Re: Clock Card?

[mjmahon@aol.com (Michael J. Mahon)]

Bill Garber wrote:

>"Michael J. Mahon" <mjmahon@aol.com> wrote in message
>20030917230428.19477.00001294@mb-m20.aol.com">news:20030917230428.19477.00001294@mb-m20.aol.com...
>> Bill Garber wrote:
>>
>> >I tried ProDOS users.disk and it claims there is no
>> >time or date. I tried reading the eprom, however I am not sure
>> >if I should read C400 or CC00. They don't display the same
>> >info. It is a 2764 which is I believe 8k so I'm not sure how that
>> >would read since $100 is all that will fit in the slot memory.
>> >How do I find out where the firmware is loading into RAM?
>> >You would think just about any clock software would find it.
>> >ThunderClock software searches the slots, finds none. NSC
>> >looks for it and finds none of course since it isn't looking in a
>> >slot at all. ProDOS seems to find nothing. I don't know what to
>> >try next.
>>
>> This clock card (the battery is a dead giveaway) was probably
>> produced before the Thunderclock and ProDOS.  (Checking
>> chip date codes and/or card copyright date might verify this.)
>
>I'm pretty sure you are right now that I have tried almost
>every piece of software for clock cards that I could find.
>Where would these date codes be located?

On most chips, there is a 4-digit code like "8315", which
means that it was manufactured in the 15th week of 1983.

>> It has an unusual amount of firmware for a clock card.
>
>One of the cards has a 2764, the other a 2708. The eproms
>have the same firmware on them AFAICT without popping
>them out and reading with a eprom programmer which I have
>coming in a few days. I will investigate further and attempt to
>disassemble then.

It sounds like there is only 1KB of ROM, then.  The 2764
is likely just a more readily available substitute for a 2708
(though a surprisingly large one!).

>> Odds are, none of the firmware is loaded into RAM, but
>> the ROM is bank-switched into the $C800 space, probably
>> by using some of the extra port bits on the 6821.
>
>I'll read from C800 on both cards and see what is there.

You will need to disable 80-column firmware by typing ESC-Q,
then use the monitor (CALL-151 or MTR).  You should then
reference the $C400 space (if it's in slot 4) first, to enable the
expansion ROM.  Then you can move $C800.CFFF to an unused
RAM location (say, $1800) by typing "1800<C800.CFFFM",
then BSAVE it (BSAVE EXPROM,A$1800,L$800) for future
reference. (If the 80-column firmware remains active, sending
characters to the display will deselect the expansion ROM
and select the 80-column firmware.)

>> In its "default" state, there is almost certainly slot ROM
>> code in the $Csxx space, and a perusal of this code would
>> reveal how the 6821 is manipulated.  It probably also uses
>> $C800.CFFx space, with the 2KB chunk of ROM mapped
>> in depending on the 6821 port settings.
>
>Can you give me exact location codes for say slot 4?

The slot ROM space for slot s is from $Cs00 to $CsFF.

>> Try disassembling the $Csxx space and see where it
>> leads you.
>
>I will as soon as you verify what location I need to read.

Start by moving $C400.C4FF to a RAM location, say $1400,
and then BSAVE it for future reference and disassemble it
starting at $1400.

>> (BTW, the 5832 requires at least 2.2 volts to keep time, so
>> there should be two cells on the card...)
>
>There is only a AA holder on the card, but there is a voltage
>boost circuit in place so that is probably correct.

Wow, that's a peculiar way to go--an "ultra low-power" boost
supply would have a very low efficiency.  But then, an AA cell
has plenty of capacity to crank out, say, 0.5 mA for six months
or more...

-michael

Check out amazing quality sound for 8-bit Apples on my
Home page:  http://members.aol.com/MJMahon/