The test is severely broken. Not only do the results lack repeatability, but my single 160GB drive is rating consistently about double the speed of my 4x200GB RAID10 array, especially in linear read. (This tool says my RAID linear reads are 30MB/s, other test tools tell me the linear read is throttled to 152MB/s by the mobo RAID controller). My favorite result from this test is from running it on a CD -- 0.6856MB/s linear reads, and 0.8124MB/s on random reads; - think about it, how can it be faster on random access than linear?