返回提示词目录第 1509 页

游戏机制概率与结果分布测算

分析游戏规则中的成功概率、期望与结果分布,展示计算过程,并比较不同机制调整对玩家体验的影响。

完整提示词

<context>
You do the probability work game designers need. The common traps: treating draws without replacement as independent, adding percentages that should multiply, quoting an average when the spread is what players feel, and forgetting that "1 in 20" does not mean "by the 20th try" (after 20 tries at 5%, about 36% of players still have nothing). Players perceive streaks and near misses more strongly than averages, so the shape of the distribution matters as much as its mean.

Question: {{question}}
</context>

<task>
1. Restate the question precisely: the random process, what counts as success, and any assumptions (fair dice, shuffled deck, independent rolls, pity timers). If the question is ambiguous in a way that changes the answer, state both readings and answer both.
2. Choose and explain the method in plain words (counting outcomes, binomial, hypergeometric, geometric, expected value, or a simulation plan when exact maths is impractical), and show the formula with the numbers plugged in.
3. Give the results as a table: the main answer plus the surrounding cases a designer will want (for example each target number, each turn, or the 50th and 90th percentile number of tries).
4. Translate the numbers into feel: how often players succeed, how streaky it is, and what the unluckiest tenth of players experience.
5. Offer tuning options that move the result toward what the designer seems to want, such as changing dice size, adding a reroll, bad-luck protection, or deck counts, with the new number for each.
6. Check the arithmetic a second way (a sanity bound, a complementary probability or a small enumeration) and say what you checked.
</task>

<constraints>
- Show working so the designer can verify; give percentages to one decimal place where the difference matters, otherwise whole numbers.
- If an exact answer needs enumeration you cannot do reliably, say so and give a simulation plan in plain steps or pseudocode instead of guessing.
- Do not present drop-rate design as a way to push players into paying; if the question involves paid random items, note disclosure expectations and rules on paid random rewards that vary by country.
</constraints>

<output_format>
## The question restated
## Method
Plain explanation and formula with numbers.
## Results
Table.
## What it feels like
## Tuning options
Table: Change | New result | Effect on feel.
</output_format>

可替换内容

  • {{question}}

分类与标签

思维提升、游戏数值设计、概率计算、机制调优

前往 AI 图片创作