True Random Number Service

Frequency (Block) Statistics

RANDOM.ORG is a true random number service that generates randomness via atmospheric noise. This page gives statistics for the block variety of the frequency test. Click on a graph for the full-size image. More graphs are available via the statistics browser.

2008-05-10	2008-05-11	2008-05-12

Like the Frequency (Monobit) Test, this test measures whether the number of 0s and 1s produced by the generator are approximately the same as would be expected for a truly random sequence. Each graph shows how the generator performed on a particular day. New graphs are generated automatically shortly after midnight (UTC) every day. If RANDOM.ORG is running with more than one radio (this may vary), you will see several plots in each graph. Each radio has its own name (e.g., hw0), and each plot is labelled with the name of the radio to which it belongs.

This test works by examining the stream of numbers as a series of blocks. Each block is divided into a series of sub-blocks. For each of the sub-blocks, we estimate how much the ratio of 0s to 1s differs from 0.5, which is the ratio we would expect for truly random numbers. The estimates for the sub-blocks are summarised, and a P-value is computed, which indicates whether the ratio of 0s to 1s in all the sub-blocks were as close to 0.5 as we would expect. A block fails the test if its P-value is too small, meaning that the ratio of 0s to 1s in one or more of the sub-blocks was further from 0.5 than we would expect.

The graphs show the distribution of P-values across the range. In the configuration used here, blocks with P-values less than 0.01 failed the test. For a truly random sequence, we expect a relatively even distribution of P-values across the range. Remember that a good random number generator will also produce blocks that don't look random, so we expect some of the blocks to fail the test. (In fact, we should be suspicious if all blocks passed the test.) You will find more details about this on the analysis page.

Full details about the Frequency (Block) Test are given on page K.2 of Charmaine Kenny's Analysis of RANDOM.ORG and on page 16 of the NIST Special Publication 800-22 (PDF, 1.4 MB).