DFI UT X58-T3EH8 continued - Part 2

[owikh84]

DFI Engineer's reply regarding my 2nd time Error Code 88 on UT..

Regarding the problem description and RMA record I got, I noticed it is a burned case of CPU Vcore PWM; our RMA engineer has replaced a new PWM IC to let it function again.

But, now it gets 8.8 error code again (8.8 is indicate there is some voltage not ready yet, BIOS can not be started). I suspected, it probably is electric-attribute changed of some components on CPU Vcore circuit and damage PWM IC again. Of course, we need to do further analysis for determining a root cause is or not at final.

My suggestion is:

Due to it is a second RMA case and CPU Vcore PWM IC is changed, it probably has electric-attribute changed of some components.

I will ask RMA ready a new board to replace your defective board. Once we get your agreement and RMA is applied, I will do the actions to arrange board to our agent as soon as we can.

Have a nice day!
DFI FAE

Jimmy
jimmy.su@dfi.com.tw

[GMX]

Discovered new BIOS today.
No idea what 'broken' PLL Voltage is, left it at default/1.8v

After 1 year, I finally spent a bit of time trying to get my CPU somewhere.
Currently sitting @ 196*20

I'm stuck @ 196 Bclck with 1.45vtt, 197+ will return 8.8. on the diagnostic LED. Have to clear CMOS to get it to post again. Is there some kind of hole?

Genie BIOS Setting

CPU Feature
Set VR Current Limit Max: Enabled
Thermal Management Control: Disabled
EIST Function: Disabled
CxE Function: Disabled
Execute Disable Bit: Disabled
Virtualization Technology: Disabled


***** Logical Processor Setting *****
Intel HT Technology: Enabled
Active Processor Cores: All

DRAM Timing
All auto except 9-9-9-25 @ 2T

Voltage Setting
O.C. Shut Down Free: Disable O.C.S.D.F
CPU VID Control: 1.356
CPU VID Special Add Limit: Enabled
CPU VID Special Add: Auto
Vcore Droop Control: Enabled
DRAM Bus Voltage: 1.620v
DRAM PWM Switch Frequency: nominal
DRAM PWM Phase Control: 2 Phase Operation
CPU VTT Special Add: AUTO
CPU VTT Voltage: 1.45v
VTT PWM Switch Frequency: nominal
VTT PWM Phase Control: 2 Phase Operation
CPU PLL Voltage: 1.80v
IOH/ICH 1.1V Voltage: 1.13v
IOH Analog Voltage: Default
ICH 1.5 Voltage: 1.7v
ICH 1.05V Voltage: 1.15v
DIMM 1/2 DQ/DQSTB Bus VREF: -00.0%
DIMM 3/4 DQ/DQSTB Bus VREF: -00.0%
DIMM 5/6 DQ/DQSTB Bus VREF: -00.0%
ADDR/CMD VREF Control: Disabled
x DIMM 1/2 ADDR/CMD Bus VREF: 110
x DIMM 3/4 ADDR/CMD Bus VREF: 110
x DIMM 5/6 ADDR/CMD Bus VREF: 110
CPU QPI Drive Strength: Normal
IOH QPI Drive Strength: Normal

PPM Function: Enabled
Turbo Mode Function: Disabled
1 core Max Turbo Ratio: 22x
CPU Non-Turbo Clock Ratio: 20x

* BCLK/UCLK/QPI Controller Settings *
QPI Control Settings: Enabled
QPI Link Fast Mode: Enabled
QPI Frequency: 4.8GT(lowest)
CPU Base Clock (BCLK): 196 Mhz
Boot Up CPU Base Clock: 155
PCIE Clock: 100 Mhz
DRAM Frequency: x10
UnCore Frequency: x20

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled

[donmarkoni]

Discovered new BIOS today.
No idea what 'broken' PLL Voltage is, left it at default/1.8v

After 1 year, I finally spent a bit of time trying to get my CPU somewhere.
Currently sitting @ 196*20

I'm stuck @ 196 Bclck with 1.45vtt, 197+ will return 8.8. on the diagnostic LED. Have to clear CMOS to get it to post again. Is there some kind of hole?

Try changing "Boot Up CPU Base Clock: 155" to 185 or 180, as it sets memory timings like you would run your system @ BCLK 155 and RAM at 155x10 and when the speed rises after that - system locks up.
I have DFI UT X58, but on the shelf now.
EDIT: To conclude: try to keep it 10-15MHz lower then BCLK.

