Overclocking i7 975 EE - Take 2 - Can't keep stable above stock...

[AJTheGamer]

I have another thread, but it seems to have died...

Since I now have more of an idea of whats going on, and have actually managed to at least raise the BCLK from 133 to 178, seems to be as high as I go as raising the VTT from 1.26v to anything higher doesn't seem to make any difference, went as high as 1.4v, but with only a 2MHz increase and it not working I figured I'd stop there and try and work on the multiplier...

I managed to raise the multi to 16x, 2.8GHz, with only 6GBs (2x3GBs) RAM installed, and it ran Prime95 stable for over 10 minutes with no errors, so I moved onto 18x (3.2GHz) and it crashes.

Same thing happens if I turn off the Turbo and run Prime95 on stock settings of 3.33GHz, it crashes.

I'll take some photos of my BIOS shortly as its slightly different in the newest update to most the templates around.

Now it seems when I try to raise the 'CPU VID' from AUTO to anything else, it wants to crash, 1.1v, 1.15v, 1.2v, 1.25v, 1.3v, 1.35v, 1.4v, 1.45v.... Thats with the 178 BCLK and 18x multi...

So how can I start to work on the CPU VID without it crashing? Should I just take it back to the last stable settings and set the CPU VID manually? I was reading about vdroop and LLC or something, but I can't seem to find those settings...

 

[Carlitos714]

Use these codes to narrow down what you need to tweek. Download BlueScreenViewer if you dont catch the codes while it crashes or you have windows 8. Post what you are getting and ill try to help you

0x101 = increase vcore
0x124 = increase/decrease QPI/VTT first, if not increase/decrease vcore...have to test to see which one it is
0x0A = unstable RAM/IMC, increase QPI first, if that doesn't work increase vcore
0x1A = Memory management error. It usually means a bad stick of Ram. Test with Memtest or whatever you prefer. Try raising your Ram voltage
0x1E = increase vcore
0x3B = increase vcore
0x3D = increase vcore
0xD1 = QPI/VTT, increase/decrease as necessary, can also be unstable Ram, raise Ram voltage
0x9C = QPI/VTT most likely, but increasing vcore has helped in some instances
0x50 = RAM timings/Frequency or uncore multi unstable, increase RAM voltage or adjust QPI/VTT, or lower uncore if you're higher than 2x
0x109 = Not enough or too Much memory voltage
0x116 = Low IOH (NB) voltage, GPU issue (most common when running multi-GPU/overclocking GPU)
0x7E = Corrupted OS file, possibly from overclocking. Run sfc /scannow and chkdsk /r

 

[AJTheGamer]

Thanks, I'll get back to you, I'll crank it to the settings that it last crashed at and see if the codes match with any that you posted.

Thanks for the reply

 

[AJTheGamer]

I was able to turn the multi up to x18 on a 178 BCLK, ran Prime for 10 minutes, turned it up to x20 and I had errors on Blend test and BSOD on Small FFTs...

I attached screen for the last BSOD... Error code wasn't on the list posted.

 

[Carlitos714]

What speed was you ram at when you got that bsod? Also can you provide a link to the ram you have?

 

[AJTheGamer]

I had everything set pretty low.

The RAM in BIOS is set to BLCK x 06 (1068MHz) with UnCore of x12 (2136MHz) timings set to 9-9-9-24. Voltage at 1.65v Its Corsair XMS3 I only have 3 of the sticks in atm.

I just tried again and got a BSOD with the error code:
0x101 = increase vcore

But I'm not sure where to start since its been on AUTO.

The rest of my CPU Options in the BIOS look like this.

