学习路径Learning path · 第 5 阶段Stage 5
先发极简版
Ship a rough version first
一句话：绝大多数独立游戏的失败原因不是不好玩，而是做完 80% 才发现方向错了。
In one line: most indie games do not fail because they were bad — they fail because the wrong direction was discovered after 80% was done.
⚠ 行业最大的风险
| 你以为的风险 | 真实风险 |
|---|---|
| 做得不够好 | ⚠ 方向错了 —— 做得多好都没用 |
| 竞争太激烈 | ⚠ 没人玩（但没人知道，因为没人试过） |
⚠ 你无法通过「更努力地做」来发现方向错误 ——
因为你是在做完之后才知道，而那时已经投入了几百小时。
⚠ The biggest risk in the industry
| The risk you think it is | The real risk |
|---|---|
| Not polished enough | ⚠ The direction is wrong — no amount of polish saves it |
| Too much competition | ⚠ Nobody plays it (and nobody knows, because nobody tried) |
⚠ You cannot discover a wrong direction by working harder on it —
because you find out after finishing, by which point you have invested hundreds of hours.
为什么 AI 让这件事可行了
过去：发一个能玩的版本 = 3~6 个月（美术、音频、关卡都要做完）
现在：⚠ 原型成本降到了可以「多发几个试」的程度
⚠ 这才是 AI 做游戏真正带来的方法论变化 ——
不是「做得更快」，而是「试错的次数变多了」。
Why AI makes this affordable
Before: shipping something playable took 3–6 months (art, audio and levels all had to be finished)
Now: ⚠ prototypes are cheap enough to "ship a few and try"
⚠ This is the real methodological change AI brings to game-making —
not "it gets done faster", but "you get more attempts in".
怎么做
1. 砍到只剩核心
发出去的版本必须有：能操作、有目标、有反馈。
⚠ 不能有：好看的美术、完整关卡、设置菜单、音效。
2. ⚠ 明确说清这是「试玩版」
⚠ 不要让人以为这是最终版 —— 否则负面反馈会盖过真实反馈。
3. 看数据，不看留言
| 信号 | 说明什么 |
|---|---|
| 多少人玩完了 | ⚠ 核心循环是否成立 |
| ⚠ 有多少人玩第二次 | ⚠ 这是最关键的指标 —— 没人玩第二次说明不够有趣 |
| 留言里问的是什么 | 「怎么玩」= 引导问题；「还有吗」= 需求真实 |
How to do it
1. Cut it down to the core
What ships must have: it is operable, goal-directed and responsive.
⚠ What must not ship: nice art, complete levels, a settings menu, sound effects.
2. ⚠ Say clearly that it is a playtest
⚠ Do not let people think it is the final version — otherwise negative feedback drowns out the real signal.
3. Read the numbers, not the comments
| Signal | What it tells you |
|---|---|
| How many finished it | ⚠ Whether the core loop holds |
| ⚠ How many played a second time | ⚠ The most important metric — nobody replaying means it is not interesting enough |
| What the comments ask | "How do I play this?" = onboarding problem; "Is there more?" = real demand |
常见失败
- ⚠ 追求「做完再发」 —— 最常见的失败模式，也最难自己察觉
- 发给朋友 —— ⚠ 朋友会给你礼貌性反馈，陌生人才给你真实反馈
- 看点赞数 —— ⚠ 点赞是礼貌，不是参与
Common failures
- ⚠ Aiming to finish before shipping — the most common failure mode, and the hardest to notice yourself
- Sending it to friends — ⚠ friends give you polite feedback; strangers give you real feedback
- Watching the like count — ⚠ a like is politeness, not participation