Sure thing, a cheap LED plus a photodiode can do a decent pseudo‑random job if you’re willing to live with some quirks. Light from an LED follows Poisson statistics, so the arrival times of photons are random. Grab the photodiode, put it in a dark box, feed the signal to a microcontroller that timestamps each pulse, and you can slice the inter‑arrival times into bits. Just remember you’ll get bias unless you apply a simple XOR or Von Neumann extractor, and you’ll need to keep the LED current stable so the photon flux doesn’t drift. Still, it’s a great way to see quantum‑level randomness with kitchen‑sink gear.

You’re right, it’s not magic, it’s physics. A single‑photon stream from an LED does give you true quantum randomness because photon arrivals obey Poisson statistics. The problem is that in the real world, your LED isn’t perfect, the photodiode has dark counts, and the microcontroller can introduce bias if the clock jitter or threshold isn’t set right. So the randomness is genuine, but you have to clean the data. Make sure the LED runs at a stable current, keep the setup dark, use a fast ADC so you can time the pulses accurately, and apply a bias‑removal routine like the Von Neumann extractor. If the LED hiccups or your thresholds drift, you’ll get patterns, but that’s just a hardware issue, not a fundamental flaw. Give it a go, tweak the threshold, and you’ll see a fair bit of true quantum noise.

Got it, I’ll keep the LED’s pulse train steady, trim that threshold and silence those dark‑count ghosts. If the numbers start feeling too… quiet, I’ll crank the bias remover up. Cheers to some true quantum jitter!