[GMX]

Thank you for the advice.
By the theory of the system tightening timings if the gap is too large between bootup and effective, what if you set them to same values? Should be the most stable?

Here are new settings, currently testing LinX for stability/temps.

Genie BIOS Setting

CPU Feature
Set VR Current Limit Max: Disbaled
Thermal Management Control: Disabled
EIST Function: Disabled
CxE Function: Disabled
Execute Disable Bit: Disabled
Virtualization Technology: Disabled


***** Logical Processor Setting *****
Intel HT Technology: Enabled
Active Processor Cores: All

DRAM Timing
Memory Lowgap: 2048MB(optimized default setting)
All Auto except 9-9-9-25 @ 2T

Voltage Setting
O.C. Shut Down Free: Enable O.C.S.D.F
CPU VID Control: 1.33125
CPU VID Special Add Limit: Enabled
CPU VID Special Add: Auto
Vcore Droop Control: Disabled
DRAM Bus Voltage: 1.650v
DRAM PWM Switch Frequency: Nominal
DRAM PWM Phase Control: 2 Phase Operation
CPU VTT Voltage: 1.44v
VTT PWM Switch Frequency: nominal
VTT PWM Phase Control: 2 Phase Operation
CPU PLL Voltage: 1.80v
IOH/ICH 1.1V Voltage: 1.23v
IOH Analog Voltage: 1.1v
ICH 1.5 Voltage: 1.7v
ICH 1.05V Voltage: 1.15v
DIMM 1/2 DQ/DQSTB Bus VREF: -00.0%
DIMM 3/4 DQ/DQSTB Bus VREF: -00.0%
DIMM 5/6 DQ/DQSTB Bus VREF: -00.0%
ADDR/CMD VREF Control: Disabled
x DIMM 1/2 ADDR/CMD Bus VREF: 110
x DIMM 3/4 ADDR/CMD Bus VREF: 110
x DIMM 5/6 ADDR/CMD Bus VREF: 110
CPU QPI Drive Strength: Strong
IOH QPI Drive Strength: Normal

PPM Function: Enabled
Turbo Mode Function: Disabled
1 core Max Turbo Ratio: 22x
CPU Non-Turbo Clock Ratio: 20x

* BCLK/UCLK/QPI Controller Settings *
QPI Control Settings: Enabled
QPI Link Fast Mode: Enabled
QPI Frequency: x36/4.8GT(lowest)
CPU Base Clock (BCLK): 190 Mhz
Boot Up CPU Base Clock: 190
PCIE Clock: 100 Mhz
DRAM Frequency: x8
UnCore Frequency: x18

CPU Spread Spectrum: Disabled
PCIE Spread Spectrum: Disabled

Updated settings.
Seems much more stable.

FYI I was still on May 11 2009 (1105) BIOS before this and could not get above 187/3.74ghz. Though I admit back then I was over OCing so I just set and forget it haha

[Mafio]

flash the latest beta (D615)

[GMX]

flash the latest beta (D615)

826 is out.

[Mafio]

826 is out.

thx

i'm going to try it

[donmarkoni]

Thank you for the advice.
By the theory of the system tightening timings if the gap is too large between bootup and effective, what if you set them to same values? Should be the most stable?

Probably. If you set it on Auto it will be the same also.

Good to hear it's working better for you now.

EDIT: BTW, try dropping ICH voltages to default, there should be no need for raising them for overclocking.

@everybody: are there reasons to put my old DFI UT X58 to the tests again since two new BIOS'es are out?

[GMX]

I cannot really comment on that since I am coming straight from the 0511 bios.
I went to DFI's site to get the 615 everybody has been reporting about most recently in this thread and discovered 826.

Once again, thanks for the support. Been a long time since I been here, happy to be back for a bit.
I will make those voltage changes to NB/SB, overall i haven't got 124C BSODS but lots of 09C/101 which makes me think the CPU is just hitting it's limit on air.


My temps also seem high and min/max range between cores @ 100% load is more than 10C. System (which I think is PWM area) reported in Smart Guardian is pretty well under control. Under 80 C 100% load for about 30min. Vdroop disabled.


Do the Switching frequency of the VTT/DRAM PWM go hand in hand? Generally how should these be adjusted?

[Mafio]

Probably. If you set it on Auto it will be the same also.

Good to hear it's working better for you now.

EDIT: BTW, try dropping ICH voltages to default, there should be no need for raising them for overclocking.

@everybody: are there reasons to put my old DFI UT X58 to the tests again since two new BIOS'es are out?

