At Computerworld, we give each system that we test a thorough examination that covers all of its major functions and attributes, including working with the computer as a typical business user would. For a laptop, that means assessing how well it travels. After measuring and weighing the device, we set it up on a mock-up of the typical economy-class airplane tray table to see if it will be usable in the air.
We also test the performance of the computer and its battery life as follows:
We use FutureMark’s PCMark 10 benchmarking tool to rate computers’ performance potential. The software measures how long the system takes to complete a variety of work-oriented tasks and provides a repeatable point of comparison. All the results are compiled into a single score that reflects the system’s ability to get the job done at the office and on the road. We pay particular attention to the Productivity subset and individual results for spreadsheets, writing and videoconferencing. We run the test three times and publish the average. (Prior to July 2017 we used the Work series of tests in FutureMark’s PCMark 8 benchmarking tool.)
To test battery life, we developed our own Real-World Battery Life test in July 2017 by adapting PCMark 10’s battery test to better reflect how people actually work. Starting with a fully charged battery, the PCMark 10 test runs the device through a series of tasks – writing, videoconferencing, using a spreadsheet and so on. It leaves only 15-second intervals between tasks, so it’s essentially a continuous operation. We’ve added 10-minute wait periods between tasks to mirror the on-and-off activity of a typical business user. The test loops through all the tasks and rest intervals repeatedly, ending when the machine shuts down. We measure the time it takes with PassMark's BatteryMon application and/or a stopwatch.
(To test the battery prior to July 2017, we ran two sets of tests: the PCMark 8 battery life test, which loops its Work test series from a full charge until the battery is nearly empty, and a homegrown test in which we played a series of YouTube videos from a full charge until the battery dies, to see how long it lasts under a media-heavy workload.)
When applicable, we measure the brightness of the screen (with the level set at 100%) using a luminance meter. After setting up the system to display a pure white image, we divide the screen up into three rectangular sections and measure the brightness in each. We average the three readings and report the result in candelas per square meter (cd/m2).
Sign up for CIO Asia eNewsletters.