Computers come in many shapes and sizes generally under the classification of either desktop or laptop models. These are further diversified across a number of sub-categories designed to appeal to a wide range of users.
- Desktops have several different sub-categories such as all-in-one models to mini-PCs and even smaller fit-in-your-pocket PC sticks.
- Laptops (also called notebooks) generally fall into two categories – the traditional 'clamshell' shape and the 2-in-1 convertible models that do double duty as a tablet. Tablets might also find use as laptops when equipped with a keyboard and possibly a mouse and maybe even a desktop dock.
Below you'll find a round-up of the approaches our experts take to test and assess different computers.
On this page:
Our expert testers
The home computer is often the hub of a whole range of technologies from storage to scanning, networking, display screens, accessories, lounge room media players and all kinds of software from operating systems to productivity programs, backup, cloud services and more.
Our expert computer testers have extensive experience in working with a wide range of consumer-level computer-related technologies, because it's not just about how the computer works but also how it fits into your home technology ecosystem.
How we choose what we test
More than any other category of household product, computer technology is constantly updating and evolving. The pace of changing technology means we constantly have to re-evaluate how we look at computers. Last year's top performers often find themselves at the level of this year's entry-level devices as much faster chipsets offer substantial performance boosts, while storage becomes cheaper and faster. Improvements come so thick and fast that some computer models have a shelf life of just a few short months before newer and better models hit the market.
So each time we test a particular range of computer products we start by researching the market to find current models to compare and we try to include models in each category from across all the major brands. This includes finding out which models we can test that will still be available on retail shelves by the time testing is completed and the results published for you.
We test in small batches and look to include the most up-to-date and popular models available in the big name retail outlets
While we try to establish a level playing field for testing across a category, the variation in range of options available can make this a real challenge. It's worth noting that there are often several models of a particular computer "family" available, having the same basic features but varying in main processor (CPU) speed, graphics processor (GPU), the amount of memory (RAM), and storage capacity and whether hard disk drive (HDD) or solid-state drive (SSD). In some tests we may buy a number of models from the same manufacturer that fit into different families.
Some laptop models may also be available in a range of colour options (with a slightly different model code for each colour). The models tested should give you a good indication of how a particular model family rates against current competitors, including the advantages and disadvantages of its overall design and any special features and inclusions.
We look to include the most up-to-date and popular models available in the big name retail outlets and buy them off the shelf or online just as any consumer would, so you can be sure that what we test is what you can buy and our results should be what you can expect.
How we test
Our testing of all computers includes comparison testing of performance and ease of use. We also include an expert evaluation of the display screen. Laptop and tablet testing also includes battery life testing.
- Performance testing includes benchmarking each model using a variety of software tools. These measure various aspects of CPU performance, memory performance and memory bandwidth, video performance, 3D graphics and storage speed. The benchmarking software is designed to measure performance under a variety of simulated workloads.
- Ease of use evaluation includes an expert assessment of connectors, design flaws and advantages, supplied security hardware and software, if any, along with the format and scope of any built-in help, support documentation and manuals, plus general design features and overall style. Where applicable we also assess the supplied keyboard and mouse/trackpad, stylus pen or other pointing device. All performance testing and ease of use evaluation is conducted under the computer's native operating system.
- Battery life testing for laptops starts with each battery conditioned to achieve optimum life, then tested under a heavy-usage scenario with the power management features set to maximum performance and screen brightness to 100%, with Wi-Fi turned on and connected to a network. Because some screens, particularly tablets, can be much brighter than typical clamshell-style laptops due to being designed for outdoor use (which affects battery life if set at the highest setting) our battery life score now takes into account calculations based on the luminance (brightness) of the screen, to create a more level playing field.
- Display evaluation takes into account colour, brightness/contrast, glare and surface reflections, horizontal/vertical angles of view with reference still images, plus Blu-ray video playback.
- Wi-Fi testing We connect each laptop exclusively to our test lab wireless router, then wirelessly transfer a set of test files across this closed network, timing this over several runs to establish consistency and averaging the final score in megabytes per second (MBps). All devices are restarted before testing and drivers updated to latest versions.
- Temperature testing for laptops. Whether on your lap or held in the hand, a device should be comfortable to use and this includes temperature. We measure the temperature of the hottest part of the underside of each laptop to see if it gets uncomfortably hot after sustained heavy usage on battery only. We consider 44 degrees the maximum acceptable comfort threshold. This test doesn't contribute to the overall score.
- Energy usage We measure power consumption in active use and on standby, calculating an annual cost based on an average use scenario applicable for that category, costed at current rates.
- Sound evaluation testing using the computer's inbuilt speakers is done by an in-house expert who evaluates the quality compared to similar devices.
- Tablet use assessment is for 2-in-1s and tablets with that can be used with a detachable keyboard. This testing takes into account how the device performs as a tablet (without keyboard) rather than as a laptop. In addition to the same CPU, graphics, storage, and Wi-Fi tests used for laptops we also consider form factor (45%), display screen (40%), and battery life (15%) from a tablet-user point of view. This score does not contribute to the overall score as non-convertible laptops are not included. Some devices may score better as tablets than laptops and vice-versa.
Our rating system
CHOICE applies the following interpretation to the scores achieved in our tests. When we describe a result as "excellent", "poor" etc, it usually relates directly to a numerical score in that range.
As laptop processing and storage performance increase, we modify our scoring method in order to best reflect the differences between the slowest and quickest models.
- 0–24 Very poor
- 25–45 Poor
- 46–54 Borderline
- 55–69 OK
- 70–79 Good
- 80–89 Very good
- 90–100 Excellent
We're on your side
In more than 60 years of making a difference for Australian consumers, we've never taken ads or sponsorship.
Instead we're funded by members who value expert reviews and independent product testing.
With no self-interest behind our advice, you don't just buy smarter, you get the answers that you need.
You know without hesitation what's safe for you and your family.
And you'll never be alone when something goes wrong or a business treats you unfairly.
Learn more about CHOICE membership today
Stock images: Getty, unless otherwise stated.
