Gnarlodious spake thus: > > and a faster processor adds lots of heat. >Huh? >Just last week someone said the newer electronics are more efficient than >older electronics, so power consumption is not significantly higher. > >My understanding of physics is that power=heat, so what's wrong with this >picture? > >If we could just get the numbers on the processor upgrades. . . At least two things affect how much power a CPU chip uses. For a given process size (typical transistor size), heat dissipation goes something like the square of the clock speed. As process size gets smaller, though, the amount of heat dissipated is less for identical clock speeds, so generally (Pentium 4's are a glaring exception) CPU's with smaller process sizes run cooler. The 7457 (fabbed at 130 nm), which is what PowerLogix (at least) is offering in its Cube CPU upgrade kits, dissipates < 20 Watt at 1.4 GHz, while the 7455 (180 nm process size) would dissipate ~ 3 times as much heat at the same speed, if it can be forced that high. The original CPU in the Cube was a 7400 or 7410. If the 7410 could have been run anywhere near 1.4 GHz, it would dissipate ~ 40 W. As an added improvement, the 745x's are capable of four instructions per clock cycle vs. three for the 7410; if Apple takes advantage of that, many things could seem even faster. ....or such is my understanding of the CPU power dissipation bidness. Joe Gurman -- "I love deadlines. I love the whooshing sound they make as they go by." - Douglas Adams, 1952 - 2001 Joseph B. Gurman, NASA Goddard Space Flight Center, Solar Physics Branch, Greenbelt MD 20771 USA