Front: What are the three ideal properties of a Random Number Generator (RNG)?

Back: Repeatability, long cycle length, and high speed.


Front: Why is repeatability important in an RNG?

Back: It allows generating the same sequence of numbers using a seed, which is critical for testing and debugging.


Front: What is the significance of a long cycle length in an RNG?

Back: It ensures the sequence remains random for a large number of values before repeating or becoming predictable.


Front: Why would a true RNG be unsuitable for scenarios requiring reproducibility?

Back: True RNGs lack a seed, making them inherently non-repeatable due to reliance on unpredictable physical processes.


Front: Name two physical processes used to generate true random numbers.

Back: Quantum phenomena or radioactive decay.


Front: What statistical property do true RNGs exhibit within an interval?

Back: Uniform distribution (equal count of each number in the interval).


Front: List three benefits of computational (pseudo) RNGs over true RNGs.

Back: Repeatability, portability across systems, and approximation of statistical properties like uniformity and independence.


Front: What is a major drawback of many pre-installed PRNGs in computer systems?

Back: Short cycle lengths and strong correlation patterns in generated sequences.


Front: What is the recurrence relation for a Linear Congruential Generator (LCG)?

Back: , where are integers.


Front: How does a mixed congruential method differ from a multiplicative congruential method?

Back: Mixed uses , while multiplicative sets .


Front: What parameters does Numerical Recipes recommend for an LCG?

Back: , , .


Front: What is the maximum possible period of an LCG sequence?

Back: At most (the modulus value), but it can be shorter depending on constants and seed.


Front: Why are PRNGs considered “random enough” for statistical applications?

Back: They closely approximate uniformity and independence in generated sequences.


Front: What key trade-off exists for PRNGs like LCGs?

Back: Speed vs. cycle length and correlation issues (fast but shorter cycles).