DFI NF4 Ultra/SLi-D BUGS and Fixes Thread

[eva2000]

Blue078, not the first report of this - seems this board needs 0.1-0.3v more vdimm for ram to run what it normally runs for some reason ?

anyone else experience this ?

[Blue078]

I'm not haveing any problems with Ballistix 4000 {micron} or OCZ Gold 4000 {hynix} they are runing at the same voltages and timings as on the NEO2.

It's just the PDP with XBL....they clocked super great on the NEO2 and in my single channel UT250-GB with only 2.7v

I thought this board was tweeked for TCCD

[OPPAINTER]

OPP :

what PSU are you using?

I use OCZ 520W.

OPP

[OPPAINTER]

opp I want your VX settings for this board. you're getting like 15mhz over mine. I know these can go higher, even in dual channel. Help a brother out.

I'll take a screeny later today of my A64 Tweak settings.

OPP

[trans am]

I'll take a screeny later today of my A64 Tweak settings.

OPP

Thanks, what about dram drive strength and data drive strength values? Those always confuse me. Is it like cas where your lower numeric value is stressed more? or does drive strength get weaker as the value increases?

[OPPAINTER]

Thanks, what about dram drive strength and data drive strength values? Those always confuse me. Is it like cas where your lower numeric value is stressed more? or does drive strength get weaker as the value increases?

I'll try and take some pictures of my BIOS settings.

OPP

[flexy]

Got my board installed yesterday, went through several MAJOR curse out battles
[...]

i dont want to sound like a dumba$$, but who installs a new system with exotic hardware like USB wheels plugged in ?

The first rule is that IF ANY PROBLEMS run the system with as few components as possible (even including disabling onboard components) and then, step by step, enable devices, plug in pci-cards etc.

AFTER THE SYSTEM IS UP AND RUNNING

ALSO...this is an interesting thread and i didnt follow THIS particular one.....can someone compile a FAQ or something more compact with all the issues and fixes so people do not have to go through the whole thread ?!

thanks !

(Btw. i didnt even know we HAD such major issues...especially i was assuming this board LOVES TCCD as everyone says.....now all of a sudden people claim issues with TCCD sticks ?)

[HARDCORECLOCKER]

i dont want to sound like a dumba$$, but who installs a new system with exotic hardware like USB wheels plugged in ?

The first rule is that IF ANY PROBLEMS run the system with as few components as possible (even including disabling onboard components) and then, step by step, enable devices, plug in pci-cards etc.

AFTER THE SYSTEM IS UP AND RUNNING

ALSO...this is an interesting thread and i didnt follow THIS particular one.....can someone compile a FAQ or something more compact with all the issues and fixes so people do not have to go through the whole thread ?!

thanks !

(Btw. i didnt even know we HAD such major issues...especially i was assuming this board LOVES TCCD as everyone says.....now all of a sudden people claim issues with TCCD sticks ?)

Now many poeple posted they had the issue with TCCD caused by using not enough Vdimm, 2,6 or so. 2,8 works fine and someone posted that the DFI even will have a bit more Vdimm than other boards.

BTW I can not understand people set the Vdimm to AUTO or 2,6 and expect a fine running system, many RAM brands require especally in DC 2,8 and maybe Oced 2.9 or more so it's always a good idea to start with 2,9 - will not damage anything and when all is running fine You can try to lower it a little.

[googles]

has anyone measured with dmm what voltages are being passed when they fix instability with more vdimm for tccd

where do you measure vdimm on this board?

[xgman]

Blue078, not the first report of this - seems this board needs 0.1-0.3v more vdimm for ram to run what it normally runs for some reason ?

anyone else experience this ?

I have 2x512 tccd running at 2.87v at 300HTT 1-1 2.5-3-3-7-1T on a beta SLI version. Also bios 0201. I can't figure out why Blue's tccd aren't getting along there.

[Blue078]

I have 2x512 tccd running at 2.87v at 300HTT 1-1 2.5-3-3-7-1T on a beta SLI version. Also bios 0201. I can't figure out why Blue's tccd aren't getting along there.

Neither can Blue

[BeachKEN]

I´ve got different problems:
When I´m listing DRAM Voltage options and confirm 2.9V or higher accepting it by Enter, it allways jumps back to 2.6V, but when using +/-, everything is fine.
Another one, setting the Vcore to 1.55V, it always gives me 1.5V, some bug in the BIOS ??? (Voltages in Enermax PSU excellent)

My Mushkin LVLII V2 with TCCD dont like more than 260 at CL2.5 4-4-11 1T at 2.8V more voltage doesn´t help at all. WTF??? I thought, this board loves TCCD´s

[AMDnewbie]

I´ve got different problems:
When I´m listing DRAM Voltage options and confirm 2.9V or higher accepting it by Enter, it allways jumps back to 2.6V, but when using +/-, everything is fine.
Another one, setting the Vcore to 1.55V, it always gives me 1.5V, some bug in the BIOS ??? (Voltages in Enermax PSU excellent)

My Mushkin LVLII V2 with TCCD dont like more than 260 at CL2.5 4-4-11 1T at 2.8V more voltage doesn´t help at all. WTF??? I thought, this board loves TCCD´s

For bios 0125, pls select your vdimm using page up/down and F10 save and quit directly. Don't go and set other things....doing this would save the vdimm...don't even press enter too......its solve on the latest beta bios....

