Test Benches and Scoring System
A test bench is a set of tests that we perform on the products we test. To come up with the set of tests that will be included in a test bench, a rigorous process is followed, during which many factors affect the final decision. Among the most important factors for including a test is whether it answers a few important questions regarding the true capabilities of the product, whether it will answer questions about what most people care about, and how accurately and consistently can it measure a specific feature.
Test benches include some usage performance indicators we call Ratings. Ratings aim at estimating the performance of a product during a specific use. To construct a Rating, we first choose which of the tests matter for the performance of the product during the specific use. Then we assign a weight to its importance. The value of the weights assigned on each test for each Rating is based on our experience regarding the test and its relevance to the specific usage. The numerical value of each Rating is assigned simply by computing the weighted average.
Test benches, ratings, and individual tests are updated on an as-needed basis. Once a test bench update is finalized, the tests included in this version and the way their respective score is calculated remains unchanged until the next test bench update is released. Each update tries to encapsulate technological advances/changes that break our current tests, or the introduction of new methodologies that result in more accurate/consistent measurements.
How to interpret the scores
10 to 7.5
This product is great for this usage or test. You will be satisfied with it. When we discuss tests or products that fall into this category, we usually use adjectives like good, very good, great, impressive, excellent, remarkable, outstanding, superb, amazing, etc.
7.4 to 6.0
It is good enough for most people, but if you care about this usage or test, it will bother you. When we discuss tests or products that fall into this category we usually use adjectives like mediocre, middling, passable, adequate, not bad, unremarkable, decent, okay, alright, acceptable, satisfactory, reasonable, fair, ordinary, etc. We also use sub-par or sub-standard when the score is at the lower end of the scale.
5.9 to 0.0
This product is below average at this usage or test. It could be an issue even if you aren't particularly picky about this aspect. When we discuss tests or products that fall into this category, we usually use adjectives like disappointing, inadequate, poor, inferior, bad, terrible, very bad, awful, etc.
Additional information
Hover over any usage ratings in a review or recommendation page to see the components. Changes in scores due to firmware updates are noted in a yellow banner at the top of a review, and any changes to overall weightings or scoring are listed below. The test bench version each product was tested in is listed on the review, next to the ratings.