Endings

Introduction

Endings are the most searched and least documented part of Say Cheese!. The official description presents the outcome as a choice between destroying a peaceful society and trying to change fate toward hope, and then hands the decision to the player. That is enough to support one claim: more than one direction is intended. It is not enough to support a list, a count, or a set of trigger conditions, and as of 2026-10-31 the creator has published none of those. This section therefore does something different from a typical endings page. It explains how to reason about outcomes in a game whose ending logic is undocumented, how to tell when a run is committing to a direction, and how to test whether a decision actually mattered. It also explains how to read community coverage without absorbing its assumptions, because almost every video on this game is titled around endings and almost none of them cite a developer statement.

What Is Confirmed About Outcomes

Outcome Reasoning

Core Mechanics

The mechanics of an undocumented ending system are mostly mechanics of observation. When you play, watch for the moment the game stops describing the world and starts asking you to decide something; that transition is the part of the run worth remembering. Immediately after the decision, note whether the crowd's behaviour changed, whether the framing of the screen changed, and whether the game gave you any explicit feedback. On a second run, change exactly one decision and repeat the same three observations. A real branch produces a different observation in at least one of them; a coincidence produces a different feeling but the same observable events. This sounds slow, and it is, but it is the only method that produces knowledge rather than belief. The second mechanic is language discipline: separate what the game showed you, what you inferred, and what someone else told you. Most arguments about endings collapse because those three get mixed together within a day of playing.

Advantages

Working this way has one large practical benefit: you stop needing to be right about a number. A player who insists the game has a specific count of outcomes has to defend that claim every time someone disagrees, and neither side can point to a source. A player who says the direction is confirmed while the count is not can simply show the official page and end the argument. The method also makes replays cheaper. If you know which single decision you changed, one extra run answers a real question; if you change three things at once, you have spent a full session and learned nothing about which one mattered. Finally, keeping your own log makes you independent of the guide economy around a twelve-day-old game. Videos published in the first fortnight were made under time pressure, and the log in your notes will still be accurate months later.

Challenges

The obvious weakness is that this section cannot give you the answer you probably came for. If you want a list of named outcomes with conditions, no honest page can provide one right now, and any page that does is repeating unverified material. A second limitation is that single-variable testing assumes the game behaves consistently, which is an assumption rather than a documented fact; if the experience is partly random, a changed outcome may reflect chance rather than a decision. The method also costs time: two careful runs take longer than watching a ten-minute video. And because the game is new, the creator may publish an official structure at any point, which would make parts of this reasoning obsolete. When that happens, this page will be rewritten from the primary source rather than patched with community claims.

Frequently Asked Questions

Quick Tips