 it gives me 270.

Has anyone tried moving ram from slot?
I don't mean all are good, but some might be.

No the ram will not die by using more than 3v. Alot of my ram had already died if it was so. I've been using mine at 3.3v for more than a year in different mobos.

EDIT:

In my testing it seems that using 3.3v rail directly, the vtt isn't changed accordingly, cus i need to set volts to 3.2v to have full stability.
Using 3.3v from rail tough, allows me higher clocks than 3.2v. So regulating the 3.3v rail may not be the better option.

[[XC] 4X4N]

I disagree. It is not a hack or a mod, it is a feature of the board and is advertised as such and I am sure a lot of people considered that when they bought the board. As such, if it doesn't work correctly, if for example the voltage fluctuates, or spikes to 5V (I would like to stress that we don't know that for sure yet, but if it does) I is a malfunction and I think DFI is liable.

My $0.02

Yes, I agree with this also. Having the option to run up to 4v vdmm was the #1 reason I bought this board, and if it turns out this doesn't work properly DFI needs to be held accountable.

[xgman]

Keep in mind that this is just the sort of reason that these companies use to NOT include these extra OC features and we better be carefull what we ask for because we may end up back at square one again.

[jinu117]

FiNcH64...It's very interesting to see some members whose not fully well informed barging in and making this interesting "from expert" comments.
Fallacy of your logic.

1) OCZ VX and Mushkin Redline requires 3.2v. 10% margine is allowed per design as usual than it leaves us at tops 3.5v right away. This problem is not just limited to any UTT but the ones that are supposed to be used at 3.2v+, etc.

2) Regarding TCCD, have you actually had TCCD? Anyone having TCCD with 2.8-2.9v with over 290 2.5-3-3-7 without additional cooling is REALLY risking it. The darned thing is hotter than VX or UTT @ 3.2v 2-2-2-5 230mz.

3) You say over 50c cetigrade. Well, my memory is never above 3-4c + case temp (which is another 1-2c+ ambient) If you read through people with issue, none of them were slacking on cooling the memory. I don't expect their to be much higher than 35-40c if that high.

so i agree that DFI is free of any responsability of memory deaths.. just because their line of nforce 4 mobos have an option of even 4.0 volts on ram,

Yes and No. DFI put the incomplete feature in, than well, we have issue. What is point of putting feature in that actually DO fail component? If this is not an issue with DFI than there is gotta be some other reason. Maybe my OCZ 600W PSU, maybe... well, bottom line is, this is happening MUCH more frequently with DFI than any other board. (s939 iteration)
Last thing we need is blindly blaming users and defending company -_-; Did anyone here say I hate DFI or something to that nature? We just want to figure out what is killing it and why and if we know how to... get this issue resolved.

PS) I am crazy enough to run 3.4-3.5 vdimm 24x7 on few other mobo I had with vdimm cap mod. Never did any memory I had fail due to over voltage or heat issues. (that is about 7gig of BH-5, 4gig of VX, 2gigs of UTT) -> one of gig of UTT is actually failed one. Now there is nothing statiscal about this but I can tell you, when I say I never had this issues around voltage I mentioned I don't mean just another pair of UTT or something.

[wtz54321]

PS) I am crazy enough to run 3.4-3.5 vdimm 24x7 on few other mobo I had with vdimm cap mod. Never did any memory I had fail due to over voltage or heat issues. (that is about 7gig of BH-5, 4gig of VX, 2gigs of UTT) -> one of gig of UTT is actually failed one. Now there is nothing statiscal about this but I can tell you, when I say I never had this issues around voltage I mentioned I don't mean just another pair of UTT or something.

It seems that UTT(UnTesTed) from winbond is not able to handle extremely high voltage as old-good BH5 or CH5 ... It seems that UTT is not 100% the same as BH5 because BH5 can not boot in CAS=3 setting , but UTT can boot in CAS=3 setting , maybe there is something different from UTT and BH5 to cause that UTT not able to handle extremely high voltage

[Tony]

It seems that UTT(UnTesTed) from winbond is not able to handle extremely high voltage as old-good BH5 or CH5 ... It seems that UTT is not 100% the same as BH5 because BH5 can not boot in CAS=3 setting , but UTT can boot in CAS=3 setting , maybe there is something different from UTT and BH5 to cause that UTT not able to handle extremely high voltage

UTT is made from older CH5 andf BH5 dies...it is the same as the old chips..its just untested by winbond....

Im running the jumper tweak with the 3.3line at 3.6v,I find it stable, cool and the ram clocks well.

[hovo73]

UTT is made from older CH5 andf BH5 dies...it is the same as the old chips..its just untested by winbond....

Im running the jumper tweak with the 3.3line at 3.6v,I find it stable, cool and the ram clocks well.

Is there a point for me to try this tweak if I can't change the +3.3V line on my PSU?

[Lastviking]

All your rams died in slot 2 or? no one killd ram in slot 1 and 3,4?

[nicholarse]

I also have problems although I'm not sure what is to blame. I'm trying to run a DFI SLI-D mobo, 4000+ sandy, 1GB OCZ PC3500 plat BH5, BFG 6800GT OC, PCP&C510 SLI etc etc

Originally I could only boot with one stick in orange slot 2 - this was the case on air and phase but now I can't boot at all whatever I memory I have (tried muskin lvII) and whatever slot. Its driving me nuts.