[AMDnewbie]

ok guys, I had spend more than 4 hours trying different settings for my tccd at cl 2.5 with various settings and tweaking . Various clear cmos settings too...

everytime I tweak alittle, evey time I memtest it with #8 to ensure it is ok because #5 would show normal....

This pair of corsair proved to hit 275 on K8NF-9 but at this board cannot scale even 240 no matter what i try...vdimm, drive strength...whatever can set also try....

the problem is the "bank interleave", by disable it, I can memtest pass #8 with no errors up to 270....

thats the infor...I don't know why for my corsair pairs with cl2.5 over here..

no problem w twinmos sp up to 250Mhz on 3.2v @ 2-2-2-7-1T w bank interleave

maybe that is the problem w some tccd sticks..it just don't like some tccd w cl 2.5..........

ha....tire but adventurous...hope the bios issue can be resolve soon..

[split]

Seeing as you have stickied this, thought I might gather some info for the thread regarding the XP-120.

Seems it fits depending on orientation, and some thought must be given to ram height.

With pipes facing the gfx:
All 4 slots usable, though max ram height is about 40mm, are your modules taller you just can't fit them under there. And if your rams are taller then about 32mm you can't mount them after HS is mounted, and due to the way the HS is mounted (you tip it over towards the slots to get those hooks in the holder) it's going to be quite hard to mount it when rams are already in slots.

Pros: All 4 slots can be populated
Cons: Max ram height is 40mm for slot 3 & 4


With pipes facing the slots:
Slot 1-3 can now be populated with ram with any height. But HS is going to be close to the gfx which could lead to problem with some third parts cooling for gfx. And slot 4 cannot hold any ram, not even low profile.

Pros: slot 3 can now hold high ram modules
Cons: Slot 4 cannot hold any module at all and HS is very close to gfx.

[Blue078]

split is that PDP with XBL memory you have in that board ???

[split]

split is that PDP with XBL memory you have in that board ???

I dunno mate as it's not mine
I gathered it all from over here.

I see it being asked on several forums, and now it's in a sticky

[flexy]

XP-120:
Problem is we NEED to install two modules in 2 and 4 since these slots give best performance and 1/3 are either flakey or dont work at all.

My two OCZ rev2 are exactly 30mm btw.

So, it comes really down to the question:

* ) Install XP-120 first ???
* ) Install mem first (in dimm2 and 4) ???

What would be the bigger problem with the modules exactly 30mm tall (like the OCZ rev2s) ?

How did others do the procedure ?

[googles]

does anyone know if the xp-90 fits without any problems?

looking at side-by-side comparisons betw the xp-120 and xp-90, it looks to me that even the xp-90 will hover over the mem slot...

can anyone confirm? coz i need to make a purchase soon

[underpar]

does anyone know if the xp-90 fits without any problems?

looking at side-by-side comparisons betw the xp-120 and xp-90, it looks to me that even the xp-90 will hover over the mem slot...

can anyone confirm? coz i need to make a purchase soon

^^ same boat...thinking of getting an XP-90

[andyOCZ]

Hey guys,

How has anyone been able to run SLI mode on the Ultra-D without the SLI bridge between the 2 cards. It does not ship with one. Maybe I'm missing something here?

[ominae]

you need the sli bridge. A bridge design might come out in time - OR you can contact DFi and ask them to ship you one, because you "lost" your bridge :P

[andyOCZ]

LOL, Thanks. Makes the board a bit pricey for a useless SLI option.

[mrlobber]

A thing to test for all who are using any kind of high-voltage loving memory (BH-5, BH-6, maybe also CH-5, Twinmos UTT, OCZ VX et al) - well, except maybe for the very big guys (OPP, OPB), as for them it's running fine anyway

I'd like to know how your memory behaves in memtest test #5 IF using 3.5 Vdimm or more (which, of course, is derived from the 5v rail).

What I have observed with my BH-5 kit and OCZ VX as well, as soon as I up the Vdimm to 3.5v or more in bios, and loop memtest test #5 for a few loops (10 is enough, you can see it then), memory starts showing memtest errors EXACTLY AT THE SAME MEMORY ADDRESS all the time. Usually if memory is not memtest stable, it usually throws out errors at various addresses (maybe in the same range but almost never in the same address, as one chip on the memory is failing, and it is usually the size of 64 Mb now). Moreover, this happens also (my VX is memtest stable up to 262 htt 1:1) on lower speeds, say, 255 htt. The only variable here which seems to impact these errors, is the Vdimm. As soon as I lower it, the errors disappear.

Of course, I'm actively cooling both the RAM sticks as well as the 5v Vdimm jumper area.

So I'd like to know whether your boards and memory exhibit the same "issue" - to see whether this is an isolated incident or not.

For BH-5, I was running OPB's A64Tweaker settings, for OCZ VX - also, except Tras was 10, Trc - 12, Trfc - 16. Using 02/01 beta bios.

Thanks.

[BeachKEN]

XP-120:
Problem is we NEED to install two modules in 2 and 4 since these slots give best performance and 1/3 are either flakey or dont work at all.

My two OCZ rev2 are exactly 30mm btw.

So, it comes really down to the question:

* ) Install XP-120 first ???
* ) Install mem first (in dimm2 and 4) ???

What would be the bigger problem with the modules exactly 30mm tall (like the OCZ rev2s) ?

How did others do the procedure ?

Install the Xp120 first, RAM sticks are not so easy to be placed in 4th slot, but after few minutes and trying different angles you will get them right in.