There's more to these numbers than simple %'s show and there's no "defrag" setting that fits all sizes. The full version of Diskeeper has 2 settings that cover both topics: defrag method & collect performance data.
I'd expect a drive with low file fragmentation but many directories all over the drive would produce lower performance when reading back multiple files (whole drive copy speed). While any few files of a large size & high fragmentation would do the same but only when reading just them since their speed increase/loss would be lost in the overall performance benchmark reading the entire drive.
Keep on benching Thane, looks like at least you haven't found zero-gain or negative effects from defragging!
joeuser wrote:
Agreed.
Thane Sherrington wrote:
Two more machines:
6/14% fragmention - 12% speed increase 27%/54% fragmentation - 15% speed increase.
On the first machine, Windows defrag said it didn't need defragging, on the second it did. Given the similarity of speed increases, I think Windows defrag doesn't know much.
T
--- [This E-mail scanned for viruses by Declude Anti-Virus]
