Red light 660nm vs near-IR 850nm for skin: dose is the part I wasn't logging

quick number for context: I bought a panel in March and logged 47 sessions before I realized I was tracking the wrong variable. The 660nm vs 850nm split is the part people argue about, and it’s real. 660 (red) gets absorbed shallower, so it’s the wavelength people cite for collagen and skin. 850 (near-IR) penetrates deeper, more of a muscle/joint argument. Fine. but that distinction doesn’t settle the thing that actually determines whether you got a useful dose: irradiance times time times distance. I was logging “10 min, evening.” I was not logging that I’d moved the panel from 6 inches to about 18 because the closer distance felt too hot. Irradiance falls off fast with distance. So two sessions I logged identically were probably delivering very different joules. The LED mask vs panel question collapses into the same gap. Masks sit on your face (short distance, low power), panels run higher power from further away. Comparing them by “minutes used” tells you almost nothing. no personal collagen n=1 to offer here, spreadsheet brain just noticed the logging hole. anyone actually tracking distance as a field?