Short entries from the bench — written down before they evaporate, filtered by subject when they pile up.
Before measuring anything, I now sit in the hall for ten minutes with the house PA off. The rehearsal room upstairs has a low hum that settles somewhere around the D below the staff; it was always there, and hearing it changes how the chord voicings get spread. The measurement later only confirms what the ears found first. The rule I keep writing down: measure after you have listened, never instead of it.
A brass bell under tungsten wash is not simply bright; it is warm where the light is warm and cool where the reflections catch the sky, and the conical flare smears that gradient over a wide angle. Under LED pars the same bell looks like a different instrument — harder, more mirror than metal. Half of what an audience reads as the sound arriving seems, on inspection, to be the instrument’s face deciding what the light will do.
Recorded a temple fair bell with a borrowed converter set too low, and the strike came back wearing a slow wobbling ghost that was not in the tower. The inharmonic partials of a large bell sit well above anything a small rig can hold, and once they fold there is no arguing them back. Re-recorded properly at a sane rate. Kept the bad file; it is the clearest illustration of aliasing I own, and better than any diagram.
Pedal B♭ on the flugelhorn, about 233 Hz, is the note I use to learn a new space: long, quiet, held until the room commits to an opinion. Some rooms lift it, some swallow it, and the difference has nothing to do with the instrument. Writing this down because it is the one test that never fails to teach me something about a hall I thought I already knew, and because the number keeps turning up elsewhere on this site.
Two lamps labelled the same warm tone arrived from different suppliers and turned the stage into two different evenings. Since then I check every rig the same way: one white card, one photograph, no filters, compared against the house daylight reading by the door. It is not science, it is housekeeping. But the band hears the difference before anyone admits to seeing it, and the card settles arguments faster than ears do.
Spent an afternoon with notebooks from several years back, mostly unreadable, occasionally priceless. The entries that survive are the ones that record what surprised me, not what I measured; the numbers were always re-measurable, the surprise was not. New rule for this site and the paper ones beside it: lead with the thing that broke the expectation, keep the reading underneath, and write as if the handwriting will get worse.