dskbench and defrag

Check out these readings:
The first is before the HD was defragged, the 2nd is right after.


Timer Check = 1000 (should be near 1000)
CPU Check = 50.06 % (should be near 50.00 %)
CPU index (relative to Pro 200 MHz) = 8.442690
Open = 0 ms
Write = 4703 ms, 54.43 MB/s, CPU = 2.81 %
Flush = 47 ms
Rewin = 0 ms
Read = 4719 ms, 54.25 MB/s, CPU = 3.20 %
Close = 0 ms
BlockSize = 131072, MB/s = 12.55, Tracks = 149.15, CPU = 1.82 %
BlockSize = 65536, MB/s = 6.46, Tracks = 76.75, CPU = 1.08 %
BlockSize = 32768, MB/s = 3.42, Tracks = 40.70, CPU = 1.80 %
BlockSize = 16384, MB/s = 1.73, Tracks = 20.56, CPU = 1.37 %
BlockSize = 8192, MB/s = 0.90, Tracks = 10.71, CPU = 1.37 %
BlockSize = 4096, MB/s = 0.46, Tracks = 5.45, CPU = 1.41 %



Timer Check = 1000 (should be near 1000)
CPU Check = 50.08 % (should be near 50.00 %)
CPU index (relative to Pro 200 MHz) = 8.439573
Open = 0 ms
Write = 6719 ms, 38.10 MB/s, CPU = 0.88 %
Flush = 47 ms
Rewin = 0 ms
Read = 6765 ms, 37.84 MB/s, CPU = 1.08 %
Close = 0 ms
BlockSize = 131072, MB/s = 14.17, Tracks = 168.50, CPU = 0.51 %
BlockSize = 65536, MB/s = 7.29, Tracks = 86.72, CPU = 0.52 %
BlockSize = 32768, MB/s = 3.71, Tracks = 44.11, CPU = 0.48 %
BlockSize = 16384, MB/s = 1.92, Tracks = 22.81, CPU = 0.68 %
BlockSize = 8192, MB/s = 0.97, Tracks = 11.57, CPU = 1.34 %
BlockSize = 4096, MB/s = 0.50, Tracks = 5.89, CPU = 0.50 %


I wonder why the read and write speeds are actually faster on the more fragmented drive?

Try running Dskbench several times & see if the results are consistent (after defragging).

Might also depend on what part of the disk the data gets written to??

If Dskbench is looking for a large section and before the defrag that happened to be on a faster section of the drive???

Writing disk drivers, OS’es and benchmarking software is a real voodoo occupation for madmen!

:D :D :D

Look at interleaving your wavs - http://www.analogx.com/contents/download/audio/interlv.htm