O.C. Shut Down Free: Enable O.C.S.D.F
CPU VID Control: AUTO
Power Saving: Standard
Super VID: OFF
VCore Auto PSI: Disabled
OCP: 140A
DRAM Voltage: 1.650v
DRAM PWM Switch Frequency: Nominal
DRAM PWN Phase Control: 2 Phase
CPU VTT Special Add: Auto
CPU VTT Voltage: 1.31v
VTT PWN Switch Frequency: Nominal
VTT PWN Phase Control: 2 Phase
CPU PLL Voltage: 1.80v
IOH/ICH 1.1v Voltage: 1.31v
IOH Analog Voltage: 1.10v
ICH 1.5 Voltage: 1.5v
ICH 1.05v Voltage: 1.05v
DIMM 1/2 DQ/DQSTB Bus VREF: -0.00%
DIMM 3/4 DQ/DQSTB Bus VREF: -0.00%
DIMM 5/6 DQ/DQSTB Bus VREF: -0.00%
DIMM 1/2 ADDR/CMD Bus VREF: -0.00%
DIMM 3/4 ADDR/CMD Bus VREF: -0.00%
DIMM 5/6 ADDR/CMD Bus VREF: -0.00%
CPU QPI Drive Strength: Normal
IOH QPI Drive Strength: Normal

 

[AJTheGamer]

Tried it at 1.3v for the CPU VID, it doesn't seem to like me changing that CPU VID, had a restart pretty quickly with Prime, with no BSOD....

EDIT - Tried the settings that worked with x18 and AUTO CPU VID, but changed the CPU VID to 1.265v and got a BSOD with error 0x0F4... Not really helpfully when I look it up...

 

[PolRoger]

Hi AJ,

I suspect that either your motherboard or your chip has problems and is not running like it should at normal spec. Ideally they should both be tested individually with other known good working parts to determine which of the two is at fault.

You could possibly try to find a friend or a tech shop who might have some compatible i7/x58 parts to test with or I guess you can run your system under clocked and look into getting replacement parts and then do further troubleshooting?

On Ebay they are now selling retired Intel Xeon W3520 (SLBEW) socket 1366 cpu for ~$15... This is exact equivalent of the i7-920 2.66GHz "D0" cpu. If a new cpu runs in a similar manner to your i7-975 then it would point to your motherboard being the problem.

http://www.ebay.com/itm/Intel-Xeon-...sor-/281376404191?pt=CPUs&hash=item41835732df

http://www.ebay.com/itm/Intel-Xeon-...US_Server_CPUs_Processors&hash=item43c0bc90d3

 

[AJTheGamer]

Alright I guess I'll look into that for thorough testing.

I guess question for now is, is it ok to do an 'unstable' overclock. By that I mean, I was able to raise the BCLK to 178, so am I ok to try using those settings and turning it to 150-160ish, turn the turbo and all that back on and see how it runs?

As long as it doesn't restart, etc I should be ok yeh? Advice here would be great

 

[Carlitos714]

Set your multi to 25x and base clock of 133. Once you get that stable, try blck of 150. Use the codes to help you OC more if you wish

Ram, just dont set it to anything higher than 1600 mhz. you can OC them later

Turbo Mode Function Enabled
CPU Non-Turbo Clock Ration (max # allowed)
QPI Frequency Auto
Uncore Frequency Auto
CPU Spread Spetrum Disabled
Thermal Management Control Disabled
EIST Function Disabled
CxE Function Disabled
Intel HT Technology Enabled
Active Processor Cores All

O.C. Shut Down Free: Enable O.C.S.D.F
CPU VID Control: 1.3
Power Saving: none or disable
Super VID: OFF
VCore Auto PSI: Disabled
OCP: 180A or the highest you are allowed
DRAM Voltage: 1.650v
DRAM PWM Switch Frequency: Nominal
DRAM PWN Phase Control: 2 Phase (can you go higher?)
CPU VTT Special Add: Auto
CPU VTT Voltage: 1.31v
VTT PWN Switch Frequency: Nominal
VTT PWN Phase Control: 2 Phase (can you go higher?)
CPU PLL Voltage: 1.80v
IOH/ICH 1.1v Voltage: 1.31v
IOH Analog Voltage: 1.10v
ICH 1.5 Voltage: 1.5v
ICH 1.05v Voltage: 1.05v
DIMM 1/2 DQ/DQSTB Bus VREF: -0.00%
DIMM 3/4 DQ/DQSTB Bus VREF: -0.00%
DIMM 5/6 DQ/DQSTB Bus VREF: -0.00%
DIMM 1/2 ADDR/CMD Bus VREF: -0.00%
DIMM 3/4 ADDR/CMD Bus VREF: -0.00%
DIMM 5/6 ADDR/CMD Bus VREF: -0.00%
CPU QPI Drive Strength: Normal
IOH QPI Drive Strength: Normal

 

[AJTheGamer]

Tried the settings you suggested, with Intel Burn Test it doesn't even finish one run before it restarts, with no BSOD... And No I can't go any higher on the Phase Control, only 1 or 2..

