Associate
- Joined
- 4 Jun 2009
- Posts
- 1,675
- Location
- Solihull/Cardiff
Alright guys, here to ask your advice again 
I've got an i7 920 on an EX58-UD5 which I've had running stably at 21x200 for a year now, but last month my power supply packed up, and I sent the motherboard back first, assuming that was at fault, but that was returned no fault found. I've since got it all back and upgraded my faulty PSU and have been trying to regain my overclock. (No I didn't back it up, yes I am an idiot)
Earlier in the week I started at stock, did 20 pass LinX runs to prove it was stable, then off I went with the overclocking, but found whatever I did LinX failed almost instantly - Even switching from auto multiplier on the RAM to manual of the same value caused almost instant failures
So I did the sensible thing and ignored LinX and started using IBT and Prime95 instead
I've since gotten up to 21x191 on 1.3V (A lot less voltage than I used last time) and it runs 50 IBT passes on Max RAM, 16 Threads fine, and runs Prime overnight without failing. I thought I was onto a winner.
Then I tried LinX again at 2048MB, and it fails after seconds unless I massively up the voltage to 1.35V
To make things even more interesting the display locks up as soon as I begin run one, so I can only go off fan speed to tell what's going on 
It unlocks as the pass ends, then locks immediately as pass 2 begins and so on.
The graphics card is a 260GTX at stock settings, all being powered by a Corsair HX 850W power supply (Up from the 700W OCZ ModXStream)
Any ideas what's going on? The system clearly isn't 100% stable, although the LinX results seem very consistent on 1.35V at ~52 GFlops.
Any idea why LinX is finding errors IBT doesn't, or why my display is locking up?
Cheers
Matt
I've got an i7 920 on an EX58-UD5 which I've had running stably at 21x200 for a year now, but last month my power supply packed up, and I sent the motherboard back first, assuming that was at fault, but that was returned no fault found. I've since got it all back and upgraded my faulty PSU and have been trying to regain my overclock. (No I didn't back it up, yes I am an idiot)
Earlier in the week I started at stock, did 20 pass LinX runs to prove it was stable, then off I went with the overclocking, but found whatever I did LinX failed almost instantly - Even switching from auto multiplier on the RAM to manual of the same value caused almost instant failures
So I did the sensible thing and ignored LinX and started using IBT and Prime95 instead
I've since gotten up to 21x191 on 1.3V (A lot less voltage than I used last time) and it runs 50 IBT passes on Max RAM, 16 Threads fine, and runs Prime overnight without failing. I thought I was onto a winner.
Then I tried LinX again at 2048MB, and it fails after seconds unless I massively up the voltage to 1.35V
To make things even more interesting the display locks up as soon as I begin run one, so I can only go off fan speed to tell what's going on It unlocks as the pass ends, then locks immediately as pass 2 begins and so on.The graphics card is a 260GTX at stock settings, all being powered by a Corsair HX 850W power supply (Up from the 700W OCZ ModXStream)
Any ideas what's going on? The system clearly isn't 100% stable, although the LinX results seem very consistent on 1.35V at ~52 GFlops.
Any idea why LinX is finding errors IBT doesn't, or why my display is locking up?
Cheers
Matt