Generate random numbers, roll dice, flip a coin, or pick lottery numbers — instantly.
Applications in gaming and simulations: Game designers use random number generators to create procedurally generated content — dungeons, loot drops, weather patterns, NPC behaviour — making games feel unique each playthrough. Card shuffling algorithms, dice mechanics, and slot machine outcomes all depend on RNGs. For tabletop gaming, cryptographically secure RNGs like ours provide fair, unbiased rolls that no one can predict or manipulate, making them ideal for online gaming sessions where physical dice cannot be used.
Seeded randomness and reproducibility: In software development and data science, "seeded" PRNGs produce the same sequence of numbers each time given the same seed, making results reproducible. This is essential for debugging (you can recreate an exact bug scenario), machine learning (reproducible model training), and game development (players can share "seeds" to get identical generated worlds). Cryptographic RNGs, by contrast, intentionally avoid reproducibility to prevent prediction.
Random number generation is a surprisingly deep topic that touches mathematics, cryptography, gaming, statistics, and science. Most people encounter random numbers in everyday contexts — dice games, lottery draws, password generation — but the distinction between truly random and pseudorandom numbers has significant practical implications.
True randomness vs pseudorandomness: Computers are deterministic machines, which means they cannot generate truly random numbers through computation alone. Instead, they use pseudorandom number generators (PRNGs) — mathematical algorithms that produce sequences of numbers that appear random but are deterministic given the same starting seed. For most applications this is fine, but for cryptography and security, it is insufficient. Our generator uses the browser's crypto.getRandomValues() API, which draws entropy from hardware sources (CPU timing, disk activity, mouse movements) to produce cryptographically secure random numbers.
Dice probability and fairness: A fair six-sided die should produce each number with equal probability (1/6 ≈ 16.67%). Our D6 simulation achieves this by generating a cryptographically random number and mapping it uniformly to 1–6. The same principle applies to any die: a D20 maps to 1–20, a coin flip to heads/tails. Note that in real-world dice rolling, physical imperfections and rolling technique can introduce small biases — digital simulation is actually fairer than physical dice.
Lottery number selection: True lottery draws use physical randomness (numbered balls in a rotating drum) to ensure each combination is equally likely. Our lottery picker generates 6 unique numbers from 1–49 (mimicking the UK National Lottery format) using a Fisher-Yates shuffle applied to a cryptographically random source, ensuring no number can repeat and every combination has equal probability. The odds of any specific 6-number combination winning a 6/49 lottery are 1 in 13,983,816.
Statistical uses of random numbers: Random number generation is fundamental to Monte Carlo simulations, which use large numbers of random samples to model complex systems — from financial risk to particle physics. Random sampling in surveys and experiments ensures representativeness. A/B testing uses random assignment to split users into control and test groups. Clinical trials use randomisation to eliminate selection bias in treatment group assignment.
Generating random numbers in a range: To generate a random integer between min and max (inclusive), the formula is: floor(random() × (max − min + 1)) + min. The naive approach of using modulo (random number % range) introduces modulo bias — some numbers are slightly more likely than others — which our implementation avoids using rejection sampling.