[rescue] Current collections...

[Geoffrey S. Mendelson]

Dave McGuire wrote:

>    No...sampling != demodulation.

I can think of a lot of fancy mathematics to convert the tapes to data,
but I think the easiest way to do it would be to read a standard wav
file (44.1k 2 channel) and look for bit transitions. It could be done
easily in C (or almost any other language for that matter), and would
produce a bit stream from a "tape".

The point being that anyone could record the wave file on almost any 
computer with a sound card (PC, MAC, etc) and then make the bit stream
out of the samples. Once you had the bit stream, you could write a seperate
program to decode it into the proper format for the computer in question.

This method appeals to me because all you need is some standard equipement
(computer, soundcard, walkman) and the tapes. 

There really is not a lot of data on them, AFIK Comodore (PET, VIC-20, C64)
wrote data at 300 bits per second as did the 8 bit Atari's. Some CPM systems 
went as fast as the 1200 bits per second. I think that the IBM PC used 1200 
bits per second, but I've never seen a tape for one.

What I don't know is the baud rate and encoding method, but I'm sure it's
well documented.

Geoff.
-- 
Geoffrey S. Mendelson gsm at mendelson.com