Great, and what exactly is the improvement seen with tri-gate transistors as an independent variable? Without that it's hard to judge what the significance is f

Great, and what exactly is the improvement seen with tri-gate transistors as an independent variable? Without that it's hard to judge what the significance is for tri-gates. I understand Ars can't test a 32nm processor with tri-gates but that doesn't make this analysis any less unscientific. Ars is not alone in this dishonesty and I think even Intel's official statement was just as confusing.