| Date: | November 17, 2003 / year-entry #132 |
| Tags: | history |
| Orig Link: | https://blogs.msdn.microsoft.com/oldnewthing/20031117-00/?p=41803 |
| Comments: | 6 |
| Summary: | My favorite bad CD-ROM drive from Windows 95 was one where the manufacturer cut a corner to save probably twenty-five cents. The specification for CD-ROM controllers indicates that each can host up to four CD-ROM drives. When you talk to the card, you specify which drive you wish to communicate with. The manufacturer of a... |
My favorite bad CD-ROM drive from Windows 95 was one where the manufacturer cut a corner to save probably twenty-five cents. The specification for CD-ROM controllers indicates that each can host up to four CD-ROM drives. When you talk to the card, you specify which drive you wish to communicate with. The manufacturer of a certain brand of controller card decided that listening for the "Which drive?" was too much work, so they ignore the drive select and always return the status of drive 1. So when Windows 95 Plug and Play goes off to detect your CD-ROM drives, it finds four of them. Apparently this was a popular card because the question came up about once a week. |
Comments (6)
Comments are closed. |
Note that it was 25 cents per card, which if they were as popular as you say, would have been a whole lot of cents. Besides who’s gonna blame the card manufacturer? Because "it’s that darn Windows thing that finds so many CD-ROMs," so they got away with it…
Honestly after starting to read your blog I’m much more understanding of the MS Windows team and any hmm…strange behaviour that sometimes occurs (hmm, maybe this is all just a marketing scam to make people less anti-MS).
I thought it would be simpler to communicate with machines, instead of humans. Well, it is quite difficult to understand and react according to other party, because there are billions of different characters and some have 2 or more characters but they appear they are the same person.
However, Raymond Chen barely and obviously proves that even though we engineers communicate with computers, we come up with the same result; people does not understand, never ever. Should an engineer have to consider another engineer’s mistake? Or should Microsoft say "Hey! That was your cheap card’s problem, not mine!"?
"Geopolitics is a sensitive subject", as Raymond Chen said. That was another story impressed me. I think universities must teach "Human Factor" course to engineers as well. Plus, "Let’s obey standards" course can be very efficient.
Standards aren’t
Microsoft does say "that’s your cheap card’s problem!". That’s what happens when you get a blue screen in NT and it says "nv4_disp" or "emu10k1.drv" in big white letters along the top. Substandard hardware and software is annoying, but it pushes the adoption of more robust technologies in response. Now we have fully protected memory per-process to protect against userspace faults bringing down the machine, and if I get a kernel fault I can pull public symbols from Microsoft to backtrace the faulty driver. Another nice benefit of the increased paranoia in modern OSes is that hardware failures are detected more often, such as bad memory.