The torch, the notebook, and the room you can’t enter
I stood at the edge of a dark room, torch in hand, and I wouldn’t step inside. I swung the beam from one doorway spot to another and wrote down what each sweep caught. Lots of little peeks, then one best guess of the whole room. Takeaway: many partial views can join into one picture.
The room didn’t play fair. A shiny thing flashed bright from one spot, then went dull when I shifted. Some corners stayed hidden behind furniture, so my notes disagreed with each other. If I tried to draw the room from only a few peeks, the sketch looked wrong the moment I moved.
The clever twist was to keep one invisible room rule instead of lots of separate bits. The rule answers two things anywhere the torch might pass: how much that place blocks light, and what colour it would send back towards where you’re standing. That way, a gleam can change with your angle without moving the object.
To make a new view, I pictured a torch-beam line going straight from my eyes into the room. Along that line, the rule gets checked again and again, and the checks get blended. Mapping: camera line equals torch beam, blocking equals how quickly the beam gets stopped, final pixel colour equals the beam’s gathered light. Takeaway: trace one line, collect what it meets.
Two small habits made the picture sharper. I started marking places in finer and finer ways, so tiny edges didn’t smear. I also stopped staring into empty air: a quick sweep found where something might be, then I slowed down only there. The torch time went where it mattered.
After enough adjusting to match the real photos, the room finally behaved. I could “stand” in a new spot and the view looked believable, even for shiny glints, because the blocking stayed tied to the room while the colour could shift with where I looked from. Before, I redrew for every angle. Now one room description carried them all, even if it took ages to build.