历史与人物bigmessowires.com
DB-19：复活一个过时的连接器
DB-19: Resurrecting an Obsolete Connector
2016 — 2024，它一次次回到首页
黑点是上首页（≥ 50 分）的那次投稿，空心点是没冲上去的投稿，点越大分数越高。
导读
这篇文章讲述作者如何为一个已经停产的 DB-19 连接器寻找制造途径，并最终小批量复刻出来。它围绕过时电子零件的采购、开模与定制生产展开，也涉及爱好者面对高起订量和供应链缺口时的替代方案。
为什么经久不衰：它把复古硬件维修中“买不到就自己造”的困境和供应链博弈写得很具体，对今天的创客、维修者和复古计算爱好者仍有参考价值。
HN 评论精选
定制停产连接器成本高、起订量高，小市场难以支撑；改造替代品可行但可靠性差。
这故事太棒了！我们遇到过类似问题：要找到能制造本已不存在的零件的人——一个28针卡边式 Super Nintendo 扩展端口连接器。唯一愿意考虑的公司要5000到8000美元初始设置费，起订量上千；总成本约15000美元，而我们潜在市场可能只有50到100销量。最后我们找到 Samtec 一个用于完全不同应用的更宽卡边连接器，用打磨机把连接器切成两半，再用环氧树脂填封一侧，做成替代品。缺点是稳定性差，切半填胶后插拔寿命最多几百次，针脚就不能可靠连接；手工打磨也大增工作量；而且它们是表面贴装，排针极小，需要专家组装。团购对我们不现实，但我仍希望像作者用3D打印等手工制作尝试最终成功。
byuu查看英文原文
This is an awesome story! We faced a similar problem trying to locate someone that could manufacture otherwise non-existent parts: a 28-pin card edge Super Nintendo expansion port connector. The only company we could find that would even consider it wanted $5000-8000 for the initial setup fee, and an MOQ in the thousands. It would have set us back around $15000, but our potential market is maybe 50-100 sales. In our case, we ended up finding a wider card-edge connector from Samtec for an entirely different application, and we dremeled the connector in half, and then filled the side in with epoxy to create this: http://www.qwertymodo.com/hardware-projects/snes/snes-expansion-port-connector The down-side is, they're not very stable. Sawing them in half and then filling one end with epoxy results in connectors that have mating cycles rated in the hundreds at best before the pins stop making clean connections. It also adds substantially to the workload to hand-dremel the parts. And they're surface-mount, with the tiniest headers you've ever seen, requiring an expert to assemble these boards. Unfortunately, a group-buy isn't really viable for us. But I'm still hopeful that at some point, we can achieve success with something like the author's attempts at hand-making things with 3D printing and such.
过时零件常需定制，先付费做原型再批量生产的分阶段合作可降低风险。
不错。这家伙为 DB-19 连接器问题折腾了一两年，终于找人做出来了。有时过时零件就得这样。几年前我让人在中国做过 Teletype 纸带打印机的纸带。和中国零件制造商打交道，常见做法是先付溢价做几个原型，确认可行后再付批量发货；一开始就付全款风险很大。
Animats查看英文原文
Nice. This guy has been struggling with that DB-19 connector problem for a year or two. Finally, he had some made. Sometimes you have to do that for obsolete parts. I had Teletype tape printer tape made in China a few years ago. When dealing with a parts manufacturer in China, it's not uncommon to make a deal where you pay a premium for a few prototypes, and after those work, you pay for a volume shipment. Paying for the whole job up front is a big leap of faith.
