iGamerKnow the game.
18+ --:--:-- UTC

RNG & Fairness

Fairnessverified 2026-08-04next review 2028-08-03

How casino games decide outcomes — certified RNGs vs provably-fair — and why every round is independent.

Digital casino games determine outcomes with a random number generator (RNG). Most licensed games use a certified pseudo-RNG seeded from an unpredictable source and tested by an accredited lab; crypto casinos often add a 'provably fair' scheme that lets players verify each result cryptographically. In both models, every round is statistically independent — the game has no memory of past spins.

Certified RNG — the mainstream model

A pseudo-RNG produces a fast, uniform, unpredictable stream from a seed. Testing labs (GLI, eCOGRA, iTech Labs) verify statistical randomness against batteries such as NIST SP 800-22, confirm unpredictability, and check that observed RTP matches the declared model. The player trusts the licence and certification chain rather than checking each individual bet.

Provably fair — the crypto model

A server seed, a client seed and a nonce are hashed together. The server commits to a hashed server seed before play and reveals the seed afterwards, so the player can recompute and confirm the outcome was not altered mid-play. It proves the result matched the committed seeds; it does not prove the RTP, nor that the seeds were generated fairly in the first place.

Every round is independent — the gambler's fallacy

Because the RNG has no memory, a run of reds does not make black 'due', and a slot that has not paid is not 'hot' or 'cold'. Each outcome is drawn fresh from the same distribution. Believing otherwise is the gambler's fallacy — the single most exploited misconception in gambling, and the engine behind most losing 'systems'.

What certification does and does not promise

A certificate attests that the RNG and RTP behaved as declared on the day of testing. It does not guarantee your session, and it says nothing about which RTP variant the operator has deployed. Fair maths is necessary but not sufficient: the game can be provably fair and still carry a house edge that grinds down every player over time.

Formulas
Provably-fair outcomeoutcome = hash(serverSeed, clientSeed, nonce)server commits to hash(serverSeed) before play and reveals serverSeed after, for verification
Worked example

On a provably-fair dice game, before you bet the server shows SHA-256(serverSeed). You add a clientSeed; each bet increments a nonce. After you rotate seeds, the server reveals serverSeed — you re-hash and confirm every outcome matched. You have proven the results were not tampered with, though the 1% house edge still stands the whole time.

Key facts
MainstreamCertified pseudo-RNG, lab-tested (GLI / eCOGRA / iTech)
CryptoProvably fair: server + client seed + nonce, hash-committed
Key propertyEvery round independent — the RNG has no memory
Certification provesBehaviour on test day — not your session or the deployed RTP

Education, not advice. This explains how the game works so you can read it clearly — it is not a system to beat the house. Every casino game carries a house edge; over enough play it wins. 18+.