[
direct link to this post]
Over Labor Day weekend I ran a series of experiments compiling the same bit of video over and over checking out different CPU speeds for my duals. The results of these experiments are attached below. The data from these experiments show that the for this test, once a minimum CPU speed of 1.5 or 1.6 GHz was achieved, the compilation time flattened out, showing that data input/output had become the limiting factor.
PL_dual_speed_temp_comparison.pdf ( 32.62k )
The entire preceding and ongoing discussion can be found here: Dual Processors and Bus bottleneck. It is suggested that you read it for full context and other updates and info.