中国厂商的起订量常可谈判，实际小批量也可能做。
好故事。补充一个小数据点：以我有限的中国制造商经验，起订量很少能撑过最初几封邮件。他们喜欢先抛一个吓人的起订量看你是不是认真，或者只是初次尝试，或者因为那个起订量绑定的是别处报的超低价。但进一步询问后常发现起订量并不真正有约束力。例如最近有人告诉我某面料设计必须订5万件，而实际上我能做不到500件，价格仍有竞争力。另外，一旦你与某人接洽，他们会对你的项目投入，甚至违反自己组织规则帮你做成。不会为你卖孩子，但肯定会和内部会计聊起订量。
bambax查看英文原文
Great story. Just to add a small data point: in my (limited) experience with Chinese manufacturers, MOQ rarely survive the first couple of emails. They like to throw at you impressive MOQ just to see if you're serious, or maybe as a first try, or maybe because those MOQ are attached to the really low prices they have quoted elsewhere. But when you inquire further you often discover that MOQs aren't really binding. For instance, I was recently told I would have to commit to 50000 pcs (50k!) of some fabric design... and in fact I'll be able to make less than 500 at a still very competitive price. Also, and this may be true everywhere, once you engage with someone they become invested in your project and will work with you, even against the rules of their own organization, to make it happen. They won't sell their kids for you, but they will most certainly chat with the in-house accountant about MOQs.
澄清 D-sub 命名：所谓 DB-9 实为 DE-9，DB 指外壳尺寸而非针数或类型。
从来就没有 DB-9 连接器；它们是 DE-9。‘DB-这’‘DB-那’的叫法源于无知地类比（正确命名的）DB-25，DE-9 取代了它。提示：D 后面的字母是尺寸。DB 外壳可容纳间距0.1英寸的两排12和13针，往下到 DE 只容纳4和5针同样间距。后来 DE 外壳用更密间距做三排针，用于视频显示器线缆。等足够多无知的人开始叫 DB-9 并按此名购买，供应商也开始这样卖。
ncmncm查看英文原文
There were never DB-9 connectors; they were DE-9s. The "DB-this" and "DB-that" naming originated by ignorant analogy with the (correctly-named) DB-25 connectors that the DE-9s were replacing. Hint: The letter after the D is the size. DBs had room for two rows of 12 and 13 pins on 1/10 inch centers, on down to DE with room for only 4 and 5 pins with that spacing. Later, the DE shell was used with closer spacing for three rows of pins, for video monitor cables. Once enough ignorant people took to calling them DB-9 and trying to buy them by that name, the vendors started hawking them that way. https://en.wikipedia.org/wiki/D-subminiature#DE-9
历次讨论
同类还值得读
开发者无薪坚持完成 Mac 图形计算器：一个秘密项目的故事
Dev stays without pay to finish Mac Graphing Calculator as skunkworks project
苹果 Mac 图形计算器（Graphing Calculator）开发者的亲述：在项目被砍后，几位工程师在没有报酬、没有正式授权的情况下，以“地下项目”方式偷偷留在公司把软件做完…
历史与人物上首页 6 次 · 2009–2022pacifict.com理查德·费曼与 Connection Machine
Richard Feynman and the Connection Machine
长文回忆费曼晚年参与 Connection Machine 并行计算机项目的经历：他分析路由器、用自创的并行 Basic 编写 QCD 程序，也涉足细胞自动机等主题。
历史与人物上首页 5 次 · 2010–2021longnow.org编辑精选负 2000 行代码
-2000 Lines of Code
folklore.org 上 Andy Hertzfeld 记录的一则苹果 Macintosh 开发轶事：Bill Atkinson 被问到写了多少行代码时，回答“-2000 行…
历史与人物上首页 4 次 · 2014–2022folklore.org编辑精选Dennis Ritchie 的第一版 C 编译器（约 1972 年）
Dennis Ritchie’s first C compiler (c. 1972)
这个 GitHub 仓库保存了 Dennis Ritchie 约 1972 年编写的最早 C 编译器源码，是 C 语言和 Unix 早期历史的一手材料。
历史与人物上首页 4 次 · 2013–2025github.com一位破解赛马密码的赌徒
A Gambler Who Cracked the Horse-Racing Code (2018)
彭博社 2018 年的长篇特写，讲述数学家 Bill Benter 如何用统计模型和算法在赛马博彩中持续盈利、累计赢得近十亿美元的故事。
历史与人物上首页 4 次 · 2018–2024bloomberg.com