Re: Three chip 6502 machine

On 2014-02-14 at 19:32:08, Marko Mäkelä (msmakela@gmail.com) wrote:

> On Fri, Feb 14, 2014 at 04:25:35PM +0200, Juhani Jaakola wrote:  
> >CQ which is Radio Amateur term for a general call.
>  
> Oh, right. Q is --.- and Y is -.--
>  
> How did I get that wrong? I should have whistled it aloud. I learned  
> Morse Code in the national service in 1995.
>  
> To bring this a bit more on-topic, were there any Commodore programs  
> for
> capturing Morse code from a radio receiver? If yes,

Yes, a friend of mine (and quite famous “scener”) made a very interesting one after he also decided to make a licence.

> how was the signal detection done?

One bit on the USER PORT for sending and another for receiving. A tiny circuit with 555 was required if only audio was available from the receiver. The algorithm was interesting though. It didn’t try to decode in (near) real time as most programs did (and failed with human engendered irregularities, especially on the straight key) and required wpm settings to be set in advance. Instead it tried to capture as much as possible of the transmission and only after encountering a predefined pause, it scanned the input and based on gathered statistics decided what was the wpm, pause/dot/dash lengths etc. It was therefore much more flexible and tolerated higher level of irregularities. Other advantage was that one didn’t have to tune the wpm settings before receiving the code. It worked more like humans - hear and decide how to decode what’s being heard. Although only after finishing to listen, that is..

> I guess with some op-amp circuit similar to what  
> exists in the tape drive?

I used it for training with a straight key. I connected the key via one transistor in the USER PORT plug to invert the level (software was expecting HI on the port bit for carrier presence, the same what would come from the above mentioned circuit or a TRX with appropriate output)

> And how would the data be recorded? Anything  
> similar
> to the *.tap format, which represents the time between output  
> transitions?

I thought of it many times and promised myself that “one day” I will thoroughly analyse how he did it. But - shame to say - I never did the analysis. Maybe that “one” day didn’t come yet ;-) What I wrote above about the way it works is what I guessed and asked the author about. But I expect that there is probably no way to escape from the need to record the time measurement in some way.

P. S. To Juhani: DR OM TNX QSO 73 DE SP1WSC K ;-)
--  
SD!

       Message was sent through the cbm-hackers mailing list