ManYue: Record visual-riddle constraints, candidates and revealed answers

Visual riddles often distribute quantities, positions and written conditions across panels. Guessing the ending can still conceal a missed condition. Rather than solving a particular published work, use a self-made, low-stakes shape puzzle to record what the question states, what is visible, which conditions candidates satisfy and what the reveal explains. ManYue provides reading and page location; the reader performs reasoning and note-taking manually.

Make a small puzzle whose conditions can be checked

Draw three boxes A, B and C. A has a circle and star, B only a star, and C only a circle. State: find the box with both a circle and a star. Add a separate panel revealing A. Number question, object and reveal panels, import them as an ordinary comic and keep the original. This trains conjunction only, with no probability, hidden mechanism or real competition. For a published work, record edition, chapter, positions and question boundaries. Establish which panels belong to the puzzle and which are decoration or later explanation; cover promotion is not a puzzle condition.

Copy written conditions individually

Before looking at the reveal, record every qualifier separately, including both, only, at least, exactly, cannot and scope words. Preserve both in the practice question; it must not become find a box with a pattern. If the question limits the choice to three boxes, an outside ornament cannot become a fourth candidate. Attach page and panel references to each condition, distinguishing quotation from paraphrase. Mark an unread qualifier as unread instead of filling it from familiar puzzle conventions. Complete conditions matter more than an immediate answer, especially preserving prohibitions rather than rewriting them as steps to perform first.

Separate visible shapes from assumed rules

On the second pass, record visible shapes only: A has circle and star, B star, C circle. Do not label correct or incorrect yet. Note visibility limits for occlusion, blur or off-panel areas. Without supporting conditions, do not assume every box contains only one object, colours encode a secret or symmetry and numerical coincidences are rules. A published puzzle may use evidence across pages, but include it only where the question permits it or context clearly connects it. Identifying each rule’s source prevents repeatedly adding restrictions the author never supplied merely to protect a guess.

Check every condition for every candidate

Create a manual candidate table, checking has circle and has star separately for each box, then checking their conjunction. A satisfies both, B lacks a circle and C lacks a star, so A meets the known question. B is not wholly irrelevant: it satisfies one condition but not both, as does C in the other direction. For a real puzzle with several conditions, retain satisfied, not satisfied and unknown states; unknown is not automatically satisfied. If several candidates pass, acknowledge that known conditions do not uniquely determine an answer, then check for omissions rather than choosing the one that looks most like the protagonist.

Write a possible countercheck for each candidate

After selecting a candidate, actively check boundaries: is the circle actually outside the box, is the star obscured and do both marks belong to the same box? In an external copy of the practice drawing, you may move A’s star outside the frame to illustrate how it stops satisfying the rule, but do not alter the original being cited. This tests your rule; it is not a changed work used to accuse the original puzzle of error. If object ownership is unclear, change the candidate to unknown and explain why. The revealed answer cannot retrospectively prove every ambiguous shape must belong where you guessed.

Compare explanations after the reveal

Open the reveal and separately record the stated answer and the reasons actually supplied. In the practice example, A agrees with the two-condition check. If a published work merely announces an answer, record that no explanation is supplied rather than presenting your reconstruction as the author’s explanation. If you guessed correctly using a wrong rule, retain that error instead of calling the reasoning wholly correct. If you missed an unread condition, record where it was found. Separate information newly supplied by the reveal from information originally available; later facts cannot make an earlier deduction inevitable.

Troubleshoot reading without assuming automatic solving

If question and reveal disagree, inspect the whole page for missed small print, cross-page panels or a mirrored scan, and confirm the same puzzle and edition. Zoom helps identify visible content; it cannot restore cropped conditions. For genuine ambiguity, state two feasible readings and their conditions rather than treating insufficient evidence as a reader-app defect. Do not search for an invented automatic-solver or recognize-all-objects entry, or overwrite your observations with an online answer. Unresolved is acceptable. A complete account of scope serves this exercise better than manufacturing certainty.

Leave and return to audit the condition chain

Leave the reader and return to the question before the reveal. Recheck every qualifier, all three boxes and mark ownership from the notes, then open the reveal and compare explanations. Keep five parts in the final record: original conditions, visible facts, candidate checks, counterchecks and reveal. Distinguish your first guess from later revisions. When discussing it, identify the spoiler scope and respect sharing permissions. Completion means another person can repeat the checks from your evidence instead of seeing an isolated answer. There is no automatic solution, invented hidden rule or confusion between guessing and proving.

Related guides and sources

ManYue branching comics: manually log nodes, choices and return routes; ManYue fictional board scenes: check positions against stated rules.

Checked on November 7, 2026: Apple public version information lists ManYue 2.0.4, requiring iOS / iPadOS 15.0 or later. Feature scope follows the official App Store description and the installed version. This article provides reading and checking methods you can perform; it does not claim step-by-step physical-device or private-server testing. Manual records, external tools and actually visible controls are identified separately, and personal analysis or code observations are not presented as verified automatic capabilities.