Overclocking Help i7 920

CptMalice · 30 Apr 2011 at 13:54

Got it to boot up at 4.0ghz using 1.268vcore, just testing for stability and checking what the temperatures are like.

Edit: Do you think a new cooler would fix the problem with the temperatures? Currently running at 80-85c.

setter · 30 Apr 2011 at 14:06

A new more beefier cooler could indeed help, but other factors to consider are, your ambient room temp, case cooling and airflow.

CptMalice · 30 Apr 2011 at 14:17

Well the case it's in now has pretty good airflow, and the room temperature isn't too warm.

So i think it's either the cooler struggling to cool the cpu or maybe i just haven't seated it/applied the thermal paste properly.

CptMalice · 30 Apr 2011 at 14:41

What would be the reccomended cooler to get at the moment? For under £50.

old gamer · 30 Apr 2011 at 16:06

the Titan Fenrir is pretty good for cooling the i7 920

CptMalice · 30 Apr 2011 at 17:55

Went and bought a new cooler, now running at 4.0ghz with 1.248vcore at around 70 degrees under full load. Trying to get it stable at 4.0ghz now.

Thanks for the help.

CptMalice · 1 May 2011 at 04:18

Blue screened when i left it running prime overnight. It ran error free for at least 1-2 hours before i left it, not exactly sure when it crashed though.

Settings were, 191 Bclk,20x Multi,Turbo Mode Enabled, 1.28vcore with load line calibration enabled, DRAM frequency 1523,Uncore 3046, QPI Voltage 1.315v ,DRAM Voltage 1.66v.

Any Suggestions?

CptMalice · 1 May 2011 at 07:27

PC just randomly restarted after about 80 mins in prime, no blue screen or anything, what would cause this? Settings are still the same as above.

paradigm · 1 May 2011 at 09:38

Are you 100% certain there was no BSOD? Have you disabled automatic restart on system failure?

Check the event viewer for the cause of the crash, and post back what error code it was, alternatively, what was the STOP code on the BSOD last night?

CptMalice · 1 May 2011 at 10:01

I'm looking in the event viewer right now, what code is it that i should be looking for?

Edit: It says Event ID-41 Source-Kernel Power, is that it?

paradigm · 1 May 2011 at 10:18

Look for an "Error" in the System Event Log, around the time of the BSOD, it might help to filter out for only events with ID's of 1001 and 1003, it will most likely come under the source of "BugCheck" and contain something along the following:

Code:

The computer has rebooted from a bugcheck. The bugcheck was: 0x0000009c 
(0x0000000000000000, 0xfffff88003501c70, 0x0000000000000000, 0x0000000000000000) 
A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: .

We want the bit following "The bugcheck was:" (So in the above example 0x0000009c, is the important bit).

CptMalice · 1 May 2011 at 10:23

The computer has rebooted from a bugcheck. The bugcheck was: 0x0000000a
(0x00000001401a3aa3, 0x0000000000000002, 0x0000000000000008, 0xfffff80002a9c969).
A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 043011-84256-01.

This is the only event with a code like that, it happened after about 4 hours when i left it overnight.

paradigm · 1 May 2011 at 10:25

0x0000000a is typically an unstable IMC or unstable RAM. Try upping the QPI voltage, if that fails (and your RAM is already at 1.66~), up the vcore again, slightly.

CptMalice · 1 May 2011 at 10:28

Ok, i'll give that a try, thanks.

CptMalice · 1 May 2011 at 11:32

Tried again at 1.25vcore and then again again with the vcore up a notch. Got BSOD twice, Codes are:

0x0000001E and 0x00000024, what do they mean? Do i just need more vcore?

paradigm · 1 May 2011 at 20:08

overclock.net guide said:
BSOD codes:
BSOD codes for overclocking
0x101 = increase vcore
0x124 = increase/decrease vcore or QPI/VTT...have to test to see which one it is
0x0A = unstable RAM/IMC, increase QPI first, if that doesn't work increase vcore
0x1E = increase vcore
0x3B = increase vcore
0x3D = increase vcore
0xD1 = QPI/VTT, increase/decrease as necessary
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)

Unfortunately they don't seem to have an entry for 0x00000024, but typically errors relate to vcore or qpi.

CptMalice · 1 May 2011 at 20:21

Yeah those were the problems, been running prime for 8 hours so far, no errors yet. Seems to be getting close to stable now.

Edit: How long should i be looking for it to run prime before deciding it's stable? 24 hours?

paradigm · 1 May 2011 at 20:49

Anything over 16+ hours I'd consider stable, I've never had anything fail beyond 12Hrs, but just to be sure, I'd give it a few hours headroom.

CptMalice · 1 May 2011 at 20:57

Ok, I'll leave it overnight and if it passes then that should be it stable, thanks again.

silversurfer · 1 May 2011 at 21:00

paradigm said:
0x0000000a is typically an unstable IMC or unstable RAM. Try upping the QPI voltage, if that fails (and your RAM is already at 1.66~), up the vcore again, slightly.

Is it correct that ram volts over 1.65 can damage the cpu mem controller