29 Nov
2014
29 Nov
'14
11:47 p.m.
So, we have the ABC, Colossus, Manchester Baby, and now the EDSAC.
Anyone for the ENIAC? Univac I? IAS machine? Whirlwind?
How come 3 of the 4 are in Britain?
On 2014-Nov-29, at 9:12 PM, John Foust wrote:
The National Museum of Computing unveils EDSAC
re-creation:
http://www.bbc.com/news/technology-30131447
30 Nov
30 Nov
12:30 a.m.
I would say that in the US you have CHM and the LCM which exhibit working
mainframes. There are also MARCH and the New Jersey (I think) Museums which
show working machines. In the UK several working exhibits have been
"mothballed". The Science Museum has discontinued Pegasus demos, my project
to restore some of the Pegasus i/o equipment at MOSI has been suspended, and
the Hartree Differential Analyser is to be removed from display. Personally
I would rather that money was expended on keeping real mainframes running
rather than building replicas,
I also note that the Baby Replica at MOSI is now around 16 years old, it
first ran in 1998. Its almost an artefact in its own right...
Dave Wade
G4UGM
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Brent
Hilpert
Sent: 30 November 2014 07:48
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: EDSAC lives
So, we have the ABC, Colossus, Manchester Baby, and now the EDSAC.
Anyone for the ENIAC? Univac I? IAS machine? Whirlwind?
How come 3 of the 4 are in Britain?
On 2014-Nov-29, at 9:12 PM, John Foust wrote:
> The National Museum of Computing unveils EDSAC re-creation:
>
> http://www.bbc.com/news/technology-30131447
11:07 a.m.
I would vote for the Whirlwind. In addition to an interesting
architecture, the machine has a fascinating history! -- Ian
On Sun, Nov 30, 2014 at 12:30 AM, Dave G4UGM <dave.g4ugm at gmail.com> wrote:
I would say that in the US you have CHM and the LCM
which exhibit working
mainframes. There are also MARCH and the New Jersey (I think) Museums which
show working machines. In the UK several working exhibits have been
"mothballed". The Science Museum has discontinued Pegasus demos, my project
to restore some of the Pegasus i/o equipment at MOSI has been suspended,
and
the Hartree Differential Analyser is to be removed from display. Personally
I would rather that money was expended on keeping real mainframes running
rather than building replicas,
I also note that the Baby Replica at MOSI is now around 16 years old, it
first ran in 1998. Its almost an artefact in its own right...
Dave Wade
G4UGM
--
Ian S. King, MSIS, MSCS
Ph.D. Candidate
The Information School
University of Washington
An optimist sees a glass half full. A pessimist sees it half empty. An
engineer sees it twice as large as it needs to be.
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Brent
Hilpert
Sent: 30 November 2014 07:48
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: EDSAC lives
So, we have the ABC, Colossus, Manchester Baby, and now the EDSAC.
Anyone for the ENIAC? Univac I? IAS machine? Whirlwind?
How come 3 of the 4 are in Britain?
On 2014-Nov-29, at 9:12 PM, John Foust wrote:
> The National Museum of Computing unveils EDSAC re-creation:
>
> http://www.bbc.com/news/technology-30131447
1 Dec
1 Dec
12:49 a.m.
On 2014-Nov-30, at 11:07 AM, Ian S. King wrote:
I would vote for the Whirlwind. In addition to an
interesting
architecture, the machine has a fascinating history! -- Ian
It's one of my favorite early machines too. It was just so far-looking in objective.
While all the other first-generation machines under construction at the time -
state-of-the-art themselves - were intended for number-crunching applications in what
would become batch-processing, Whirlwind was leap-frogging them by 10-15 years in vision,
targetting real-time simulation with human interaction.
One of the SAGE blockhouses should have been kept intact with the full & functioning
equipment installation.
On Sun, Nov 30, 2014 at 12:30 AM, Dave G4UGM
<dave.g4ugm at gmail.com> wrote:
I would say that in the US you have CHM and the
LCM which exhibit working
mainframes. There are also MARCH and the New Jersey (I think) Museums which
show working machines. In the UK several working exhibits have been
"mothballed". The Science Museum has discontinued Pegasus demos, my project
to restore some of the Pegasus i/o equipment at MOSI has been suspended,
and
the Hartree Differential Analyser is to be removed from display. Personally
I would rather that money was expended on keeping real mainframes running
rather than building replicas,
I also note that the Baby Replica at MOSI is now around 16 years old, it
first ran in 1998. Its almost an artefact in its own right...
