Re: Mark space ratio in 1531 data signal

Den Tue, 10 Dec 2019 17:15:14 +0200 skrev Marko Mäkelä
<msmakela_at_gmail.com>:
> On Tue, Dec 10, 2019 at 04:00:55PM +0100, Mia Magnusson wrote:
> >>For writing the tape head is driven to saturation or not. It is not 
> >>very complicated.
> >
> >Rather saturation in either polarity.
> 
> A 50% mark/space ratio might have the benefit that it is immune to
> any change of signal polarity.
> 
> The 264 series machines wire CASS READ to a level-sensitive I/O pin, 
> while all others (PET, Vic-20, C64, C128, CBM II) connect it to 
> something that detects a falling edge. So, only for the 264 series it 
> might matter which polarity is used. In reality, the 264 series uses
> an even slower encoding (and slightly different high-level format)
> than the other machines. See the source code of my C2N232 firmware
> and related software (c2nload, c2n) for more details.

Had a look at the schematics of the PET (both 2001 and 8296, didn't
bother to check any else). The first cassette port is hooked up to the
6520 while the second is hooked up to the 6522. The 6520 seems to be
programmable so that you can select which edge it gets triggered by.
You'd need to change that programming each time the input is triggered
to be able to detect both edges though. So at least in theory it seems
like it's possible to detect the pulse ratio when using the first
cassette port on a PET.

The CBM-II B and P uses the _FLAG input on their 6526 CIA, so it's the
same as C64/C128.

Side track: Of the other 8-bit computers with "digital" cassette
recorders I can think of, the Sharp MZ range is polarity sensitive
while I'm not sure if polarity matters on the Spectravideo and
the Swedish ABC 80 and ABC 800 computers.



-- 
(\_/) Copy the bunny to your mails to help
(O.o) him achieve world domination.
(> <) Come join the dark side.
/_|_\ We have cookies.