[Coco] History of Tandy/Radio Shack Cassette Recorders

[John E. Malmberg]

Aaron Banerjee wrote:
> I'm sure just about everyone's aware of this, but just in case:
>    Back in the "good 'ol days" when we used cassette to store our
> programs/data, there was a difference in cassette recorders.  Some
> cassette recorders were labeled "Computer Compatible", which meant that
> they did not invert the signal played back by the tape (if you're just
> listening with your ear, it doesn't matter whether or not the cassette
> recorder inverts).

In all of the time that I have looked at home computers and cassette 
data storage, this is the first that I have heard of this issue of 
inverting the signal, and it does not seem to make sense.

Since it is a frequency, not level modulated signal that is AC coupled, 
I find it hard to believe that anything would care if the signal was 
inverted.

The phase inversion should only matter with a DC coupled signal, and the 
level of the signal indicates a one or a zero, something not done on any 
of the home grade computers.

I suspect that much more effort was put into convincing people that they 
needed a more expensive "Data grade" recorder than was needed.
A data grade recorder is only needed for 9600 baud and higher.

Data recording on a standard cassette runs at a constant level and more 
fully saturates the tape than an audio recording.  At even the high 
speed of 1200 baud it still provides a large amount of signal, much more 
  than is needed to reliably extract it using the detection circutry in 
a color computer or even the old model I.

For the home computers of the day, the requirements for data recording 
were far less stringent than what someone would want for a high-fi 
recording.  The media and the transports could be cheaper than what an 
audiophile would put up with and the computer would be quite happy.

The most important thing was having a reasonably constant speed, and 
next not distorting the signal too much.

I used the cheapest tape recorders and the cheapest tapes, and I got 
excelent results.

>    There is a US Patent specific to the coco which is the CLOAD/CLOADM
> routine (including the sourcecode).  The invention was that the coco
> doesn't care whether or not the tape recorder inverts -- you can use your
> walkman to record/load coco programs with a coco.

The coco decoder is a zero crossing detector, and it is not really that 
much advanced from the detector on a TRS-80 Model I/III/IV.

The difference is that because the COCO generates a stepped sign wave, 
the recorded signal does not distort as much.  The analogue audio 
amplifier and tape head effectively smooth out the waveform.

The other TRS-80s used square waves and that was what made them more 
sensitive to the tape recorder.  If they overdrove the tape recorder, 
then an accurate signal could not be played back.

>  I thought I'd be funny
> and use an 8-track for CLOAD/CSAVE once (by the way, there's a reason that
> never caught on...). 

8 tracks had the problem where they did not provide good contact with 
the pinch roller.  The commercial versions of the them at the radio/TV 
studios have a slot where the pinch roller is inserted by the deck. 
Other than that they look about the same.

This makes a big difference in the wow and flutter, and a computer would 
care about that.

> I don't have the patent with me, but the inventor
> was Dale Chatham.  I'll look it up if anyone's interested (who doesn't
> already have the patent).
>              - Aaron

I am interested if there are any active patents that someone writing an 
emulator, or building a home made "sound" attachment for an old laptop 
might run up against.  I would expect by now that all the patents had 
run their course.

-John
wb8tyw at qsl.net
Personal Opinion Only