Related Articles

Overclocking

You saw the results in the charts... we reached a maximum stable clock speed of 3.32GHz - a 25% increase over the stock 2.66GHz, and a respectable 9% increase over the X6800 Extreme Edition.

Basically there were three factors to the successful overclock - four if you count the updated BIOS.

1. Figuring out the fastest stable memory speed
2. Figuring out the fastest stable memory timing
3. Setting the FSB speed and multiplier
4. Getting the right Vcore

You can see that CPUZ was reporting the processor running at merely 1.296V at 3320.1MHz, with a 415MHz FSB - which is equivalent to a 1660MHz FSB data rate as the bus is quad data rate.

That's a whopping 13.3GB of potential bandwidth!

In order to run at 3.32GHz we:

• used the excellent pre-release GA-965P-DQ6 motherboard
• set the Vcore to 1.3375 (which the BIOS promptly ignored)
• left the DDR voltage at default (otherwise computer would not boot)
• set the DDR2 timing to 4-4-4-12 @ 1037MHz (2.5 memory multiplier)
• set the processor multiplier to 8
• set the FSB speed to 415MHz (yes, that high)
• used the Nautilus watercooler

Frankly, I suspect we could have gone higher with a more tuned BIOS; with the current one we could not get the processor to work with Vcore higher than 1.40V - which is strange, because the first BIOS (which had MUCH poorer memory performance) we got with the board was able to POST with 1.56V.

We used the Nautilus water cooler, so there was plenty of temperature headroom - but don't worry, this is not the end of our Conroe overclocking experimentation.

I found it very impressive that Conroe would run at 3.32GHz with a mere 1.375Vcore.

You can see that the memory is running as DDR2 1037.6 - or PC8300 - with 4-4-4-12 timings.

What really impressed me however was being able to run the FSB at 1660MHz. Actually, I was able to run it at 1700MHz, but the system was not stable at that speed. This pre-release BIOS was not stable if the PCIe speed was not left at "Auto", so I am not sure what speed PCIe was running at.. but I am certain it was WAY over spec, as there was the occasional boot sector corruption during overclocking. I shudder to think that I may have been running the PCIe bus at 207MHz, more than twice its rated speed.

The current BIOS gave a lot of control over DDR2 timing, however it did NOT like it when DDR2 or FSB voltages were increased.