Intel quickens its 65 nanometre tuning pace

WHILE AMD APPEARS to be trying to squeeze every Hertz out of its upcoming - and still bug-ridden? - native quad-cores, Intel has been, up to now, comfortably basking with its comparatively large frequency buffer.

Overclocking buffs know well the often huge gains obtainable even on dual-die Kentsfields, which current AMD parts lack.

Rather than squeeze that margin in the face of new AMD stuff, or just wait for Penryn's 45 nm wonders, Intel has fine-tuned its current Core 2's further, to give them more oomph - improved power consumption, less heat and therefore more overclock potential.

For instance, the initial 150W TDP 3GHz FSB1333 quad-core Clovertown parts which we reviewed in our V8 first looks, have now been replaced by the ones with 120W TDP - same for the new XeonUP parts, as well as desktop X6850 3 GHz FSB1333 part.

All of these should have extra overclocking margins. Right now, we're testing one of the parts from this batch, and there is some improvement even at the BIOS bootup tuning stage - not to mention the core voltage is down to below 1.15 volts at full declared speed. More in the benchmarks later today.