William Donzelli wrote:
> So thay are Data General "mostlies". Will they run early Nova code,
> before Data General started to expand the architecture?
Yes. The new instructions were created (in the same fashion as
DG later did with the Eclipse) by using instructions that were
effectively no-ops. Anything that was willing to run on an
unmapped Nova 800 should be willing to run on the 16xx machines.
Did you get programmer's consoles with any of these machines?
All ROLM machines implemented the Nova I/O bus, albeit with
very different physical characteristics. This produced an
interesting problem with respect to one of the largest sales
of the 16xx -- the GLCM/SLCM program at McDonnell/Douglas.
The 1602 and 1666 (one of each, I believe) are used in the
erector and launch systems; when the Nuclear Assurance folks
found out that the Nova I/O bus lacked parity they freaked.
> These machines do not have the gaskets or a fill port. The 1666B has
> massive heat sinks on the side, however.
...which eventually evolved to enclosed plenums with forced air in
the Hawk/32 (which implemented the MV instruction set).
DesignData (www.designdata.com) ended up with most of the prints
and spares for ROLM machines; they may be able to help with
more detailed technical information.
--
Chris Kennedy
chris_at_mainecoon.com
http://www.mainecoon.com
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
Received on Sat Dec 22 2001 - 18:05:25 GMT