Athlon FX-60 Review - 3
Power Consumption: BIOS
10: Manchester 2000
9: Winchester 2200
8: Venice 2400
7: Toledo 2400
6: SanDiego 2800
5: FX60
4: FX60 (2860)
3: Clawhammer 2200
2: Newcastle 2400
1: Clawhammer 2400
We measured the dependency of the power consumption on the core temperature sitting in the Hardware monitor of the CMOS setup utility. The CPU fan was disconnected in order to reach the higher temperatures. As described earlier, increasing die temperature causes increasing power comsumption under the same conditions.
Power Consumption: WindowsXP Idle
Two surprises right here, first, the FX60 does not draw more power than the X2-4800+. Second, unlike the Venice that, when overclocked, showed a substantial increase in power consumption, the FX60 is almost completely agnostic to clock speed ( e.g. 2860 MHz) - as long as there is no increase in core voltage.
Power Consumption: Prime95
Along the same lines, under full load, the FX60 actually draws less power than the X4800+. Keep in mind, though, that we are comparing two processors with different manufacturing dates and the lower power consumption may reflect the fine tuning of the silicon mixture in the Dresden Fab. Nevertheless, the numbers are very impressive.
CaligariTrueSpace 5.1
Adam Trachtenberg's "Vases" scene is still the most reliable benchmark we know of for measuring multicore performance since the runtime reflects both multicore and HyperThreading configurations. Shown above is a screenshot taken with the Pentium4 840 ExtremeEdition, showing all four logical cores in action in the form of individual scan lines
|
