**Official DFI LanParty UT P35-T2R Review/Overclock/Guide Thread**

[Praz]

It is entirely possible that something was damaged, but if there is something damaged, I can't see it readily. I've disassembled again to visually inspect each of the chips and nothing is visually identifiable. I'm going to reapply this stuff and try again.

I've looked into boot codes but all I've found is something about "Enable A20 Line", and I honestly don't know what that entails.

Good chance one of the PWM FETs were damaged. Stopping at hex code 26 is an indication of a faulty CPU or that it is not receiving the proper power.

[fireice2]

Here's my latest GTL insight...

For the above SS I used the following settings (BIOS)

Multi = x8.5
FSB = 445
Divider = 333/800 (5:6)
Vcore = 1.30000V
NB = 1.37v
VTT = 1.28v
Vdimm = 2.1v

GTL 58/58/58

When I used 62/62/60 it would fail in less than 1 minute. When I lowered the settings to 60/60/60 it streched to around 17 mins.

Seems like wolfdales like low GTL settings.

[Roger_D25]

Fireice2,
I see from your post above that you have your "VTT" set to 1.28v in BIOS. Well is that what you set it to or is that your actual voltage reading (either by using the BIOS Hardware Monitor or using a DMM)? I only ask because to get my actual "VTT" to run at 1.2v I need to set the actual BIOS voltage to 1.3v? Just curious if others need to do the same thing? By the way thanks for your help with GTL's!

[mrcape]

BTW I had the same finding that the GTLs had to be notched waaay down to high 50s to get prime stable with low vcore. Not sure how low you can go with those or what the effect would be.

[fireice2]

@roger
yes I set my VTT to 1.28v in BIOS. The real output is around 1.15v as indicated by smartguardian.

@mrcape
GTL tuning has become more challenging and a PITA with these 45nm procs. We really need to find a good starting point. My guess is that a good starting point would really be around 55-60.

[karateo]

@fireice2 VTT of 1.28 isn't it too high for a E8xxx?
Although the vcore is too low, i think that the VTT is responsible for degradation.

I try to stay at 1.12 !!!

[emoners]

Matt, is setting the VTT upto 1.28v good for 24/7? coz i saw this post: http://www.xtremesystems.org/forums/...2&postcount=19

unclewebb: I've been comparing the Intel datasheets today for the E6000 vs E8000.

As previously mentioned, the FSB Termination voltage has dropped from a recommended 1.20 volts +/- 5% to 1.10 volts +/- 5% with the switch to 45nm.

The E6000 had an absolute maximum voltage of 1.55 volts but the maximum VID was 1.50 volts. When operating between 1.50 and 1.55 volts you were in that gray area where "neither functionality nor long-term reliability can be expected."

With the E8000 the absolute maximum has dropped 0.10 volts to 1.45 volts but the maximum VID has dropped even further down to only 1.3625 volts. This implies that the new gray area for the 45nm cpus is from 1.3625 to 1.450 volts where "neither functionality nor long-term reliability can be expected."

It might be excessive FSB Termination voltage that is degrading some of the new 45nm chips or it might be good old excessive CPU voltage. Most XS overclockers would not consider 1.3625 volts to be excessive but it's possible that even one run above 1.3625 volts was enough to degrade the performance and long-term reliability of your chip.

[Roger_D25]

Just to verify, fireice2's actual VTT is only 1.15v, on the DFI P35 you need to set the VTT high in BIOS because it undervolts a bit.

[fireice2]

i'll try setting lower VTT. 1.15v doesn't seem too much to me.

[karateo]

Just to verify, fireice2's actual VTT is only 1.15v, on the DFI P35 you need to set the VTT high in BIOS because it undervolts a bit.

as I now Smartguardian reports false readings...
Also, as I have read this mobo strictly follows bios settings, so I guess you actually give 1.25+ VTT to your cpu

[Praz]

I haven't found any benefit yet in manually setting GTL values with the 8400 CPU.

[fireice2]

@karateo

the lowest that the board would allow me to set is 1.20v and the BIOS and SG would report 1.1v

@praz

I'd test again using while disabling GTL controls.

as of now I lowered VTT to 1.20v (BIOS) and OCCT would crash in about 30sec.

[karateo]

@karateo

the lowest that the board would allow me to set is 1.20v and the BIOS and SG would report 1.1v

you should update to the latest beta 09/01/08 bios with updated VTT tables for 45nm allowing for lowest 1.1 (which is the stock setting)

I mostly agree with this post
http://www.xtremesystems.org/forums/...29&postcount=7
keep VTT lower than 1.2 bios set and play with vcore up to 1.4 full load

[karateo]

@Praz exactly!
I have found no gain playing with GTL's yet.
I only set something like 63,63,67 as you suggested in case of bios overvolting.
I thought it was just because i am a noob overclocker

[fireice2]

i flashed back to 12/24 recently.

Will try updating bios when I get more time to play around. I still have an exam comming up so I'd rather not mess with my settings too much for the moment.

[emoners]

@Praz: is SG reporting your voltages right? so you're running 4Ghz at 1.32v VCore & 1.12v VTT?

[Roger_D25]

I just took a few minutes and measured the actual "CPU VTT" voltage with DMM to compare it with the BIOS settings. Here are the results I got on my board (BIOS 11/07)

"CPU VTT" BIOS Setting | Actual Measured Voltage
1.20v | 1.098v
1.25v | 1.147v
1.30v | 1.171v

I then set my "CPU VTT" to 1.20v in BIOS and measured the actual voltage under heavy load (Prime95 Small-FFT's)

1.20v | 1.098v (steady, the voltage didn't budge at all)

I hope that helps a bit.

[Praz]

Actual measured values are VCORE - 1.334, VTT - 1.121. One or more of the voltages may be able to go lower. I also have C1E and Vdroop enabled.

[fireice2]

@praz

what BIOS are you using?

[Praz]

@praz

what BIOS are you using?

01/09 BIOS.

[fireice2]

I tried disabling GTL controls OCCT would fail after 40mins or so.

Will try 1/09 later.

BTW, I was previously stressing CPU Only in my previous runs.

[Delemon]

I've disassembled and reapplied TIM and still no go. I inspected the PWM chips and I couldn't see anything with the naked eye, all corners sharp and no scratches. As a last ditch I'll try and see if another PSU might help. I think I've got an old OCZ lying around to try.

EDIT: Extra PSU was no help. I don't have an extra CPU or mobo to test. Anyone have a used T2R they want to sell? =) Looks like my water cooling fund has turned into replacement mobo fund.

[Roger_D25]

My fingers are crossed for you Delemon, hope you get sorted! Hopefully it is just some little thing you have been looking over...

[Speederlander]

Will this board run a graphics card at 16x and a RAID controller at 8x or do both slots drop to 8x when both are populated?

Thanks.

[MikeMK]

I just took a few minutes and measured the actual "CPU VTT" voltage with DMM to compare it with the BIOS settings. Here are the results I got on my board (BIOS 11/07)

"CPU VTT" BIOS Setting | Actual Measured Voltage
1.20v | 1.098v
1.25v | 1.147v
1.30v | 1.171v

I then set my "CPU VTT" to 1.20v in BIOS and measured the actual voltage under heavy load (Prime95 Small-FFT's)

1.20v | 1.098v (steady, the voltage didn't budge at all)

I hope that helps a bit.

Roger, I take it you are using 1224 bios. That has HUGE vdroop issues on 45nm CPU's in regard to VTT.

Give the 0109 bios a go, it solves that problem by adding new Voltage tables.