25 Aug
2011
25 Aug
'11
9:22 a.m.
Brent Hilpert wrote:
In a similar vein, I don't suppose any of the
(few) 7030s (STRETCH)
have survived ..?
I suppose it depends on what you mean by "survived". Certainly there are
none that are operational, or even that could theoretically be made
operational without fabricating a lot of replacement equipment.
CHM has most of the Livermore machine, which unfortunately is missing
the core memory, disk, and the original console typewriter. When the
machine was auctioned, one person bid on it to try to preserve the
system, but others bid on the core memory and the console typewriter.
Apparently the person that bid on the typewriter really only wanted a
Selectric typewriter, and didn't need the special one used for the 7030.
CHM is displaying the 7030 programmer's console with a normal Selectric.
(There were probably less than a dozen of the Selectric model for the
7030 manufactured.)
CHM also has a portion of the one that was originally at MITRE then
spent time at BYU. Reportedly a professor at BYU saw it on a government
auction listing, bid on it without having a clue how massive the system
was, and sent some students with a pickup truck to fetch it. Somehow it
was arranged for IBM to pay to have it professionally moved to BYU, and
to get it working again there. (Installing a 7030 and getting it
operational was a herculean feat.) BYU also acquired the Los Alamos
7030 as a source of spare parts, so it is possible that some of the
Stretch hardware at CHM might have come from that machine. When BYU
decommissioned the system, they carefully deinstalled it by chopping all
of the cables. (This is typical of how large computer systems get
deinstalled.)
There's a sad story about the fate of the one-of-a-kind HARVEST system
(7950) that IBM built for the NSA. Harvest coupled a 7030 with a 7951
stream coprocessor, special two-hole-per-core high-speed core memory
(7952), a 7955 "Tractor" robotic tape system, and a 7959 high-speed
exchange (I/O channel controller). When the NSA decommissioned it in
1976, they wanted to give it to IBM for their historical collection.
However, being government property, it had to be put up for auction.
IBM's historian bid $20,050 for it, more than a year's worth of his
budget, and won it. Unfortunately not too many years later someone else
at IBM decided that they didn't want to store it any longer.
I have never been able to find any details of the ultimate disposition
of the other six 7030 systems.
Very little software for the 7030 still exists. The Computer History
Museum has printed listings of some software, and Al has scanned some
(or perhaps all) of it.