Dan, Find a large file, that is hopefully reasonably sequential on the disk.
It would not make a difference. In fact, the CPU utilization will probably be even lower thanks to two levels of disk cache (one on the hard drive itself, and one in Windows), and caches love sequential data.
But think about it. The existence of the disk cache in Windows should give you a clue as to why your theory is bunk. Don't tell me the compression program now has to figure out whether the data is in I/O or in memory, because that's all handled by the OS. All the compression program does is request data from the device driver. The device driver then loads the data (either using DMA or getting it out of the Windows disk cache), and THEN the program processes the data IN MEMORY.
Tenchu
Market Data 
Markets 
AI