 

[Carlitos714]

So even at stock clocks you are getting a non bsod restart?

 

[AJTheGamer]

So even at stock clocks you are getting a non bsod restart?

Yes it seem so, I believe if I try those settings with AUTO CPU VID it makes it through a few passes on IBT, but either way non BSOD restart.

 

[Carlitos714]

To me it sounds like a psu or ram problem. Or maybe your cpu is bad. You have a psu testet by any chance? Run memtest on your ram.

Ive had similar problem before but I cant remember if it was my psu , ram or cpu

 

[AJTheGamer]

To me it sounds like a psu or ram problem. Or maybe your cpu is bad. You have a psu testet by any chance? Run memtest on your ram.

Ive had similar problem before but I cant remember if it was my psu , ram or cpu

Pretty sure its not the PSU, its brand new, and last night I tested stock settings with a brand new Kingston stick of ram, single stick, and it still crashed, and I've ran memtest on all my ram with no errors.

CPU seems to run fine, I turned the base clock up to 150, and it was running at about 4.0GHz with the turbo, played some games and had no issues, I might try to run the stock test again without the temp monitor running, since I have a good idea of the temps its getting at that level and its more than fine, as HWMonitor and others still cause restarts at lower settings that are stable with RealTemp, RealTemp only shows the CPU temps tho, so maybe at a certain level RealTemp interferes as well... Really hoping its not hardware, CPU seems more than fine otherwise...

If it is for some reason RealTemp, what could it be that the Temp Monitors are doing that are interfering with the stress test?

Might sound stupid but I've not turned windows defender or anything like that off, so will also do all that next attempt, try and take every variable out of the equation before I spend any money on a CPU or Mobo to test my current ones on...

 

[Carlitos714]

Pretty sure its not the PSU, its brand new, and last night I tested stock settings with a brand new Kingston stick of ram, single stick, and it still crashed, and I've ran memtest on all my ram with no errors.

CPU seems to run fine, I turned the base clock up to 150, and it was running at about 4.0GHz with the turbo, played some games and had no issues, I might try to run the stock test again without the temp monitor running, since I have a good idea of the temps its getting at that level and its more than fine, as HWMonitor and others still cause restarts at lower settings that are stable with RealTemp, RealTemp only shows the CPU temps tho, so maybe at a certain level RealTemp interferes as well... Really hoping its not hardware, CPU seems more than fine otherwise...

If it is for some reason RealTemp, what could it be that the Temp Monitors are doing that are interfering with the stress test?

Might sound stupid but I've not turned windows defender or anything like that off, so will also do all that next attempt, try and take every variable out of the equation before I spend any money on a CPU or Mobo to test my current ones on...

Try AIDA64 or try coretemp. I have seen RealTemp cause problems. Also remember that your gaming temps will be lower than you stress test. AIDA 64 also has its own stability test. Try uninstalling all cpu monitor programs. OCCT also has its own stress test and temp monitor.

I dont know what it might be with realtemp, but its gotta be something.

do some stress with realtemp on. when it does a restart uninstall realtemp and try the test again.

 

[AJTheGamer]

I tried AIDA and CoreTemp, as well as OCCT, they all caused restarts at stock.. With AUTO CPU VID it gets through one pass on IBT, while 1.3V setting definitely causes the PC to restart quicker, like 30 seconds...

Tried the stock settings without any monitors running, and ran IBT, same deal... Seems it might be hardware, my bets on motherboard, but thats just a gut feeling.

If its not a hardware issue, and its just a voltage thing, what might we be needing to play with? I've read a few i7 9xx OC things on this DFI Lanparty Jr T3H6 board, as well as the T3H8, and some people had to play with some of the DIMM voltage stuff, random things I've got no idea about... I just remember something along the lines of 'some preferred normal, some +15%, some -15%, etc' not exactly sure what they were referring to or what to adjust without advice (don't want to damage anything, especially if somethings already dodgy)