Stake Enginestakeengine

Why Use the Math SDK?

Building slot games usually means working through dense math to balance payouts, hit rates, and player engagement. That work takes time and resources. The Carrot Math SDK removes much of that load by offering:

  • Ready Frameworks: Begin from customizable templates or sample games to move faster.
  • Mathematical Precision: Simulate and tune win distributions with discrete outcome probabilities so you stay in control of the mechanics.
  • Direct Integration: Outputs are shaped for the Carrot RGS so you can move to production quickly.
  • Scalability: Built-in multithreading and optimization tools keep large simulations efficient.

Who This Is For

The Carrot Math SDK suits developers who want to:

  • Build custom slot games with original mechanics.
  • Tune payouts and hit rates without heavy manual calculation.
  • Produce detailed simulation output for statistical review.
  • Publish on stakeengine.org with as little friction as possible.

Static File Outputs

Physical slot machines (and many iGaming titles) produce results in real time by running game logic on the RGS or backend. When a round is requested, a cryptographically secure random number generator picks a reel-stop for every active reel, and play continues from that board. The downside is scale: a single reel strip can hold 100+ symbols, and with five reels that is 100^5 (10 billion) unique boards. Calculating payouts or Return to Player (RTP) exactly is often impractical, so large simulations are used to estimate results. Stake Engine requires every game outcome to be known at publish time. Storing instructions for all possible outcomes is not realistic. A subset of results is used to define the game instead.

Those outputs split into two parts: 1. game logic files and 2. CSV payout summaries. The game-logic files hold an ordered list of key details such as symbol names, board positions, payout amounts, and winning symbol positions. Each simulation in those files has a matching CSV row with the simulation number, selection probability, and payout. When a round is requested, the RGS uses the CSV or lookup table to pick a simulation number, then returns a JSON payload from the game-logic file for that number so the web client knows what to render, while the player's wallet is updated with the payout. Splitting the two files also lets us calculate RTP and core win-distribution stats exactly at publish time.

Get Started

Open the technical notes and see how the Carrot Math SDK can change how you build games. With strong tools, sample titles, and detailed docs, you have what you need to ship engaging, mathematically sound games.

See Math SDK technical details for more.