Auto as far as I am aware? Well, I have never changed the setting in the BIOS so I'm not sure what the default value is.
Liam
i5 Overclock
- Thread starter LEK-89
- Start date
Auto as far as I am aware? Well, I have never changed the setting in the BIOS so I'm not sure what the default value is.
Liam
I wouldn't have thought so - especially as you were runing small ffts - had it been a blend test there may have been an excuse to up the memory voltage by an increment or two. You've already tested extensively for memory stability (previous thread) so, for now, we'll assume it's not a factor for such a modest clock.
I can't remember but i think it's either low or high (i'm not sure it has an auto - forgotten TBH) - if it passes at 1.3V you could try auto voltage again but this time with 'high vDroop' selected.
The reason for this is that at 4.4Ghz at auto you were maxing at 1.27V - with 'high vDroop' enabled it may give you sufficient juice at load to keep things stable. If it does then you will still have the idle voltage benefits...
Yeah I thought it was unlikely that the RAM would affect it, as I have been running it set at 1600Mhz 1.5v manually for the past couple of days now on stock CPU speeds without a problem.
Thanks - Liam
Last edited:
It shouldn't with the static 1.3V - but would if you tried the auto voltage at 4.4 again.
I've just stopped Prime and gone back into the BIOS. Vdroop is indeed set to Auto, the only options are to change it from auto to Low vDroop. Shall I change it to low and do a prime test again with auto CPU voltage?
Liam
What voltage was CPUz recording using 1.3V at full load?
Ok I will continue the Prime test again with no settings changed.
Thanks again Plec - Liam
Soldato
- Joined
- 22 Mar 2009
- Posts
- 5,787
Plec why do you recommend small ffts ? as this would not class as fully stable as it must do blend test instead to ensure it 100% stable !
I haven't and didn't recommend 'only' running small ffts.
If you read the posts i've mentioned blend will need to be run also as its another valuable test for stability - it's just that small ffts stresses the CPU more. I just happen to run small ffts first...
Soldato
- Joined
- 22 Mar 2009
- Posts
- 5,787
ah right, ok fair enough. I just re-run prime at small ffts now to see if it was anything to do with my cpu not stable enough as if it passed the cpu then it could be my memory issues on the blend test failed. Thanks Plec.
No worries.
I may regret asking this, but what stabilty issues are you having?
Do they appear in stress testing apps only or are they random events during general OS use.
The reason i ask is that a clock can sometimes pass every stress test going but still fail in day to day usage.
Stress testing is a great early indicator for clock instability but, unfortunatley, it isn't the 'be-all and end-all' in guaranteeing a stable clock.
Soldato
- Joined
- 22 Mar 2009
- Posts
- 5,787
I don't know yet to be honest, since I brought i7 2600K and the board Z68 and had so many cold boot issues here:
as I was pretty upset myself over this cold boot issues and this is far the worse clicking sound noise in my experience, never had like that before on my old EX58 board.
Got BSD code 0x124, 0x101, 0xD1, 0x116 on stock speed. Took the board back to the shop and they tested for 24 hours and told me it was stable @ 4.6GHz no problem and the board had been replaced a new one, same board (But I can tell they never swap the board cos of I did check the serial number as it was the same as before) took the board home and now facing SSD issues problem with freeze, try HDD and it seem ok and passed 4.6GHz and put SSD back in, got BSD straight away with 0x124.
Got BSD code 0x124, 0x101, 0xD1, 0x116 on stock speed. Took the board back to the shop and they tested for 24 hours and told me it was stable @ 4.6GHz no problem and the board had been replaced a new one, same board (But I can tell they never swap the board cos of I did check the serial number as it was the same as before) took the board home and now facing SSD issues problem with freeze, try HDD and it seem ok and passed 4.6GHz and put SSD back in, got BSD straight away with 0x124.
So is the clock essentially is stable without the SSD installed?
If it is have you tested to see if the machine is stable at stock with the SSD?
Soldato
- Joined
- 22 Mar 2009
- Posts
- 5,787
What happen is:
My SSD is windows 7 x64 bit with all drivers coming from my old EX58 UD5 board and I just put new board in and the i7 2600K and windows automatically found new hardwares and updated. I put Gigabyte CD in and updates all drivers, and checked all my hardwares in the system and all look good. Then few days later got BSD.
My HDD is a clean installed fresh of windows 7 on it with this new board and chip and had no issues.
Soldato
- Joined
- 22 Mar 2009
- Posts
- 5,787
Damnit
Prime has just error'd again on Worker no 3, 40 minutes into the test!
Do you think that it's best to step down to 4.2Ghz at auto voltage and go on from there?
Thanks - Liam
I know the feeling, I had that as well. Pretty pain in the butt.
That really depends on you and what you want to get out of the chip.
But as you seemed quite happy with 4.4GHz and cool/silence seems key i would be temped to test 4.3GHz at auto voltages.
I know it's only dropping 100MHz but it could make all the difference.
Or, alternatively, you could try pumping ~1.31V @ 4.4GHz as CPUz was only registering 1.288V(?) (at 1.3V) so you have room to maneuver.