**Official DFI LanParty UT X48-T2R(and LT) Discussion

[CERO]

I just got my board, and it is my first DFI board, do any of you have some settings for a Q6600, espacially I never worked with GTL and dont know what it does, yes working with asus board is easy, I got a Q6600 that needs 1.288vcore for 3.6ghz and 3.825 ghz for 1.376vocre, and i got some pair of ballistix that are capable of 1200mhz 5-5-5-12 @ 2.00v. settings for 450x8, to start with? or 425x9? Thnx!

[wilson]

By Host Controller, are you refering to the Modules or the Northbridge itself...

Its the northbridge as it cant process all the data from all the dimm slots fast enough when set at full speed. Thats why two 2gb stick are better as it only use`s two dimm slots and this helps the mhc.

[wilson]

I just got my board, and it is my first DFI board, do any of you have some settings for a Q6600, espacially I never worked with GTL and dont know what it does, yes working with asus board is easy, I got a Q6600 that needs 1.288vcore for 3.6ghz and 3.825 ghz for 1.376vocre, and i got some pair of ballistix that are capable of 1200mhz 5-5-5-12 @ 2.00v. settings for 450x8, to start with? or 425x9? Thnx!

Try this for now
s.b 1.51v
n.b 1.35v
cpu vtt 1.250v
gtl 1/2-110
gtl 0/3-105
gtl nb 80

[Afterburner]

Its the northbridge as it cant process all the data from all the dimm slots fast enough when set at full speed. Thats why two 2gb stick are better as it only use`s two dimm slots and this helps the mhc.

That means I should not do 4 sticks @ 4-4-4-1 @ 1000mhz if I could ?

[JohnMike]

That means I should not do 4 sticks @ 4-4-4-1 @ 1000mhz if I could ?

Yes you can, only depends on your Ram. I have my setup with 4x1GB @1100, 5-4-4-12... Works fine!

[wilson]

i can run all 4 of mine at 1100Mhz but over time this is suppost to stress the NB and cause damage. Now if this is not true then can someone tell me different as i would like to be able to run mine at full speed without the thought of damage.

[Afterburner]

Agreed.....

[Praz]

From the Intel X48 data sheet. Doubtful Intel would spec a configuration that would damage the MCH.

• Using 1 Gb device technologies, the largest memory capacity possible is 8 GB,
assuming Dual Channel Mode with four x8 double sided un-buffered non-ECC or
ECC DIMM memory configurations. Note: The ability to support greater than the
largest memory capacity is subject to availability of higher density memory
devices.

[Psycho_eddie]

Windows boot times are considerably lower and level loading on games beats the hell out of my 2 brothers machines without raid 0.

Just got vista up and running Raid 0

The Config Raid array screen is displayed upon each bootup and lasts good 2 seconds... didnt have this holdup before so im not seeing any improvenent in boot times here!

Can disable this raid config being displayed each bootup? If not then going sell HD seems more hassel then its worth perhaps?

[Fluke_Dallas]

The first values that I posted were correct. The X38 and X48 BIOSes are wrong. I have just spoken to DFI and they are going to release a new BIOS with corrected values.

Any word on when the new BIOS will be out? Thanks.

[Clunk]

Any word on when the new BIOS will be out? Thanks.

Read back a bit, it isn't needed, I've updated the tables.

[Effet]

Try this for now
s.b 1.51v
n.b 1.35v
cpu vtt 1.250v
gtl 1/2-110
gtl 0/3-105
gtl nb 80

How is it possible to have different GTL Ref voltages on cpu 0/1 and 2/3 when GTL voltages are derived from CPU VTT? That's not making sense dude :P

Read this to undersand GTL voltages better: http://edgeofstability.com/articles/Article.html

[Big Lar]

@ Clunk Thank you Very much for the GTL settings, they helped alot.

