Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

LOSTCIRCUITS	SHORTCUTS: Dual Core vs HyperThreading Tech Specs Test Setup Memory Performance WorldBench5 3dsmax CineBench 2003 3D Gaming Power/Temperature Coefficient Power Hogs 3D Rendering Power DOOM3 3DMark2001SE 3DMark'05 Final Thoughts Give Us Some Feedback on this Review
AMD Athlon64 X2 "Toledo" Dual Core on Single Die
(Review by MS May 9, 2005)

Power

As mentioned in our earlier report, it is fairly straightforward to isolate the power that is drawn by the CPU as opposed to the entier system. The dual core CPU makes this approach especially rewarding in that the power consumption of the processor itself changes with the number of resources that are busy at any given time. In other words, just on the basis of CPU power consumption is it possible to get some insight into CPU utilization and the threaded nature or lack thereof of the software running.

Idle

The first mile stone in our measurements was WIndowsXP sitting idle a few minutes after booting in to the OS without launching any applications. Since the X2-4800+ (Toledo) is based on dual "Diego" cores, we expected the power consumption approximately 2.4 x of what we monitored in the case of the Venice core. The reason for the extra 40% power draw is the increased L2 cache size that, as shown in the differential between Newcastle and Clawhammer adds some 21% power overhead to each core.

With 24 W under idle conditions, the power consumption of the X2-4800+ is exactly where we expected to see it. The Venice core rating may have been slightly underestimated since at lower power draw the systematic read errors are relatively greater than under load.

Temperature / ICC Relation

As ICs warm up, they draw more power. We established the temperature / power relationship by reading the power consumption under a defined load (BIOS HW Monitor) while recording the die temperature. The CPU fan was disconnected to avoid reaching a plateau, especially with the Winchester and Venice cores.

Clawhammer vs. Newcastle vs. Winchester vs. Venice vs. Toledo (X2) core. With increasing die temperature, the power needed to keep the CPU running or performing a defined workload will increase in an almost linear relation. The Toledo core behaves in a rather civilized manner here compared to the older 130 nm process-based CPUs but keep in mind that the actual load is probably either distributed over the two physical CPUs or else toggled between them.

next page: => Power Consumption: Power Hogs: Prime95, SiSoft Sandra =>