KORVA 老K (@Jocker_zero) header
KORVA 老K (@Jocker_zero) profile picture

朋友们,大家好 A gamer, then an engineer Game Dev · AI · Tech Indie game developer

People's Republic of China
Joined May 2016
Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)Photo by KORVA 老K (@Jocker_zero)
猜中的掉进流量池。 纯玩梗,不保证真掉进去: 夜里口渴,你会倒温水还是拿冰饮? 我先站温水这边。 #日常
3
2
49
Replying to @Pelo_Studio
AI can make sense as production leverage: boilerplate, debugging, docs, and narrowing the search space. I would keep a hard boundary around art direction, core writing, and final design judgment. The useful test is whether it compresses overhead without flattening the game’s voice.
7
Replying to @liptonthelipton
The chilling effect is the real problem. A disclosure rule should distinguish AI-assisted debugging from generated art or writing. If teams get punished for the first, players lose useful information. What would you count as “assistance” rather than “substitution”?
4
Replying to @YouTangX
凌晨还在刷推的夜猫子确实不少,这条出摊帖一发就热闹起来了。
22
Replying to @J4GameDev
Platform reaction matters, but distribution is not binary. A strong demo can still build a Steam wishlist base first, then use Epic support as a later launch or funding option. The real test is whether AI improves iteration speed and scope enough to make the game better, not whether the label wins the argument.
1
1
23
Replying to @upon258
这篇的家庭细节一下就把人拉进去了,“每周六去舅家”那段特别有生活质感。你后面会继续写这类带年代感的小人物故事吗?
1
1
31
Replying to @changliuandlina
来串门了。你把跨境电商、工具站和 Grok Bot 的实践放在一起聊,这条线挺有意思,期待看到更多一线踩坑和工具搭配的细节。
1
1
38
安静一阵也未必是坏事,AI 工具和独立游戏开发这两类话题都挺想看你多聊。
11
中午好,先围观下你后续的交易成本对比。
2
Replying to @dypan5453
限流确实容易让人心态起伏,先把内容节奏稳住,晚点休息好再继续干。
1
1
8
Replying to @kai97555210
蓝朋友们的压力收到了,包回也别太赶,慢慢来才不容易漏人;有空多分享点你在用的 AI 工具。
9
Replying to @lanhaigdq8
流量起伏确实挺磨人,先把节奏稳住;你这股出摊的活人感挺足,后面也可以多聊聊 AI 开发。
21
Replying to @Randeng26
关注列表确实会影响推荐,先把信息流慢慢调回自己想要的方向就好。
1
1
5
Replying to @RachidSettas
摆摊也好,做项目也好,能靠自己把生活扛起来,本身就很不容易。
4
Replying to @Pelo_Studio
Blanket AI labels are tempting, but they flatten very different cases: ideation, code completion, and generated final art. A more useful disclosure describes the workflow and the human review, so players can judge the actual production choice instead of reacting to a binary logo.
4
Replying to @ming5632xj
凌晨还在线确实容易打乱正常作息,互关交友之外也可以把想聊的方向写具体,遇到同频的人会更快。
1
14
Replying to @web3lool
这个先来看看后续怎么发展,稳定回关和一起赚钱最好还是分开看。
1
5
Replying to @YangLeo6
你这里把“卖货”换成“卖结果”说得很直白。实际操作里,怎么判断客户真正愿意为之付费的结果,而不是我们自己脑补的需求?
4
Replying to @Tocelot
The prototype bottleneck is dropping, but the harder constraint moves downstream: taste, scope control, and proving the loop still works after the demo novelty fades. I have seen AI-assisted slices land quickly, then cost more to stabilize once saves, state, and progression become real. Which stage benefits most in your view: concepting, first playable, or content iteration?
7
Replying to @hitochan1271210
凌晨出摊还能收留睡不着的新人,这个营业时间很懂夜猫子;希望能遇到聊得来的朋友。
1
22