PDP-6s at MIT

Lars Brinkhoff lars at nocrew.org
Sat Dec 17 06:44:02 CST 2016


Noel Chiappa wrote:

>     > I wonder how the AI PDP-6 was used back in the day.  I suppose
>     > ITS development moved to the KA10 using the virtual memory
>     > pager, leaving the 6 behind.  But it was still attached as a
>     > slave CPU, right?
>
> Yes. There is a document, "February 1972 ITS Status Report", AIM-238,
> which is from a point in time when both CPUs were in operation as a
> dual-CPU system, with paging on the KA10, and it contains a
> considerable amount of technical detail.

Thank you!  Reading that, and the system documentation, makes it quite
clear.

> The earlier "ITS 1.5 Status Manual" might also be interesting; it
> covers a point in time when I think they only had a PDP-6.

It mentions the sparkling new PDP-10, but it seems the PDP-6 was the
primary CPU at the time.

> I'd be interested to hear anything about the DM PDP-6 which you find
> out.

Most of my notes end up in GitHub, but I'll try to keep you in the loop.

Early Project MAC Progress Reports mention both the AI and the DM system
as PDP-6/10 machines.  I suppose it's likely they both used the same
master/slave configuration.  However, there's one thing I found that
may indicate that the DM machine went in an other direction.

Alan Bawdens wrote a draft paper about ITS (maybe what later become the
PCLSRing paper) that says:

> Sequence of machines:
> AI pdp6
> DM pdp6
> AI ka10 (first as attached processor; role switch)
> DM ka10 (pdp6 flushed)
> ML ka10 (first non-special-I/O-oriented system)
> MX kl10 (expand ML group to high performance)
> AI ks10 (begin renaissance)

This describes the progression of machines used in ITS development.  Of
interest here is the "pdp6 flushed" note.  The obvious interpretation is
that the DM group stopped using the PDP-6 when they got their KA10.


More information about the cctalk mailing list