|
|
|
|
|
The Data Analysis section reports the results of several different randomness tests. Populate this section by clicking the Analyze button.
The Sample Size is the number of bytes analyzed for the test.
The Average is the average value of all random bytes analyzed. This value will tend to 127.5 in large samples of truly random data.
The Entropy value represents the information content in the data. This value will tend towards 8 bits per byte in large samples of truly random data.
The Bit Averages show the average value of each of the 8 bits in all the bytes of data. These values will tend to .500 in large samples of truly random data.
The Serial Correlation Coefficient measures the extent to which one byte of data is correlated to the next. This value will tend towards zero correlation in large samples of truly random data.
The Chi-square test is a common statistical test for randomness. The p-Value of the Chi-square test statistic gives the percent likelihood that the observed data was generated by a truly random process (note: p-Value is scaled by 100). If you consistently get values less than 1 or greater than 99 for the p-Value, it means that there is less than a 1% chance the software is generating random data. In this case, you should try changing your audio settings to get more random data. High quality random data will show p-Values between 10 and 90 about 80% of the time.
