Hard time deciding? Let nuclear 8 ball help you

thealch3m1st1 pts0 comments

radmon — RadiaCode-103

Count rate

Dose rate

rolling avg<br>raw<br>0.5s<br>1s<br>2s<br>4s

window<br>all<br>10s<br>30s<br>1m<br>5m<br>10m<br>30m<br>1h

cps/dose max<br>adaptive<br>peak hold<br>reset peak

value<br>raw<br>Δ delta

count rate — cps

dose rate — µSv/h

temperature — °C

cps↔dose corr —

GMC pulses —

log y

accumulated<br>time range

10m<br>1h<br>6h<br>24h

— cps live

— · — cps accumulated

2D<br>3D

⏸ streaming<br>window<br>5m<br>10m<br>30m<br>1h<br>3h

log y<br>reset view

♪ sound<br>step

——

click an energy to highlight it · drag up/down to scroll time · wheel to zoom

2D<br>3D<br>line<br>log

window<br>10s<br>30s<br>1m<br>5m<br>10m

span<br>30s<br>1m<br>5m<br>10m

update<br>0.5s<br>1s<br>5s

reset view

♪ sound<br>step

count-rate FFT · fs 1 Hz · 0–0.5 Hz

2D: drag up/down to scroll time · wheel to zoom (pick a window to return to live) · Nyquist 0.5 Hz at 1 Hz sampling (periods ≥ 2 s)

⏸ pause

candle<br>PDM signal

window<br>10s<br>30s<br>1m<br>5m<br>10m<br>30m<br>1h

height<br>count n<br>rate n/dt<br>demod τ

400 ms

DC level<br>AC fluctuation

each candle = one sampling window that detected ≥1 pulse · height = pulses in that ~6 ms window (taller = pile-up) · width ∝ sampling period · hover for time, count, dt & instantaneous rate · green=1, amber=2–3, red ≥4

the decay pulse train read as a pulse-density-modulated bitstream: each pulse is a "1", and the demodulated signal (low-pass filter, time constant τ ) recovers the instantaneous count-rate as a continuous waveform · faint stems = the raw per-bin density (the carrier) · DC = absolute rate, AC = fluctuation about the mean (Poisson shot noise) · readout shows live cps, mean & modulation depth σ/μ

↺ reset accumulation

accumulating…

RadiaCode —

GMC-800 —

RadiaCode ⊕ GMC (XOR) —

Spectral RNG waterfall —

Spectral RNG — all channels concatenated —

true hardware randomness from radioactive decay · RadiaCode : Von Neumann on consecutive per-second counts (c₁>c₂→1, c₁g₂→1, else 0 — unbiased for a Poisson process) · XOR : the two independent streams combined (whitening) · left = the live bit grid (TV-static), right = the power spectrum of the whole stream since accumulation began (Welch averaged, Hann-windowed; a good RNG is flat/white ≈ 1.0)<br>spectral waterfall : an independent RNG per RadiaCode energy channel (Von Neumann on each channel's de-accumulated per-spectrum count increments) · y = energy, x = bit index (newest at right), hue = energy · busy low-energy channels fill fast; sparse high-energy channels lag (blank = no bits yet)<br>all channels concatenated : every channel's bit buffer joined end-to-end (channel order, low→high energy) into one pooled stream — the full spectral entropy as a single bit grid + spectrum

Ask a yes/no question, then shake.

Shakeextractor<br>Spectral<br>Von Neumann<br>SHA-256

Bits are extracted from radioactive-decay count fluctuations — true hardware randomness.<br>Von Neumann pairs consecutive per-second counts (c₁>c₂→1, c₁waiting for live count samples…

count rate window energy time radiacode

Related Articles