Calculate the probability of obtaining at least K copies of one target drop by a chosen draw count. Every trial gives either one target drop or none. This is a game-balancing calculator, not a connection to a live game.
On attempt a since the last success, chance is min(100%, base + max(0, a − soft + 1) × increment). Hard pity forces success at its attempt. Every success resets the attempt counter, even before the requested copy count is reached.
Dynamic programming tracks failures since the last success and copies obtained; no random simulation is used. Probabilities use double-precision arithmetic. Median, 90% and 99% thresholds are shown only if reached within the chosen horizon. The gray comparison has no pity.
This does not model rate-up guarantees, multiple items per draw, duplicate exclusion or inventory collections. Maximum 1,000 draws, 10 copies, hard pity 500 and 6 million state updates per run.