From: David Murray on
There was also a kernel mod out there to drive a 40x4 LCD display. I
actually did something similar with my commodore DTV to drive a 20x4
LCD display but I made a TSR program instead of a kernel mod.
From: Andreas Kohlbach on
Jonno Downes wrote on 25. July 2010:
>
> just curious as to what (if any) custom kernals existed (or even,
> still exist)?
> I know about speed loaders like Jiffy DOS, but was there anything more
> extreme? e.g.
> - monitor / debugger in ROM?
> - replacing BASIC ?
>
> I've seen a fair bit of custom ROM images for the apple 2 ("crack
> ROMs" etc as well as industrial control firmware) but not so much for
> the c64 - maybe any custom code like that ended up in a cartridge
> instead

I guess in 1986 a friend and me added one of those floppy-turbo-load [1]
codes to an EPROM. The friend created a small circuit board where both
kernals (the original and the modified on the EPROM) were mounted, and a
switch attached outside next to the expansion port, to switch between
them before boot (it would crash if you switched while operating :-). We
also changed the "boot message". So no need to first load the turbo-load
from a floppy anymore.

[1] When I (German) was in the UK with my class (I guess in either late
1985 or early 1986) I learned that UK mostly had Sinclairs at that
time.

Well I went to a computer shop in Canterbury where we were to see how a
UK shop looks like. I always had tons of floppies with me to swap files,
just in case. :-)

Anyway, the clerk never heard of a turbo loader for the 1541 and was kind
of "Wow, that is unbelievable! That rocks! OMG!". I left him a copy on a
floppy. IIRC one of those I was also changing the load message to
something like "Ank's Turbo Load". Of course I didn't wrote the code of
the loader itself. But I wonder if this copy was spread in that area of
the UK, assuming a floppy-turbo-loader for the C64 was unknown (?) in
that area.
--
Andreas
From: Andreas Meerbann on
> Is the last second (when the bit is 'not transmitted') used for
> synching to the start of the next minute? If so, how do you get that
> synch sygnal through 1 data bit?

Yes, basically the system requires an internal clock running more or
less exact. This is the clock that's being displayed. The external
signal is used to adjust that internal clock in regular intervals -
can be from once per minute to only once per day (used by wrist
watches to save battery life)
To sync the internal clock the software first waits for a couple of
radio pulses to pass by and checks if the distance between these
pulses is roughly one second. When this is the case it indicates that
the reception is stable. Then the program waits for a missing pulse
(or rather a puls that occurs about two seconds after the last one).
This pulse then is the first puls of a minute and decoding of the bits
starts here (short/long puls decoding) for one minute until all bits
of the current date and time are collected. Then the internal clock is
updated with that information. This is usually done at the first pulse
of the following minute.

For extremly high accuracy requirements the carrier frequency of the
Radio signal is also linked to the oscillator frequency of the atomic
clock. A more advanced system can lock to that frequency and use this
as a time base. But that's more for scientific or very special
industrial purposes and not required for a simple clock.

> The only externally synched clocks I've seen here in Australia use GPS
> or NTP
so obviously Australia has just skipped this era of steam-punk
electronics and is now much more advanced than the old fashioned
Europe... ;-)

> interesting! and interesting as well that (from the description) no
> compression was being used? I would have thought that sort of
> transmission would have been a perfect candidate for RLE?
Yes, well I think the successor systems had real digital transmission
with some kind of compression technique. The service I was decoding
was purly analog. The picture content was frequency modulated grey
scale (colour pictures came in three versions for red, green and blue)
and each line had a constant time duration in the transmission. I
think it was similar to the early FAX transmissions that have been
done over the phone line these days ("FAX group 1" and FAX group 2"
are the key words). Nowadays there are similar transmissions under the
key word "slow scan TV" or "SSTV". This is quite popular amongst the
ham radio people.

I now do remember that I also was trying to send a fax over the phone
line by using a plain and unmodified C64. My room mate had an old fax
device which was still capable of the old "group 2" standard which was
using a simple transmission technique. I tried to produce the correct
beeping sounds with the SID to send a picture to that fax machine.
Unfortunately I was not able to achieve a 100% exact timing. Also the
modulation of the picture content was very hard to achieve because (I
think) it was QPSK modulation. It was not possible to control the SID
exactly enough to produce such a signal. So the only achievement I
made was to send a very big black square to the fax machine. It all
ended when the fax machine finally was broken (I hope not due to my
trials) and the next one did no longer support "group 2" transmission.

cheers,
Andi