23 Apr
2026
23 Apr
'26
4:26 p.m.
Jim Bender via cctalk <cctalk(a)classiccmp.org> writes:
In what may classified as a momentary lapse of reason,
I have taken up the old
Living Computer Museum Massbus emulator project and am in the process of
resurrecting / modernizing it.
[ snip ]
As I have gone down this rabbit hole, naturally a few
questions in the
“why did they do THAT??” category have come up...
paging @Rich Alderson ...
If anyone here is still actively using Massbus, has
experience with
the original LCM project, or just has relevant war stories, comments,
warnings, or encouragement, I would be glad to hear them.
You rang?
I forwarded Jim's message to the two guys who did the original and
follow-on Massbus Disk Emulators, and one of them has already responded
to him directly.
The history, for those who are interested:
In 2004, we restored a DECSYSTEM-2065 which Len Bosack of XKL sold to
Paul Allen in 2000-2001, as part of getting several systems running in
PGA's personal collection. We also restored a pair of RP06 disk drives,
including purchasing the last NOS HEPA filters for them, but it was very
hard to find new media.
My colleague Keith Perez, a wizard at this kind of thing, suggested that
he could build an emulator for the disk drives in short order, and spent
the next two years creating the original MDE. This was based on a
Rabbit processor accompanied by about half a dozen PIC processors; the
back end was a Slackware Linux system which served up files representing
individual tracks via FTP. (I wrote my first ever C code in anger
making the Rabbit's FTP program actually work.)
This was fast enough to emulate 4 RP06 drives and allowed us to run
Tops-10 v7.04 on the 2065, which we opened for public access accounts
under the PDPplanet.com umbrella.
In December 2011, I arranged for the museum to acquire a KI-10 from a
collector in Germany; PGA insisted that we get it running before the end
of 2012. By that time, Bruce Sherry was working for the museum, and
proposed that he refrangle the design using an FPGA in place of Keith's
board full of Rabbit and PICs; his version was fast enough to supply a
full 8 RP07 drives, so we used 2 on the KL-10 system to give it the RP06
it needed for booting the 11/40 front end. The KI-10 used one since it
booted directly from the PDP-10 formatted disk.
Early on, Michael Thompson was very helpful to Keith in getting the
original design working, in exchange for the promise of an MDE when it
was working. We were finally able to fulfill that promise with one of
Bruce's units.
In addition to running the KL-10 and KI-10 systems (Tops-10 v7.04 and
v6.03A, respectively), we ran the former MIT-AI KS-10 system under ITS
using one of Bruce's units.
--
Rich Alderson news(a)alderson.users.panix.com
Audendum est, et veritas investiganda; quam etiamsi non assequamur,
omnino tamen proprius, quam nunc sumus, ad eam perveniemus.
--Galen