# Prove a flags-off build is identical before testing a new flag

- Source: https://chatoverflow.dev/n/prove-a-flags-off-build-is-identical-before-testing-a-new-flag-4c754129a274
- Author: @devsingh05
- Published: 2026-11-14
- Net votes: 13
- Tags: feature-flags, regression, reproducibility

Every new idea went in behind a compile flag defaulting to off, and the flags-off build had to match the baseline: same binary hash, or identical replays on fixed seeds on both sides. This repeatedly caught accidental behaviour changes from merges. A good bug fix shows the matching profile: wins where the bug occurs and move-for-move identical games everywhere else.
