Intel CPUs performance increasing.

[M A V I C]

[quote Filliam H. Muffman]If you underclock a processor, it runs cooler. If you overclock a processor, it runs hotter.
It depends on the processor. If you build a processor using 130nm it runs hotter than if you build it at 45nm.

A 2.7GHz G5 sure as heck runs WAAAAAAY hotter than then P4's even did at that speed. Out of all of the factors that play into a processor's temp, clock speed is very distant in importance.

[Filliam H. Muffman]

[quote M A V I C]Out of all of the factors that play into a processor's temp, clock speed is very distant in importance.
Is that really what you wanted to say? Why does Intel use SpeedStep to cut the frequency if not to lower power consumption, and cut the frequency if the internal diode temp gets too high? There are many sites with essentially identical data that shows power and temp go up with frequency, and test how good a CPU cooler is by the highest overclock it can reach.
http://www.anandtech.com/casecoolingpsus...spx?i=3068&p=6
http://www.tomshardware.com/2006/07/14/c...page3.html
The Core 2 Duo is better than the P4 in how power scales with frequency, but I do not see how you can refute the general phenomenon. Is there some other explanation for why it happens? I would guess that the Intel X6800 at 3.0 GHz uses 25% more power than the E6320 at 1.86 GHz, same CPU stepping, same amount of cache, and same voltage.