Random Number Generator

A random number is a value chosen so that no one can predict it in advance, and so that every allowed value has the same chance of being picked. The simple generator at the top draws one whole number between the two limits, counting both ends, so the default range of 1 to 100 has exactly 100 equally likely outcomes. It is a quick way to settle who goes first, pick a raffle winner from numbered tickets, choose a random page in a book or roll a virtual die of any size.

The comprehensive version handles everything else. It accepts decimal limits, returns as many as 1,000 values at once, and can work with decimals of up to 999 digits by doing exact integer arithmetic instead of ordinary floating point. Turning off duplication makes it behave like drawing tickets from a hat without putting them back, which is what you want for lottery-style picks or for choosing a random sample of rows from a list.

Computers are deterministic machines, so they cannot invent randomness out of nothing. Most software uses a pseudo-random number generator (PRNG): a formula that starts from a seed value and produces a long sequence of numbers that pass statistical tests for randomness. Given the same seed, a PRNG repeats the same sequence exactly, which is useful for simulations that must be reproducible but unacceptable for passwords or encryption keys.

True random number generators instead measure an unpredictable physical process, such as electrical noise, the timing of radioactive decay or tiny variations in hardware. Modern operating systems mix such sources into a cryptographically secure generator. This page uses that secure source through the browser's Web Crypto interface and maps the raw bits to your range with rejection sampling, which removes the slight bias that the common shortcut of taking a remainder introduces.

Random numbers are used far beyond games of chance. Statisticians rely on them to choose survey respondents and to assign patients to treatment groups in clinical trials, so that the groups differ only by chance. Engineers and scientists run Monte Carlo simulations, repeating a model thousands of times with random inputs to estimate risks that have no neat formula. Software testers feed random data into programs to uncover rare bugs, and teachers use random picks to call on students fairly.

Frequently asked questions