Trying to figure out what Voltages i need to run on this setup, i seen a lot with Q6600 but not 6700, i overclocked to 3.36 from 2.66, 800mhz memory didnt change the volt because i dont know what numbers are good for stable. i played SC2 for about 3 hours ran perfect checked speed fan the hottest the Cores got was about 60 to 65, then tried Battlefield 2 played for 20 mins pc crashed.. do i need to change to voltages? if so to what numbers? running (2) evga GTX 460 768mb SC SLI, 4gb ram windows 7 6 bit 750w PSU, thanks
If you're going to overclock you should read about it first. One of the things that is always pointed out is that you need to check your stability when you make changes. You don't just try an OC, say "hey it booted!" and start to game.
Here's what you need: HWMonitor, Core Temp (or Real Temp or Speed Fan), Prime95, Intel Burn Test (or LinX or OCCT), and Memtest86+. CPUZ is good too just to check current settings.
Run Intel Burn Test (IBT) on High stress for 5 runs. If it passes, you're pretty stable. If it doesn pass, you're not stable and to get it stable, you need more voltage. For CPU OCing, the main voltages to adjust are Vcore and VTT/IMC/QPI (interchangeable names, it'll have one of them). Start by raising Vcore 0.01V or so at a time. After ~0.1V increase, do the same to VTT. Generally VTT is lower than Vcore. Once you pass 5 IBT runs, do it again with 10 runs. If it passes that, you'll want to run Prime95 large FFT for about 8 hours.
The other half of OCing is that it affects your RAM too. If you increased the FSB then that increased your RAM speed, meaning you need to increase the timings too for it to be properly stable. Prime95 blend is a fairly quick way to determine if the RAM is ok. IBT can also have issues when the RAM is acting funny. Make sure your RAM is running at it's rated voltage. Memtest86+ is the best way to determine RAM stability but it can take a while to run the test, and requires a boot disk.
The other thing is, running the super clock GTX 460 might have unstable OCs. Obviously if it's from the manufacturer then it's their fault and they should replace it for you. An unstable GPU OC can cause a PC crash. You can test GPU OCs with FurMark, but make sure you run GPUZ along with it to monitor the temps (under Sensors tab you have GPU core temps and VRM temps). If the temps get too high after a few mins and it crashes, then it's a bad OC... also if doing that turn the fan up.