Bandicoot Lab
Every device score on Bandicoot Lab is built from real measurements — no subjective ratings. Here’s exactly what we test, how we test it, and how each metric feeds into the scores you see on the site.
Almost all scores are designed to be improved on. That’s to say, very few scores have a ceiling. Some do — if a display is perfectly color-accurate, then it’s perfectly color-accurate. Others have a theoretical limit, like charging speed, in which the theoretical best is 0 minutes — but that’s not happening any time soon. Most scores should get higher and higher over time. Scores were originally calibrated to a benchmark of around 950–1000. Sub-scores roll up into category scores via weighted averages, and category scores roll up into the overall score the same way.
The Overall Score is a weighted composite of all category scores, designed to give a single number that represents the total device experience. Each category is weighted by its real-world importance to daily use.
While our tests are comprehensive, there are some missing pieces. At this time, for example, we’re unable to test cellular performance and cellular drain. This is due to the complexity of accurately measuring different networks, limiting variables like unknown cell tower upgrades, and so on. As time and resources allow, we’d like to add these tests to our methodology.
Scoring formulas are periodically refined as testing methodologies improve and new devices push the boundaries of existing scales. When formulas change, all devices are rescored simultaneously to maintain fair comparisons. Multi-lens devices (foldables, multi-telephoto setups) have adaptive scoring that accounts for the number of cameras and their capabilities.
Think our methodology has a problem? Think it could be better? Reach out! We’d love to hear from you.