Date   : Sat, 27 Nov 2010 11:36:38 +1100
From   : plexer@... (James McGill)
Subject: Electron Ferranti ULA reverse engineering progress

I (am probably) wrong, but I think you need a zero crossing detector
as well, to recover the clock in cases where you have a large number
of 1's or 0's in a row (and assuming the data isn't manchester encoded
or similar).

Use of an external clock is insufficient because of variations in the
speed at which cassettes were played back.

On Sat, Nov 27, 2010 at 10:47 AM, Pete Turnbull
<pete@...> wrote:
> Rick Murray wrote:
>
>> Speculation: Band-pass filter on the two freqs, presence of signal
>> triggers a positive state for either logic one or logic zero, which is
>> then clocked out as a stream of bits... That's my thought. It wouldn't
>> need to be terribly complicated as the input is fairly well defined.
>> Just a bit of leeway on the freqs and clocking in order to cope
>> gracefully with the minor variations in tape speed.
>
> Yup, some modems used to do that. ?Actually you only need one filter,
> and a fairly narrow one at that -- presence of a tone at that frequency
> is a logic zero ("space" state), anything else is a one ("mark" or idle
> state). ?But some modems did have two filters, one for each tone, in
> order to provide a carrier detect (either filter passing signal
> carrier present).
>
> --
> Pete ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Peter Turnbull
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Network Manager
> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?University of York
>
> _______________________________________________
> bbc-micro mailing list
> bbc-micro@...
> http://lists.cloud9.co.uk/mailman/listinfo/bbc-micro
>