(Which number generator has different results given the same input data?

A true random number generator (TRNG) produces outputs derived from nondeterministic physical processes (e.g., thermal noise, oscillator jitter, radioactive decay, or other hardware entropy sources). Because the underlying phenomenon is not algorithmically determined by an input seed in the same way as a PRNG, repeated “inputs” (or identical conditions from a software perspective) do not yield the same sequence; the outputs vary unpredictably. By contrast, a pseudorandom number generator (PRNG) is deterministic: given the same seed and internal state, it produces the same output sequence, which is useful for repeatability but means security depends on seed secrecy and proper seeding. “Prime” is not a generator type, and “sequence” is too generic and does not imply nondeterminism. In cryptographic systems, TRNGs (or hardware entropy sources) are often used to seed cryptographically secure PRNGs (CSPRNGs), combining high-quality entropy with efficient generation. Therefore, the generator that can produce different results for the “same input data” is a true random number generator.