Labs

GPS devices

We tested a wide variety of GPS-based devices in this Labs, with an incredibly broad range of features and capabilities, so coming up with meaningful tests was a challenge. Car satnav systems are particularly complex, with most having a mind-boggling list of capabilities, while the recreational devices have a wide variety of specialisations.

Performance

We went back to basics with the in-car systems and concentrated on three primary areas: GPS speed and accuracy, route calculation and on-the-road navigation performance.

The last thing you want when you jump in your car is to have to hang around waiting for a satellite fix, so we timed how long each device took to lock on to a satellite signal when initially switched on. All the devices were connected to a car power outlet - sometimes satnavs take longer to lock on to a signal if they're running on battery power alone.

We then used each device to plan a 20-mile circular route, designed to incorporate most types of road you're likely to encounter in the UK: urban streets, country lanes, a stretch of motorway and a major A-road. This test resulted in some subjective scoring, reflecting several factors, including (but not restricted to) speed and ease of trip planning, simplicity of address entry, positional accuracy, screen-update speed, voice instructions and route choices.

For the recreational GPS devices, we devised a similar set of tests. You don't want to be standing around on the street waiting minutes on end just to record your position, so as with the car GPS systems we timed how quickly each one locked on to enough satellites to provide a position. We gave extra points for devices that give some indication of progress towards this goal.

We used each recreational device on at least one 24-mile round-trip cycling commute from north east London. This route takes in mostly open terrain, but finishes among the tall buildings of London's Fitzrovia district, where many such devices struggle to maintain accurate position. We also took the hiking GPS devices on a walk in Epping Forest - Greater London's largest green space. It's not overly rugged terrain, but with plenty of trees hindering the devices' line of sight it's a challenge to hold and maintain an accurate position.

Throughout, we took note of positional accuracy (where applicable), the clarity, readability and flexibility of the screen, how comfortable the system was to wear and use, and build quality - essential for a device that's likely to take a beating in the outdoors.

Features

Car satnavs are complex beasts, and some have piles of extras - such as photo viewers and media players - that are likely to see limited use. As a result, we awarded points only for core features - those that will have genuine day-to-day usefulness. We assume most people will use a dedicated player or their mobile phone to listen to MP3 files, for instance.

But as well as core features, such as hands-free phone compatibility, traffic information and speed-camera information, we reward innovation in the navigation engine. Devices with text-to-speech capabilities score extra points, as do those with lane-assist systems to help with complicated motorway junction negotiation.

For the recreational devices, we assess each in the context of its respective sub-group to avoid skewing the results. So we rewarded fitness watches for their flexibility and their ability to set multiple heart-rate zones, for example. And, more importantly, we assessed the power and ease of use of their PC-based training analysis and planning tools. The hiking devices get more points for better quality screens and route-planning abilities, and the quality and availability of on-board mapping.

Value for Money

The Value for Money score is based on a weighted average of the Performance and Features scores; we then factor in the cost of each device (including delivery) for a bang-per-buck result.

Overall

The Overall rating is a straight average of the Performance, Features and Value for Money scores. Occasionally, it may appear higher or lower than expected due to rounding.