Moon Wards: how Neon Eclipse's doubling multiplier works

Neon Eclipse is live on Stake. Rei is a spirit hunter who cuts winning clusters apart in a neon night city. Each cleared cell leaves a Moon Ward behind. Repeated clears advance that ward along a doubling ladder, up to ×128. In Total Eclipse free spins, the wards stay for the whole feature.

We put the multiplier on the grid, so the position of a cluster matters alongside its symbols and size. Here is how we account for those positions, when the wards change, and where their growth stops.

A ward belongs to a cell

We use a 6×6 board. A paying cluster needs at least 5 identical symbols connected horizontally or vertically. Winning clusters pay and disappear together, the remaining symbols fall, and new symbols drop in. That sequence repeats until the board has no paying cluster.

The ward belongs to the cell where a symbol was cleared. It stays at that position as symbols move through the grid. Another clear at the same position advances it again, regardless of which symbol formed the cluster.

The ladder starts with an unmarked cell. Its first clear leaves a mark, its next clear advances it to ×2, then further clears take it through ×4, ×8, ×16, ×32, ×64 and ×128. A mark has no multiplier contribution yet.

For each cluster, we add the values of all wards of ×2 or higher on its cells, then multiply its paytable value by that sum. Marked cells add nothing. If there are no contributing wards, the cluster receives its ordinary paytable value. We add the wards together rather than multiplying them by one another.

The order matters: we read the wards before removing the cluster, then advance the cleared cells. A clear that creates a ×2 ward does not use that newly created multiplier for its own payout. It leaves the ward available for a later cluster at that position.

Total Eclipse keeps the grid

In the base game, we clear every ward at the start of each spin. Wards can develop through that spin's tumbles, but the next base spin starts with an empty ward grid.

Eclipse Seals trigger Total Eclipse from the initial board: 3 award 10 free spins, 4 award 12, 5 award 15, and 6 or more award 20. The base tumbles finish before the feature starts. A triggered Total Eclipse begins with no wards, so the wards earned on the triggering base spin do not carry across that boundary.

Once inside the feature, we keep the ward grid between free spins. Each spin starts with the wards left by the previous one. Symbols change, but the cells retain their marks and multiplier values. If 3 or more Eclipse Seals land during a free spin, they add 5 more free spins, with the same persistent ward grid.

That persistence changes the span of the mechanic. We track cleared positions across the feature instead of only within a single spin. The wards still need a paying cluster over their cells to contribute; a lit cell does nothing on its own.

Where doubling stops

A doubling ladder makes each later upgrade larger than the previous one. Keeping those values between free spins lets the same cell continue advancing across the feature. We therefore have to account for both repeated clears and clusters covering several developed wards.

The cell limit is ×128. Once a ward reaches it, another clear leaves it at ×128. This gives the ladder a fixed highest state while keeping the same rule for advancing cleared cells below it.

That limit applies to an individual cell. A cluster can cover several wards, so its sum can exceed the value of a single ×128 ward. The cell ceiling alone does not set a ceiling on the total payout of a round.

We also cap the whole round at 10,000× the base bet, in every mode. The base spin and its Total Eclipse feature share that limit. Each paying step can add only the amount remaining below the cap. Reaching it ends the round immediately, including any remaining free spins.

These limits act at different levels: ×128 bounds the state stored at a cell, while 10,000× bounds the round total. Neither number describes how likely a particular outcome is.

RTP covers the whole distribution

Neon Eclipse's theoretical RTP is 96.50% in every mode. We calculate it from the weighted payouts of complete rounds, divided by the cost of the mode. The calculation includes the wards, the feature and the round cap, rather than treating the doubling ladder as a separate return on top.

Our math tables contain simulated rounds with weights. We calibrate those weights and make an exact correction to reach the RTP. The ceiling is part of the payout rules used for that calculation; choosing a ceiling does not, by itself, produce the target average.

The distribution matters as well as the average. Our math notes identify the compounding ward sums in the feature as a source of volatility. A multiplier that persists across free spins changes how unevenly payouts are spread, even when the weighted average remains fixed.

We check the ordering as well as the arithmetic: our model tests cover the ward ladder, sums, reading before removal, saturation at ×128, base resets, feature persistence and the round cap. Those are separate behaviours that all have to agree with the rules.

For more on weighted returns, read how we test a slot's math. RTP is a long-run average, not a prediction for a session. A persistent ward grid records what has happened inside the current feature; it promises nothing about what comes next.

18+ only. Stake.com is not available in every country, including the US, UK and Australia. Gambling can be addictive, so play responsibly. Responsible gambling on Stake