@ Praz You are a wealth of information kind Sir. Just wanted to let it be known I value all the help from both of you

Larry

[Praz]

How can it possible to have different GTL Ref voltages on cpu 0/1 and 2/3 when GTL voltages are derived from CPU VTT? That's not making sense dude :P

Read this to undersand GTL voltages better: http://edgeofstability.com/articles/Article.html

Each GTL setting in the BIOS has it's own opamp. The input of each opamp is tied to VTT. The value set in the BIOS determines the output of each amp.

@ Big Lar
Thank you. Just trying to help out somewhat. No sense all of us having to bang our heads against the wall if it can be helped.

[zsamz_]

How is it possible to have different GTL Ref voltages on cpu 0/1 and 2/3 when GTL voltages are derived from CPU VTT? That's not making sense dude :P

Read this to undersand GTL voltages better: http://edgeofstability.com/articles/Article.html

does it have to make sense lmao
as long as it works

[Effet]

Each GTL setting in the BIOS has it's own opamp. The input of each opamp is tied to VTT. The value set in the BIOS determines the output of each amp.

But those values seems to be quite high for VTT volt at 1,20? Or have I got it all wrong now? And there's a difference of +5 between 0/1 and and 2/3.

[emoners]

is the UT still available? all i can find at MotherboardPro is the LT-version

[RGone]

is the UT still available? all i can find at MotherboardPro is the LT-version

So far as we know, the UT has not yet been released.

[zsamz_]

everything much hotter now with those high gtls i gonna lower thaem a bit

[Jor3lBR]

But those values seems to be quite high for VTT volt at 1,20? Or have I got it all wrong now? And there's a difference of +5 between 0/1 and and 2/3.

Guys just read the thing a bit more carefully...

The value isn't high, the real value readings are the ones that you should look for so please search for the REAL value and forget about the ones you see in bios...

So when you see a 110 in bios it's actually 0.812v measured with a DMM by Clunk and the differences between the GTL's is because Clunk is trying to make our life even easier, so to get both GTL's on the ball park of 0.81v for a 1.213vtt (1.250v in bios) as an example, you should use 110 on one and 105 on the other that will give a 0.812-0.816 on each (67% of 1.21vtt)!

Go with Clunk's chart that you will see the stability happening. But ofcourse some boards like different percentages, my X48 loves 69% so I used the chart to find the real 69% on each GTL...

[Clunk]

I should point out that the voltages were measured by Jimmy @ DFI

[adya]

just finished running prime for 2 hours , cpu ram mixed. all passes ! yesss!
so far i have 475fsb x8 =3.8ghz memory =1140mhz.
still sometimes it when i start it up , it turns on and off. but it did that with a 450fsb. i think its ok.... i hope lol
in bios i have -cpu vtt1.250
gtl 1/2 110
gtl 0/3 105
n.b gtl 80
is that right even for dual core? i read before that gtl 0/3 was for quad?
also my nb is at 1.456v is that too high? should i try reducing it abit?

[Praz]

just finished running prime for 2 hours , cpu ram mixed. all passes ! yesss!
so far i have 475fsb x8 =3.8ghz memory =1140mhz.
still sometimes it when i start it up , it turns on and off. but it did that with a 450fsb. i think its ok.... i hope lol
in bios i have -cpu vtt1.250
gtl 1/2 110
gtl 0/3 105
n.b gtl 80
is that right even for dual core? i read before that gtl 0/3 was for quad?
also my nb is at 1.456v is that too high? should i try reducing it abit?

GTL0 is for core 0 and GTL1 is for core 1. Whatever GTL value that brings stability to your system is the right value.

[adya]

ahhh k , now i get it. i must of misread it before.
thanks praz.

[adya]

FATAL ERROR: Rounding was 0.4997558594, expected less than 0.4
Hardware failure detected, consult stress.txt file.
Torture Test ran 2 hours, 45 minutes 32 seconds - 1 errors, 0 warnings.
Execution halted.
guess it means unstability. should i try uping my cpu vid?maybe to 1.47v ?