Athlon FX-60 Review - 2
HyperThreading to MultiCores
Some Numbers
Test Configurations
To put things into perspective, we are using dual and single core configurations for comparison along with a dual processor Opteron system. The point is not to say that one is better than the other, rather it is to give some idea where a given system would currently be, and whether there is really any need for upgrading or not. Every processor and every system has its own strengths and weaknesses and depending on one's preferences, some of the issues will be irrelevant whereas others may influence a purchasing decision -- or not.
Software Configuration and Benchmark Results at One Glance
Software: 32-bit
Windows XP Corporate Edition
Service Pack 2
SiSoft Sandra 2005
Lightwave [8]
3dsmax 8
Cinebench 2003
DOOM3
FarCry (Patch 1.3)
Prime95
Caligari TrueSpace 5.1
Call of Duty 2
SPEC ViewPerf 8.1
Power Consumption Monitoring
We used the same power measurement setup as in our previous reports. Briefly, for isolated CPU power measurements, we used a Fluke 80i-410 AC/DC current probe in combination with a Wavetek Meterman 33XR multimeter to measure current through the isolated +12V supply lines feeding into the CPU VRM. To increase granularity of the measurements, we ran the supply lines in a triple loop through the clamp. The clamp itself was calibrated using a BK Precision model 1692, 30V 40 A DC power supply. Since there is a temperature dependency of the probe, we monitored the zero-current offset at the beginning of each measurement as well as at the end of each run. If the values drifted during we retook the measurements. Despite these precautions there are possible deviations of the read-out from the real current, however, these errors mostly affect the lower (processor idle) measurements. We estimate that the errors should not be more than 10% at the lower end of the data and less than 5% in the mid and higher data range. Moreover, since the same procedures were applied to all processors tested, there may be an offset in the absolute numbers, however, the relation of the individual cores to each other with respect to power consumption should be fairly accurate.
Bear in mind that the power measured is the input power going into the VRM rather than the regulated output power at the individual supply phases. However, the current VRMs are operating in the order of approximately 85-90% efficiency in the worst case. This means that the power measurements are slightly inflated across the board by approximately 10%.
High Level Benchmarks, SiSoft Sandra
Memory Bandwidth Benchmark
Given the fact that the memory controller is on-die, it is not overly surprising that the FX 60 performs better than its slower siblings.
CPU Multimedia Benchmark
One thing that is quite interesting is how much mileage the AMD processors get from the dual core design compared to the Intel CPUs. Note the FX57 performance compared to the FX60 performance and, on the other hand, the P4 670 performance compared to any of Intel's dual core offerings. |
