Disasters and Recovery

From: Eric Smith <eric_at_brouhaha.com>
Date: Sun Jan 17 23:12:27 1999

> I think it's fair to assume that some kind of microprocessors will exist
> in 100 years time.

Sure. My point was that if the format was designed to be so easy to read
that a microprocessor isn't necessary, then it should be easy for someone
to build or fix a CD-ROM drive (with or without a microprocessor).

If the format had been designed with very powerful microprocessors in mind
(as is the case for anything designed now), the format would probably be
very complicated, and it would be much more difficult to engineer your own
drive.

> Philips actually sold a CD-ROM drive that was based on a CD player. The
> modifications were a new microcontroller for the servo system (so that it
> could accept commands from the PC to move tracks, etc), a gate array to
> extract the data bits from the off-disk bit stream and a PC interface
> card with an ASIC + buffer RAM on it to get that data onto the PC.
>
> I have one. I even have the service manual for it. While it doesn't give
> much information on the PC interface, it includes enough info to figure
> out how it works.

I'm not sure which one you're referring to, but their first CD-ROM drive, the
top-loading CM-100 (aka DEC RRD-40, apparently) was based on the same
mechanism as their first Marantz brand CD-Audio player. Aside from the
CD-Audio chip set, there are no ASICs or other difficult-to-reverse-engineer
parts in it, or on the CM-153 host interface. They just took the audio chip
set, disabled the interpolation, and added an 8051 and about 35 TTL and CMOS
SSI chips to massage the data. Since the Philips CD-Audio chipset was
reasonably well documented, it was really easy to figure out how everything
worked and disassemble the 8051 code. Try that with a current 40X CD-ROM
drive. :-)
Received on Sun Jan 17 1999 - 23:12:27 GMT