[wtz54321]

UTT is made from older CH5 andf BH5 dies...it is the same as the old chips..its just untested by winbond....

Im running the jumper tweak with the 3.3line at 3.6v,I find it stable, cool and the ram clocks well.

But why BH5 fail to boot in CAS 3 setting but UTT is ok ?

[P_1]

But why BH5 fail to boot in CAS 3 setting but UTT is ok ?

there are utt bh chips and utt ch chips afaik not even the ch utt chips can boot at cas 3

[celemine1Gig]

there are utt bh chips and utt ch chips afaik not even the ch utt chips can boot at cas 3

Plain wrong! You wanna see some Winbond based RAM at CL3, no problemo. All my UTT based sticks can boot at CL3. Also had a stick of rebadged CH-5 RAM some month ago (was one of the first CH-5; week 14 2003) and even that one did CL3. The question if the RAM does CL3 or not is just a matter of SPD-programming. Nothing more and nothing less.

[P_1]

Plain wrong! You wanna see some Winbond based RAM at CL3, no problemo. All my UTT based sticks can boot at CL3. Also had a stick of rebadged CH-5 RAM some month ago (was one of the first CH-5; week 14 2003) and even that one did CL3. The question if the RAM does CL3 or not is just a matter of SPD-programming. Nothing more and nothing less.

o my corsair sticks with ch5 wouldnt boot at cl3

[Snark42]

Hey all. Just lost a second stick of VX yesterday. This happened to me about 5 months ago, and one again yesterday. This sucks. I was using 5v line but only 3.2vdimm. I had 2x80mm fans blowing directly onto the RAM. Both of them died in the orange slot closest to the CPU (slot 2 I presume).

I've read this entire post and will be doing the following to prevent anymore mystery murders ( )...

1-buy a true 24-pin native power supply
(http://www.dfi-street.com/forum/show...2&postcount=34)
2-go back to the 3.3v feed.
3-RMA my PC3200 Gold VX and get PC4000 Gold VX
4-update to a BIOS that lets me use slots 1&3 dual channel
5-Pray to Jebus

Hopfully this will work.

[brandinb]

i also killed some ram with this board. at first i blamed the board but now i realize it was my fault. i was changing ram and didnt unplug the psu. i always did it this way on older mobos and it never harmed anything. now my ram is dead my fault.
i bet 5 bux that 90% of the ram "killed" by this board is a user error. unplug the psu from the wall and drain it with the power switch then open the case trust me on that!
________
Adelyn

[celemine1Gig]

o my corsair sticks with ch5 wouldnt boot at cl3

You didn't get it, did ya? I also have Winbond based RAM that does NOT boot at CL3, but the reason why it behaves like that is just the programming of the SPD-ROM. That's all! If the SPD-ROM was programmed differently, this RAM could also run at CL3.

[saaya]

ahhh so thats why some sticks with winbond ram do cl3 while others dont? hmmm
are you sure about this?

[Magnj]

stay off the freakin 5V line. Do the 3.3V jumper trick run up to 3.5V Vdimm and be happy.( note 3.3V rail above 3.7V + SATA drives = nasty ugly bad )

[persivore]

I think that the 'spike' of 5v when booting could be a contributing factor to the death of all of the winbond based RAM.
On most boards, before the BIOS is loaded, the board will give out 2.8 to 2.9v Vdimm. Once the BIOS has been loaded, the voltage returns to whatever is set in BIOS. With the DFI NF4, the 'spike' of 5v at boot is probably being produced because there is no BIOS loaded to tell the Vdimm voltage regulator what voltage to give out. A spike of 5v being fed to RAM could kill it, even if it were only for a very short space of time.

Some boards will also supply voltage to the RAM from the 5v standby line from the PSU when there is no power on. If this is the case with the DFI NF4 board, you could be feeding RAM 3.3v+ even when the board is in 'standby'.

I'm not saying that this is definitely what is happening, or that DFI is in anyway responsible for this. This it is just a theory.

[ADM]

I think that the 'spike' of 5v when booting could be a contributing factor to the death of all of the winbond based RAM.
On most boards, before the BIOS is loaded, the board will give out 2.8 to 2.9v Vdimm. Once the BIOS has been loaded, the voltage returns to whatever is set in BIOS. With the DFI NF4, the 'spike' of 5v at boot is probably being produced because there is no BIOS loaded to tell the Vdimm voltage regulator what voltage to give out. A spike of 5v being fed to RAM could kill it, even if it were only for a very short space of time.

Some boards will also supply voltage to the RAM from the 5v standby line from the PSU when there is no power on. If this is the case with the DFI NF4 board, you could be feeding RAM 3.3v+ even when the board is in 'standby'.

I'm not saying that this is definitely what is happening, or that DFI is in anyway responsible for this. This it is just a theory.

This sounds to me like a good explanation. Looks like I'll be RMAing my sticks next week and start using the 3.3v line from now on.

[eclypse]

Well like i said before i had 2 sticks of G-Skill PC4400 LE TCCD 512MB in the system and i only use the 3.3 line for vdimm and they died on me shortly after flashing to the 310 beta bios.

Voltage was at 3.2v on the vdimm and it was getting 3.24v occording to the bios and the Zippy 700 watt P/S was treaked with 3.50v via the 3.3 pot.

Ran fine for months like that.