[Coco] It's a small win, but a win nonetheless

[Dave Philipsen]

I think the reason that they use a sine wave is that if you use a square 
wave the fast rise/fall time of the signal can sometimes cause 
unintended consequences (like harmonics, etc.) with analog circuitry.  
You're right that the cheap analog circuitry will eventually render it 
as some sort of rounded off version of the square wave but I have found 
that it's better just to start with a square wave in the first place 
because of the limitations of the analog circuitry.

That being said, I just finished a little project for someone who was 
using data-on-tape circuitry to read signals but they had long since 
abandoned the "on tape" part.  They are now playing the data directly 
from a DVD player.  So I wrote a program that took their data and 
transformed it into a PCM / wav file using biphase modulation.  In that 
particular case it worked quite well just shaping the data as square waves.

Dave


On 3/12/2017 3:03 PM, Brett Gordon wrote:
>
> I thought it interesting tandy choose a sine wave.  Seems overkill,
> considering (1) a square wave would be mushed into a sine by the cassette
> deck anyway, and (2) demodulation just uses zeroq!
> crossing.
>
> It would be interesting to see if some other encoding would work
> better/faster... like QAM, or MultiFrequency Keying (which is like qam).
>
> Cool project Jim.
>