One rounding, not many: why a provably-fair verifier must match the server bit for bit
Originally published on the Betkyo Journal , where every figure is read from the game engine's source. Every provably fair round on this site turns a 256-bit hash into a number between 0 and 1 and then into an outcome. The server does that in Kotlin; the browser verifier does it in JavaScript. Both languages hold numbers as 64-bit doubles with 53 bits of precision , and a hash has more bits than…
A provably-fair verifier must match the server bit for bit to ensure accurate outcomes. This is because a hash, consisting of thirty-two bytes, needs to be converted into a number between 0 and 1. The server handles this conversion in Kotlin, while the browser verifier performs it in JavaScript. Both languages use 64-bit doubles with 53 bits of precision, which means that the hash, having more bits than that, requires some rounding during the conversion process.
If the server rounds once and the verifier rounds multiple times, the resulting numbers may differ by the last bit, leading to potential discrepancies in game outcomes. To prevent this, the engine constructs the number as two exact pieces added together in a single operation, dividing the first four bytes by 2³² and the next bytes by 2⁵⁶ or 2⁶⁴.
This ensures that the only rounding occurs once, matching the server's conversion method. The verifier, like the server, uses the same arithmetic to compute the result, making a rejection truly reflective of the game's fairness. By avoiding multiple rounding operations, the provably-fair verifier helps maintain trust in the integrity of the game's outcome.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