BIOS D615 make the board pretty fast.
in term of stability it's about the same as D429.

i can do the same things with both of them, but with D615 i gain at least 2 or 3 seconds at super-pi 32M.

I cannot really comment on that since I am coming straight from the 0511 bios.
I went to DFI's site to get the 615 everybody has been reporting about most recently in this thread and discovered 826.

Once again, thanks for the support. Been a long time since I been here, happy to be back for a bit.
I will make those voltage changes to NB/SB, overall i haven't got 124C BSODS but lots of 09C/101 which makes me think the CPU is just hitting it's limit on air.


My temps also seem high and min/max range between cores @ 100% load is more than 10C. System (which I think is PWM area) reported in Smart Guardian is pretty well under control. Under 80 C 100% load for about 30min. Vdroop disabled.


Do the Switching frequency of the VTT/DRAM PWM go hand in hand? Generally how should these be adjusted?

PWM switching frequency make a sligth diffirence, for DU you don't need to adjust them.

Vcore vdroop Disabled in bios make all the voltage more stable, if you use it Enabled in bios pay attenction at the other voltages cause maybe vcore is ok but VTT etc etc under load droop a bit too much.

[p2501]

Any idea if the board supports some six-core action with the newer BIOSes, like 970 or 980x? Regarding PLL, is it still nailed to 1.8v or can we go below that now?

[Mafio]

it support six core and seems that can oc them quite well

http://hwbot.org/community/submissio..._980x_6612_mhz

[p2501]

That's wonderful, so I'm just selling my 920 and not the board. =)

Any info on the PLL issue?

[Mafio]

That's wonderful, so I'm just selling my 920 and not the board. =)

Any info on the PLL issue?

with D615 bios PLL can't be setted below 1.8 volt, probably with D829 will be the same thing.
BTW, i don't know if with 980x PLL lower than 1.8 volt can be usefull but with classified + 920 or 950 lower the pll don't give me 1 mhz more than when i use 1.8 volt.

[cstkl1]

with D615 bios PLL can't be setted below 1.8 volt, probably with D829 will be the same thing.
BTW, i don't know if with 980x PLL lower than 1.8 volt can be usefull but with classified + 920 or 950 lower the pll don't give me 1 mhz more than when i use 1.8 volt.

+1million
thats y u dont see ppl running multiple times linx 100 runs

just leave it at 1.8v
the pll thing is a myth and gives temp stability to unstable systems

[p2501]

Okay, well I've just heard that it may bring stability with reduced temps. If that's not the case then it doesn't matter anyway. Thanks for the insight. =)

[Mafio]

+1million
thats y u dont see ppl running multiple times linx 100 runs

just leave it at 1.8v
the pll thing is a myth and gives temp stability to unstable systems

some ppl say that low PLL change cold bug, probably it's true but not with all cpu.
with my 4 i7 45nm lower PLL don't make any diffecence.

[cstkl1]

Okay, well I've just heard that it may bring stability with reduced temps. If that's not the case then it doesn't matter anyway. Thanks for the insight. =)

well it doesnt affect prime. since prime is a continous on custom blend

but try linx
u will see mix results on long runs.

[donmarkoni]

I didn't believe lowering CPU PLL is of any help, until I bought Gulftown.

[GMX]

What is the limiting factor of uncore frequency?
Specifically which voltage parameter(s)?

[Mafio]

VTT voltage

[UNSTABLE_HERO]

Thank you for the advice.
By the theory of the system tightening timings if the gap is too large between bootup and effective, what if you set them to same values? Should be the most stable?

Here are new settings, currently testing LinX for stability/temps.



Updated settings.
Seems much more stable.

FYI I was still on May 11 2009 (1105) BIOS before this and could not get above 187/3.74ghz. Though I admit back then I was over OCing so I just set and forget it haha

for 24/7 oc its better running with vdroop enabled to avoid spikes!better put cpu drive strength to NORMAL for stability! strong 9 to 10 times makes things worse!give around 1.36-1.37 CPU so with vdroop enabled @ load should be around 1.32-1.33!

[Forin]

There is new stable BIOS. D908.

[Ccaution]

24GBs max RAM now. Nice!

No new boards - but still supporting the oldies. That's how I spell "Customer Care"

[donmarkoni]

WTF? They are making more new and good BIOS'es now that the LanParty line is officially dead, then before.
Should I take my DFI UT X58 from the shelf and try it again? Are there any significant changes, (like fixed RTL timings, not +/-n)?