This is the actual PCB layout of the
revision X8A PCB for the Atari "Sweet-16". This is where things get
interesting. Click on the above image to examine it more
closely... Note that the layout is entirely different from the
production 1200XL computers. Take particular note of the upper right
corner of this PCB design - there is a 50 Pin header footprint and the
traces all lead to the major IC's on the board which means this was the
very first PBI (Parallel Bus Interface) which appears to have been meant
as an internal device interface versus the later XL designs which
brought the PBI directly out of the system to an Edge Connector.
Next thing to note is the Upper
Center edge of the PCB, note the 2 DB-15 footprints versus an Atari SIO
footprint. This directly correlates with the early Atari Sweet-16
spec's document detailing the inclusion of 2 15 pin connectors to
replace the original SIO and would allow for peripherals to
automatically power up when the system powered up.
The departure from the X1A-X8A design
to suddenly a radically different X10A design that appears at the same
timeframe in July 1982 means that there must've been a side by side
"Sister Project" of the Atari 1000 series being built and it moved into
place to become the production machine leading to a disaster product
launch for a system with plenty of sexy looks, but nothing under the
hood, so to speak.
Several things we have to take away
from the original 1200XL with the expansion header are this:
Every revision of the new 1200XL
OS - Revision 10 (which was the replacement for the Atari 400/800 OS
Revision B, never had any support for handling Parallel Devices.
Revision 11, which came out in the last several months of the Atari
1200XL production run (Jan 1983 to end of June 1983) also had no
support for Parallel device handling.
We would have to assume the
Expansion header was just buss line connections with no intelligent
management of any devices connected to it. So this could end
up limiting its expansion capabilities similar to how the 800 even
it if had more than one "Slot 3" would've had difficulty handling
more than one expansion device board.
Atari's RF/EMI engineer noted
that the X8A and earlier was very noisy. So this too could
lend to why the header is removed from the design.
The 1200XL may have simply been a
stop-gap release product and then 3 months after its release the
more formalized designs of the "Surely" - the 600XL and the
replacement for the 1200XL - the "Surley Plus" - the 800XL would now
have all of the Parallel Device support and the design for an
external Expansion System is also created.
The Atari Museum is currently
investigating the possibility of having a PCB design firm that still
works with physical Padmaster films to reproduce actual samples of the
X8A revision motherboard, the real "Sweet-16" computer system to allow
for further study and examination.