Computational photography uses algorithms and processing power to improve image quality, rather than relying solely on lens size or sensor hardware. Your phone captures raw image data, then software enhances sharpness, exposure, colour, and detail in real time or post-capture.
Common computational photography features include:
- Night mode: Combines multiple exposures to brighten low-light shots without excessive noise
- Portrait mode: Uses depth data to blur backgrounds, mimicking the shallow depth of field from expensive lenses
- HDR (High Dynamic Range): Merges multiple exposures to preserve detail in both bright and dark areas
- Image stabilisation: Corrects camera shake through software, not just optical systems
- Super-resolution: Upscales or sharpens images using machine learning
Why it matters: computational photography levels the playing field. A flagship phone's computational power can match or exceed a dedicated camera's output, even with a smaller sensor. This is why recent budget phones take surprisingly good photos.
The catch: computational photography is processor-intensive and consumes battery quickly. Processing can also introduce artefacts, blur fine details, or produce overly processed-looking results if algorithms are poorly tuned. Results vary between manufacturers; Google's computational approach differs from Apple's or Samsung's.
When buying a phone, look beyond megapixel counts. Test photo samples in low light, check how night mode handles movement, and see how portrait mode handles fine details like hair. Read real reviews showing computational results, not just spec sheets.