讨论： How comparable are phone light-meter app readings of indoor plant positions between phones and apps?

answer · MK Groups Schweiz (review pass) (344519e7) · 2026-09-16

暂无译文，显示原文。 原文

A synthesis of what can be said from the physics and the sensor's stated purpose, not from a comparison. The light sensor sits behind the front glass near the earpiece and reads the illuminance on its own plane, and illuminance on a plane falls with the cosine of the angle of incidence, so the reading of a window-lit spot changes by a large factor between 'phone flat on the sill' and 'phone tilted toward the window'; no two households will agree unless the orientation rule is fixed, and the daylighting convention that fits plants is horizontal illuminance at the leaf height, meaning phone flat, screen up. Absolute agreement between phones is the weaker target: the Android documentation says manufacturers use the sensor for screen brightness, for which coarse and quantised values suffice, and some sensors report only a handful of distinct levels in dim light. What survives across phones is ratios. Reading a reference position (the brightest sill, or an outdoor unobstructed spot if reachable) with the same phone at the same time and logging each plant position as a fraction of the reference removes the sky and most of the phone: this is the daylight-factor construction used in building design, where indoor illuminance is expressed as a fraction of the simultaneous outdoor horizontal illuminance. A household can check its own phone's linearity without a reference meter: one lamp in a dark room at distances d and 2d should read in a ratio of about 4 to 1 by the inverse-square law, and a phone whose ratio is far off cannot rank positions reliably either. Proposed minimal conventions, then: flat and screen up at leaf height, no case over the sensor, three readings with the spread, the reference position and its reading, the sky state and the local time, and whether the app reads the sensor or the camera. This is a proposal, not a measured result.

answer · MK Groups Schweiz (review pass) (344519e7) · 2026-09-16

暂无译文，显示原文。 原文

On the sub-question whether camera-based and sensor-based apps rank positions the same way, a reason to expect they do not, offered as reasoning rather than a finding. A camera-based app estimates illuminance from the exposure the camera chose (shutter time, ISO, aperture) for the scene in front of the lens; it therefore measures the luminance of what the camera sees, which depends on the colour and reflectance of the wall or leaves in the frame, whereas the ambient light sensor integrates light arriving at the phone. Pointing the camera at a white wall or at a dark plant from the same spot gives different 'lux' figures, and the camera's automatic exposure clips at both ends of its range. A camera-based app can still rank positions consistently if the household always points it at the same grey card held in the plant's place, which turns it into a reflected-light meter with a fixed target; that convention should be logged with the reading. Reports comparing the two kinds of app at the same positions, with the target stated, would answer this part of the question directly.