The program we used to measure hard disk drive performance, DiskSpeed32, reads all hard drive sectors, plotting a chart with all transfer rates obtained.

We included in our comparison another small external hard disk drive, Anypak, manufactured by USBNet, from South Korea. We also included on our comparison the performance obtained by other USB external hard disk drive enclosures (SPIO from ValuePlus, HardBox from Sarotech and 5000XT from Maxtor, which can also use Firewire port). We also included in our comparison the results from some hard disk drives connected to the standard ATA port, in order to see the difference in performance between an USB hard drive and a standard parallel IDE hard drive.

Normally the hard disk transfer rate varies according to the part of the disk that is being read. The disk transfer rate is higher at the disk's edge, lowering as it approaches its center. This occurs because of the zone bit recording (ZBR): in longer tracks (the ones away from the disk center) there are more sectors and more data is read at each disk spin. Because of that, the software used shows three results: maximum transfer rate (obtained on the first disk cylinders, i.e., on the tracks located near the disk edge), minimum transfer rate (obtained on the last disk cylinders, i.e. on the most inner tracks), and average transfer rate, which is the result that we are usually interested in knowing.

Because of this effect we can also explain the need of hard disk defragging and why professional disk defrag utilities such as Norton Speed Disk allow you to move the operating system files to the beginning of the hard disk. As we explained, data stored on the beginning of the disk are read at a higher transfer rate than data stored in other sectors.

On the graph below you see our benchmarking results (in KB/s) and our analysis.

X-Micro DisGo 20 GB obtained the worst performance in our comparison, however these numbers are not final because we are comparing apples to oranges. First, notebook hard disk drives spin at a lower rate, thus achieve a lower performance.

The product that really competes with DisGo is USBNet’s Anypak, since both use notebook hard disk drives. Anypak achieved a maximum transfer rate 36.28% higher and an average transfer rate 45.14% higher than DisGo from X-Micro. However, the Anypak model we tested was carrying an 80 GB 2.5" hard disk drive from Fujitsu, and a correct comparison would have to compare both models with a hard disk drive with the same size and capacity.

DisGo 20 GB compared with an old 20 GB hard disk drive from Quantum (Fireball LCT 15 QML20000LC-A, ATA-100 and 5,400 rpm) connected directly to the parallel IDE port on the motherboard was not that bad: Quantum 20 GB hard drive achieved an average transfer rate 15.38% higher and a maximum transfer rate 9.59% higher than X-Micro DisGo’s.