Test Results

We rated these hand-held computers on the basis of their usability, features, performance, technology, implementation, and price, all on a scale of five stars--except for price, which we factored in at 10 percent for the Windows CE machines and 5 percent for the proprietary machines. Because the CE machines are essentially commodity items, price becomes a key determining factor. The features set and usability factor higher for the proprietary machines, as people buy them for their specific features.

We derived the performance rating by averaging the results of our speed tests. First, we measured each system's throughput during a file transfer from the hand-held computer to a laptop. The second test consisted of a download from the laptop to the hand-held.

To judge usability we graded the unit's characteristics in a few key areas. We evaluated the device's ease of use when transferring data, the screen size, and the backlighting. We judged their keyboard functionality, system labeling, and hardware documentation. We also looked for a reset button on the systems, and we gauged the size and portability of the A/C adapter. A hand-held earned extra points if it came bundled with applications.

To evaluate features, we looked at the processor speed, the screen, the data entry options, and the amount of memory that's included. We paid close attention to systems with existing upgrade paths. Because many users of hand-held PCs are mobile workers, communications options such as a modem, an infrared port, and a docking station are crucial. Sound, power supply, and battery life were also important in our features scores.

Test Methodology

For our performance tests we timed a serial download of 300 files. These included 100 schedules and 200 contact lists, which were 28.2- KB .SCD files. We transferred the files to and from a Dell Latitude XPI CD with 32 MB of RAM and a 166-MHz processor.

Our performance testing consisted of three downloads from the laptop to the hand-held, as well as three uploads from the hand-held to the laptop. After completing three file downloads, and before starting the uploads, we reset the system by popping out both the spare batteries and the primary batteries and replacing them. We also used the reset button to assure proper flushing of the static memory. We did this to make sure that the data bank was empty, putting the machine in the same state for each iteration of the test.

After each upload or download, we replaced a new, uncorrupted file with the transferred file. Additionally, we did all our performance tests with the A/C adapter, and not the battery, powering the unit because battery levels c an affect the results. All cables and A/C adapters used in our tests came with the particular system being tested.

---

Evaluations in this report represent the judgment of BYTE editors, based on tests conducted by NSTL, Inc., as documented in a recent issue of NSTL's monthly PC Digest. To purchase a copy of the full report, contact NSTL at 625 Ridge Pike, Conshohocken, PA 19428; (610) 941-9600; fax (610) 941-9950; e-mail: editors@nstl.com