How We Tested

We tested the performance of the 16 removable-media drives with NSTL's InterMark benchmarks. The tests measure the drives' throughput, CPU utilization, and sequential and random read/write capabilities. To round out the evaluation, our product testers check the drives for ease of use and features.

We ranked the 5-1/4-inch optical drives in three categories: best overall, best high performance, and best low cost. The best-overall selection is based on a formula of performance scores (70 percent), as well as ease of use (20 percent) and features (10 percent) ratings. The best high-performance drive has the highest performance numbers, without any consi deration for usability or features. The best low-cost drive incorporates the best-overall formula of performance, usability, and features scores, but there's a $2300 pr ice cutoff.

Test Methodology

To test the optical drives, we used an Adaptec AHA2940W PCI SCSI adapter to attach the drives to a Windows 95-based workstation with an S.A.G. 133-MHz Pentium CPU, 32 MB of RAM, a SCSI-based 2-GB hard drive, and a SCSI-based CD-ROM drive. Six of the drives came bundled with a SCSI adapter. We tested these drives with both the standard Adaptec adapter and the one that shipped with the product. For our evaluation, however, we used just the test scores obtained using the standard adapter.

We connected the optical drives externally to the system via a SCSI adapter. The system's hard drive and CD-ROM also connected internally to the same SCSI adapter. Windows 95 recognized the SCSI hard drive, external optical drive, and the CD-ROM, assigning drive letters at boot time; the OS noted the external optical drive as a removable-media device in the "My Computer" listing.

Testers formatted the optical drives with one large partition bef ore starting performance testing. Nine of the drives arrived with a software utility to format the media and create partitions. In these cases, we used the vendor-supplied software to format and create partitions on the media. We used Windows 95 to format all those drives that did not come with any software tools.

Ease Of Use And Features

To evaluate ease of use, we made sure that the vendor-supplied documentation had clear installation procedures, intricate hardware specifications, and a well-organized table of contents and index to guide the reader through the manual. Testers also checked to see if status indicators and buttons were easily recognizable and considered how easy it was to set up and configure the units. The features scores reflect the range of supported software and hardware platforms, reliability specifications, SCSI drive controller specifications, and other options.

---

Contributors

C handrika Mysore, Project Manager/NSTL

John McDonough, Technical Editor/NSTL

Susan Colwell, Technical Editor/BYTE

The Lab Report is an ongoing collaborative project between BYTE magazine and National Software Testing Laboratories