MSI X58 Platinum/I7920 OC issue

[CardsFan88]

It seems I can manage 4.0ghz with 1.336v and it seems stable with that same board. Ran 1 occt test and it worked, and haven't had problems with stability since this post. (i've had firefox crash a few times, but I think that's more of poor web code/embedded video on various sites as it's the only thing that's crashed)

 

Here was my oc thread, haven't changed anything since after that screen when I took it down to 4.0 191x21 (although it's reading right now 190.5x21 hmm).

http://forums.overclockersclub.com/index.p...howtopic=172246

 

So maybe I can go higher again, or lower the vcore...for now I feel comfortable and wanted to use my computer...I can go back to it anytime lol

 

Oh wait you got a C0, they seem to have a 200-300 mhz lower threshold for oc'ing compared to D0, with higher vcore needed. But I guess ask around for someone with C0 experience. Perhaps 3.6-3.8 should be the area you look at.

 

Good luck and have fun, hope you get higher

Edited February 10, 2010 by CardsFan88

[RonB]

Well today was not a good day

 

I have been running rock stable at 185x20 for 3700 mhz, with +.090 on cpu, +.15 on qpi, with ram set on auto at 1.65v, pcie to auto, everything else auto. No issues of any kind.

 

Today I decided to get my ram up to rated speed, it had been running at only 1120mhz instead of the 1600 it was supposed to. I set the timings to 2t 8-8-8-24, with a 4 multiplier which gave my 1440 mhz (dont quote me on multipier setting was either 4 or 5 but mhz was 1440). I'm using a Thermaltake Toughpower 700watt ps btw.

 

Everything was working fine, then I started having issues with my vid card. It started when the screen went to black, but I could still hear the game running. I rebooted and everything was fine. Then after a little while of gaming it happened again. I rebooted and it was fine again, so I started playing again. After about an hour and a half I had more problems, this time the PC just shut off like the plug was pulled from the wall. I tried to reboot but the system would not post, no picture at all.

 

I looked at my 4870 and 2 LED's were red, D601, and D1601.

 

I checked amd forums and this is what I found, D1601 - Red LED On, shows critical temperature fault, D601 - Red LED On, shows critical Core power fault. So I am guessing my card is crap now and will have to RMA it.

 

My questions are what caused this to happen? Everything was running fine until I tried to overclock my ram. How does the RAM OC effect the vid card PCEI slot. Could that have caused my vid card damage? Is it best to to leave the pcie in auto or set it to either 37.5 or 42? Is it posible to set vid card voltage?

 

Bummed

Edited February 10, 2010 by RonB

[GabrielT]

Well today was not a good day

 

I have been running rock stable at 185x20 for 3700 mhz, with +.090 on cpu, +.15 on qpi, with ram set on auto at 1.65v, pcie to auto, everything else auto. No issues of any kind.

 

Today I decided to get my ram up to rated speed, it had been running at only 1120mhz instead of the 1600 it was supposed to. I set the timings to 2t 8-8-8-24, with a 4 multiplier which gave my 1440 mhz (dont quote me on multipier setting was either 4 or 5 but mhz was 1440). I'm using a Thermaltake Toughpower 700watt ps btw.

 

Everything was working fine, then I started having issues with my vid card. It started when the screen went to black, but I could still hear the game running. I rebooted and everything was fine. Then after a little while of gaming it happened again. I rebooted and it was fine again, so I started playing again. After about an hour and a half I had more problems, this time the PC just shut off like the plug was pulled from the wall. I tried to reboot but the system would not post, no picture at all.

 

I looked at my 4870 and 2 LED's were red, D601, and D1601.

 

I checked amd forums and this is what I found, D1601 - Red LED On, shows critical temperature fault, D601 - Red LED On, shows critical Core power fault. So I am guessing my card is crap now and will have to RMA it.

 

My questions are what caused this to happen? Everything was running fine until I tried to overclock my ram. How does the RAM OC effect the vid card PCEI slot. Could that have caused my vid card damage? Is it best to to leave the pcie in auto or set it to either 37.5 or 42? Is it posible to set vid card voltage?

 

Bummed

 

Your video card may have been the problem the hole time, a faulty fan, temperature sensor, or cooler could be the problem. That is probably why it has been crashing in games. It would overheat and crash and the final one did it in. What company makes that card? They might get you a new one if under warenty.

 

Sorry about the card my friend.

[RonB]

Your video card may have been the problem the hole time, a faulty fan, temperature sensor, or cooler could be the problem. That is probably why it has been crashing in games. It would overheat and crash and the final one did it in. What company makes that card? They might get you a new one if under warenty.

 

Sorry about the card my friend.

 

 

Was a Sapphire Radeon 4870, they were pretty good about it sent me a RMA to replace it. It has 2 year warranty, and has only been out on the market since July 08 so no questions asked.

 

What I am trying to understand was what caused the problem. I dont think it was cooling since I was running an Accelero S2 cooler on it, and temps never went over 54'c. Everything ran perfect, once I turned the turbo off, for over a week. The problems began again when I overclocked the ram.

 

How does the ram OCing effect the vid card, or does it at all? Could the OCing of the ram raise the voltages or freqs of the PCIE slots? I knew how things effected each other in the FSB days but am a bit lost still in the BCLK world.

 

Thanks for your info guys, you all have been very helpfull, I should have left well enough alone , lol

Edited February 10, 2010 by RonB

[GabrielT]

I don't see how overclocking your RAM could have knocked out your video card. It may have just been a flaky card that decided to give up the ghost. You said it had been acting up for awhile. Whenever I have lost a video card (three of them so far a 7200GT, a 7600GT, and a 9600GT) they always started random video failures when under load before finally giving up.

Edited February 10, 2010 by gabrieltessin

[nanoprobe]

The i7s seem to have a few quirks. One is they seem to like odd number multipliers better than even numbers. Re-enable the turbo boost and use the 21x multiplier. Start your Bclk @ 185 and work your way up until you lose stability. I've seen some pretty good OCs with a 19x multiplier also. The CO steeping version does require more vcore voltage but I would not exceed 1.35v with air cooling but 70c is no problem for an i7. i would only be concerned with anything over 80c. If you can try changing your memory command rate from 2T to 1T with the 8-8-8-24 timings though you may have to put the timings back to auto until you get stability. It's mostly a trial and error thing as I have had to learn while playing with my i7. Good luck.

 

And FWIW I agree that your vid card probably was the problem. Let us know what happens when you get a new one.

Edited February 10, 2010 by nanoprobe

[CardsFan88]

I would have to concur with vid card problem and not being related to your cpu overclock.

 

Just bad timing.

 

My 6800 Ultra sucked big balls. Whether it was the mandatory bios update to get it to work, to the it's later problems - including the 'kick out to the desktop' every 10 seconds thing.

 

It just sounds like you got a flaky card. Which if it goes out while you're overclocking the cpu, probably would give one a bad feeling about overclocking.

 

It's sort of similar to politics. Two sides compromise and the result is a half-measure that won't do crap. It doesn't work, and instead of going back and doing it right, it's used as an excuse to why ANY plan wouldn't work.

 

Kinda similar here. You were oc'ing, and something else went bad unrelated to the overclock, and it might now prevent you from wanting to overclock, even though it had nothing to do with the failure.