Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more

A puzzle game can offer 7,304 board configurations without storing 7,304 separately authored puzzles. In a developer-published account of Meowdoku, Focss describes starting with 913 stored 6×6 puzzles and applying eight rotations and reflections to each. The result is a compact, deterministic way to present geometric variants while keeping each base puzzle’s solution available for validation. These are derived configurations—not 7,304 independently authored or independently verified puzzles.

How does a puzzle game reuse puzzles without showing the same board?

Meowdoku’s described system begins with a finite bank of 913 6×6 entries. Each entry stores a map of the board’s regions and a solution. Rather than generate a new puzzle during play or store every possible presentation separately, the game transforms a bank entry into a variant.

For a square board, the eight symmetries are four rotations, each optionally combined with a reflection. Applying those eight transforms to each of the 913 entries gives the reported total:

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

913 × 8 = 7,304 derived board configurations

This is a count of configurations produced from the bank, not a claim that every configuration feels novel to players. The developer’s account describes the implementation and its scale; the figures are not independently audited.

Why store the solution as one number per row?

In the game’s format, a solution places exactly one cat in each row. That makes it possible to encode a solved board compactly as a sequence containing the column index of the cat in each row, rather than recording every cell as occupied or empty.

A row-to-column sequence is convenient to store, but it is not itself a two-dimensional shape that can simply be rotated or reflected. The described procedure therefore expands the sequence into a boolean grid, applies the same geometric transform to the grid as to the region map, and then converts the transformed solution back into the compact row-to-column representation.

Transforming both the region map and the solution together matters: the answer must continue to match the board after its geometry changes.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

How are levels selected reproducibly?

The developer describes selecting a bank entry and a transform deterministically from the level number. The bank index cycles with the level, while the level number also determines which transform to apply. Given the same bank and selection rules, the server can reconstruct the same board from its level number instead of relying on the client to send a complete board state.

The server-side validator checks that a submitted solution does not repeat a cat’s row, column, or region. Because the server uses the same bank and rules, it can validate the answer against the intended puzzle.

Daily challenges

For the daily challenge, the account says the UTC date string is hashed into a bank-index and transform pair. That provides a shared selection rule: players worldwide using that implementation get the same puzzle for a given UTC date, without requiring a server-side schedule to assign each player a board.

What are the trade-offs against generating or authoring every puzzle?

The developer’s rationale is to move work away from the moment a player advances to the next level. The approaches differ in when solvability is checked, how much preparation is needed, and where computation falls.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Approach When solvability is checked Authoring or preparation Work during play Variation
Runtime generation with a solver As puzzles are generated, using a solver to find or check a solution. Requires generation and solving logic rather than a prepared bank of finished entries. The developer says backtracking would sit on the player’s next-level path; this is the author’s rationale, not a measured universal performance result. Can generate boards at runtime; the account does not quantify its possible variety.
Hand-author every board During creation and verification of each board. Every board must be made and checked; the developer says this does not scale well. Prepared boards avoid runtime generation work. Each authored board can be distinct, but the number available depends on authoring effort.
Precomputed bank with geometric variation During bank preparation and verification; the server also validates submitted solutions. Requires a bank of puzzles and solutions, then applies transforms to derive configurations. Selection and transformation replace runtime puzzle generation in the described flow. The reported 913 entries and eight symmetries yield 7,304 derived configurations.

The bank approach does not eliminate verification work; it concentrates it before play. Nor does a correct transform repair a faulty source entry. As the developer notes, a bad puzzle or solution in the bank can be reproduced across its geometric variants.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

What does the reported data footprint mean?

Focss reports approximately 67 MB of JSON across 34 files. The bank is split by grid size and variant, fetched on demand, and served with immutable caching. The client cache preloads only the next five levels. These are implementation details reported by the developer, not independently measured benchmarks, and they describe this game’s setup rather than a general requirement for puzzle games.

In practical terms, the design trades a prepared data bank for less dependence on runtime generation. Splitting files and fetching them as needed avoids requiring the client to preload the entire reported bank at once, while immutable caching suits content that does not change at a given URL.

What the 7,304 figure does—and does not—tell you

The arithmetic explains the headline, but it should not be mistaken for a count of wholly separate puzzle designs. Each group of eight comes from one bank entry under the described symmetries. Whether those variants feel meaningfully different to players is not established by the account.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Focss summarizes the intended balance in the developer article: “The bank stays small, and the player still sees a new board every level.” That is the author’s description of the design goal, rather than an independent finding about player experience.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.