Erik wrote:
thanks for your email. Given this information, I
think
my stuff is not of use for your project...
We have the schematics for the IBM controller,
but don't have one. We'll
need to build our own to connect between the IBM 1620 computer and the
plotter.
I do not know the details on this, but maybe it is easier to
rip an other plotter using stepper motors and connect the clock
and direction signals to the 565 directly. There should be quite
a lot plotter out there suitable for this.
To maintain as much historic authencity as we can to the whole system, we'd
rather build a clone of the IBM controller.
What kind of processor does the IBM controller use? In
somewhat
later Calcomp plotters (as mine from 1972) a OPC (online plotter
controller) was included. The original Calcomp PCB contains a
6800 processor which interprets signals from a RS232C interface.
This PCB generates the signals needed to drive the plotter (i.e.
step and direction pulses). Since this OPC was optional it might
be easier to obtain than the IBM part? Maybe they are identical?
I probably overstated it when I called it a controller. It is discrete
logic that does a little hand shaking with the IBM 1620 to retrieve plot
"digits", a simple decode of the digits, then drive the plotter lines with
the proper timing. It's a very simple mechanism. I was avoiding calling it
an interface because that was how I was referring to the interface circuitry
within the plotter. IBM did the design of the "controller" and used the
"SMS" logic cards that the 1620 is built with. No CalComp circuitry or
design was used.
The IBM 1620 was a 1960-vintage computer. The plotter was available on the
machine in 1962.
For the Calcomp OPC I have got schematics and ROM
dumps. It has
the same name as the language it understands "PCI906" and of
course includes a character generator...
I don't think that this would help us.
The
information we are looking for is: 1) detailed descriptions of the
CalComp HCBS [basic plot] library; 2) the algorithms used by CalComp to
implement the plotting functions; and 3) the vector character font data
that CalComp used.
Sorry, I do not have anything of the mentioned...
I understand. It's amazing to me how available the document was and yet we
haven't been able to find one yet. CalComp was always very tight with their
source code (a big part of their IP), so I'm not surprised that we can't
find that.
We have no
application programs for the IBM 1620 that do plotting
[although
we have 200,000 punched cards of programs and data]. We'll be writing
the
demo application in addtion to the basic plot library.
That sounds great - I wish
you the very best for this
project,
We think that it will be very exciting for museum visitors to see. Most of
the people will never have experienced a plotter.
Thanks,
DaveB