What is an entropy source? Source, RNG and DRBG
An entropy source is the physical component that produces unpredictability. An RNG is the complete construction that turns that into usable output. A DRBG is the deterministic algorithm inside it that expands a seed. The three are routinely conflated, and untangling them makes the SP 800-90 series readable.
Three things, three jobs
- The entropy source is physical. It produces raw output whose unpredictability comes from a process in the world, and it is the only part that creates entropy rather than moving it around.
- The DRBG is deterministic. Given a seed it produces a long stream, efficiently and reproducibly. It cannot create unpredictability; it can only preserve and expand what the seed carried.
- The RNG is the whole assembly: source, conditioning, DRBG, health testing and the policy connecting them.
Why the conflation is expensive
Marketing material describes a product as "a random number generator" and leaves the reader to guess which part was evaluated. A certificate might cover the DRBG mechanism, the entropy source, or the whole module — three very different statements, and the difference decides whether the certificate answers your question.
When someone says "our RNG is certified", the follow-up is always: certified to what, covering which component? Verifying a source you did not build is built around that question.
How the SP 800-90 series divides the problem
The structure maps directly onto the three terms. SP 800-90A specifies DRBG mechanisms. SP 800-90B covers entropy sources, their validation and their health testing. SP 800-90C describes how the two compose into a complete construction.
Read as three documents about one thing, the series is confusing. Read as one document per component, it is straightforward. What 90B validation actually proves and how 90C composes them cover the two that matter most here.
Where the entropy actually enters
Only at the source. Everything downstream is bookkeeping — good bookkeeping, carefully specified, but bookkeeping. If the source produces less entropy than the design assumed, no conditioning function and no DRBG mechanism will notice or compensate. That asymmetry is why this pillar concentrates on the physical end.
Keep reading
- Entropy fundamentals
How much entropy does cryptography actually need?
Entropy demand is calculable: DRBG instances multiplied by seed size divided by reseed interval, plus whatever consumes entropy directly, plus headroom for bursts. For most systems the answer is far below what a hardware source produces, which is exactly why sizing honestly is more useful than quoting a peak rate.
- Entropy fundamentals
Entropy conditioning, pools and reseeding
Conditioning compresses raw entropy-source output so the result carries close to full entropy per bit, usually with a hash or block-cipher construction. Reseeding policy then decides how often that material refreshes the DRBG — an architecture decision that governs how long a state compromise stays exploitable.
- Entropy fundamentals
Entropy vs randomness, a distinction that matters
Randomness describes how output looks; entropy measures how much unpredictability is actually present. A stream can be perfectly random-looking and contain no entropy at all — which is precisely what a good pseudorandom generator produces, and why statistical testing cannot establish security.