NJDevilsFan21
08-23-2007, 07:51 AM
I think I somehow damaged my computer chip... not fully, but just slightly in that I can no longer get overclocks like I once could with this Q6600.
I had a whole chart, and was making an entire chart of stable settings from 3GHz all the way to 3.6GHz on my Q6600. Anyway, some of the settings that I found stable were...
3006MHz - 1.1750V bios, +0.05FSB, +.05MCH [7 hours stable]
3195MHz - 1.2125V bios, +.1FSB, +.1MCH [2 hours stable]
3330MHz - 1.2625V bios, +.1FSB, +.1MCH [1 hour stable]
3510MHz - 1.375V bios, +.15FSB, +.15FSB [5 hours stable]
3600MHz - ?
Last night I decided to stress test 3600MHz. I set 1.375V as a 'safe' start. I loaded prime, and watched my temperatures climb. After 15 minutes the cores were at 67C 67C 63C 63C - which was warm, but below Intel's thermal spec of 71C. I decided to check up on it an hour later, which I did. Temps were slightly higher, at 68C 68C 65C 65C - but the small FTT Orthos was still running strong. I ended up falling asleep - and about an hour and fifteen minutes later, the beeping from my computer woke me up - Orthos had failed at with approximately 2:25 done. So, I go back into the bios, and bump up VCore one notch. I check temps, again the highest the got after an hour was 68C 68C 65C 65C, so I fell asleep and woke up again to beeping. This time it almost made it to the three hour mark. I bump up the vcore again one notch, I believe I was at 1.39375V at this point in the bios, and go to run the stress test again. This time the computer just froze. At this point I was exhausted and just wanted some sleep. So I shut the computer off.
This morning I woke up, and decided I wasn't going to try to find a stable point anymore. I went back to 3195MHz which I ran on a regular basis. I put in my settings, and load Windows. I notice that Windows took extremely long to load, and that AVG had crashed. I load up prime, and it fails in 1 second. What?!? So, I recheck my bios settings, and everything is what I had them at before. I reload Windows, and prime crashes again in a few seconds. I run large fft for a second, and that also crashes. So I thought maybe it was RAM problem. But Memtest86+ checked out fine, and I even switched out the Ballistix I was running for an identical pair. But prime still fails in a few seconds.
I don't know if I should blame myself for this - I thought I was being safe by staying under 70C loaded. I have a HX520 and I measured the 12V rail at 3510MHz and it was rock solid. No ripple at all. It was 11.98V at idle and 11.97V under load constantly. The RAM... well I already explained above. Maybe something went wrong with the P35-DQ6, but I just set 1.3V in the bios and 3195MHz primed fine for a few minutes. I have no idea how to test for motherboard issues either, when it boots fine. I did reset CMOS and started from a fresh bios.
So... :( :down: :shakes: :shrug: :confused:
What happened?!?
Edit: By the way, the cores fail randomly. It's not like Core 1 or Core 2 are the two that constantly fail. Every single one 1/2/3/4 has been the first to fail at some point.
I had a whole chart, and was making an entire chart of stable settings from 3GHz all the way to 3.6GHz on my Q6600. Anyway, some of the settings that I found stable were...
3006MHz - 1.1750V bios, +0.05FSB, +.05MCH [7 hours stable]
3195MHz - 1.2125V bios, +.1FSB, +.1MCH [2 hours stable]
3330MHz - 1.2625V bios, +.1FSB, +.1MCH [1 hour stable]
3510MHz - 1.375V bios, +.15FSB, +.15FSB [5 hours stable]
3600MHz - ?
Last night I decided to stress test 3600MHz. I set 1.375V as a 'safe' start. I loaded prime, and watched my temperatures climb. After 15 minutes the cores were at 67C 67C 63C 63C - which was warm, but below Intel's thermal spec of 71C. I decided to check up on it an hour later, which I did. Temps were slightly higher, at 68C 68C 65C 65C - but the small FTT Orthos was still running strong. I ended up falling asleep - and about an hour and fifteen minutes later, the beeping from my computer woke me up - Orthos had failed at with approximately 2:25 done. So, I go back into the bios, and bump up VCore one notch. I check temps, again the highest the got after an hour was 68C 68C 65C 65C, so I fell asleep and woke up again to beeping. This time it almost made it to the three hour mark. I bump up the vcore again one notch, I believe I was at 1.39375V at this point in the bios, and go to run the stress test again. This time the computer just froze. At this point I was exhausted and just wanted some sleep. So I shut the computer off.
This morning I woke up, and decided I wasn't going to try to find a stable point anymore. I went back to 3195MHz which I ran on a regular basis. I put in my settings, and load Windows. I notice that Windows took extremely long to load, and that AVG had crashed. I load up prime, and it fails in 1 second. What?!? So, I recheck my bios settings, and everything is what I had them at before. I reload Windows, and prime crashes again in a few seconds. I run large fft for a second, and that also crashes. So I thought maybe it was RAM problem. But Memtest86+ checked out fine, and I even switched out the Ballistix I was running for an identical pair. But prime still fails in a few seconds.
I don't know if I should blame myself for this - I thought I was being safe by staying under 70C loaded. I have a HX520 and I measured the 12V rail at 3510MHz and it was rock solid. No ripple at all. It was 11.98V at idle and 11.97V under load constantly. The RAM... well I already explained above. Maybe something went wrong with the P35-DQ6, but I just set 1.3V in the bios and 3195MHz primed fine for a few minutes. I have no idea how to test for motherboard issues either, when it boots fine. I did reset CMOS and started from a fresh bios.
So... :( :down: :shakes: :shrug: :confused:
What happened?!?
Edit: By the way, the cores fail randomly. It's not like Core 1 or Core 2 are the two that constantly fail. Every single one 1/2/3/4 has been the first to fail at some point.