Dave Wade
G4UGM
--
Ian S. King, MSIS, MSCS
Ph.D. Candidate
The Information School
University of Washington
An optimist sees a glass half full. A pessimist sees it half empty. An
engineer sees it twice as large as it needs to be. -----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Brent
Hilpert
Sent: 30 November 2014 07:48
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: EDSAC lives
So, we have the ABC, Colossus, Manchester Baby, and now the EDSAC.
Anyone for the ENIAC? Univac I? IAS machine? Whirlwind?
How come 3 of the 4 are in Britain?
On 2014-Nov-29, at 9:12 PM, John Foust wrote:
> The National Museum of Computing unveils EDSAC re-creation:
>
> http://www.bbc.com/news/technology-30131447
30 Nov
30 Nov
12:02 p.m.
Anyone for the ENIAC?
I'm very friendly with the Mauchly / Eckert families and I know for fact
that they want to do something similar.
But there are the usual obstacles: space and money.
Their primary goal is to gather all the remaining ENIAC pieces (which
are curently are various museums around the world) and replicate the
missing pieces. However they would also be happy with a full-scale replica.
We do have the space at the InfoAge museum here in NJ; naturally the
families would prefer a location in metro Philadelphia.
1 Dec
1 Dec
12:50 a.m.
On 2014-Nov-30, at 12:02 PM, Evan Koblentz wrote:
If someone has a spare 2M$...
I do wonder about the logistics one would run into doing a machine the size of the ENIAC.
Maybe there's more NOS stuff out there than I realise, but sourcing 18000 tubes and
associated components is different than a few hundred. Granted, you don't have to
build out the full complement of accumulators.
Didn't someone have 2 of the accumulators powered up, or am I completely
mis-remembering something?
Anyone for the ENIAC?
I'm very friendly with the Mauchly / Eckert families and I know for fact that they
want to do something similar.
But there are the usual obstacles: space and money.
Their primary goal is to gather all the remaining ENIAC pieces (which are curently are
various museums around the world) and replicate the missing pieces. However they would
also be happy with a full-scale replica.
We do have the space at the InfoAge museum here in NJ; naturally the families would
prefer a location in metro Philadelphia. 
6:31 a.m.
I do wonder about the logistics one would run into
doing a machine the size of the ENIAC. Maybe there's more NOS stuff out there than I
realise, but sourcing 18000 tubes and associated components is different than a few
hundred. Granted, you don't have to build out the full complement of accumulators.
Building a complete ENIAC replica would indeed be hard. If only exact
proper NOS parts were used, I might just say impossible. As always, it
seems, one can blame the capacitors. Even the micas from the 1940s -
and I mean the good mil spec ones - are starting to fail in large
numbers. I am not sure what ENIAC actually used (my guess is a mixture
of papers and micas), but I might thick modern types stuffed into
vintage cases would be the only reasonable route. Getting NOS paper
and micas from the era is not hard, but they are at best only a step
or two away from the trash can.
The tubes might not be too bad to source, if the project could stand
using used, tested pulls. I do not think any of the tube types in
ENIAC are rare types (if someone out there has a list of the 20 or so
types they used, please post it), regardless of what the audiophiles
think. I might even think that I could even source a majority
percentage out of my stock.
Original wire would be hard to stock in such quantity, but there are
some vintage reproductions that could be obtained. Sockets, resistors,
connectors - not much of a problem, given time.
Finding original relays might be very difficult to find in NOS, or
even good, used condition. I do not know what type ENIAC used, but
finding exact vintage replacement relays for any project can be a hair
pulling experience. If the relays needed are "telephone" types, that
would be a plus, as modifications could be made due to their modular
nature. I would not want to be the guy in charge of that project.
Transformers - that may be where all the money goes. Once again, I am
not sure what they used, but big plate and filament transformers are
getting quite pricey. I suppose there may be luck involved if the
ENIAC used massive transformers for the plate voltages for the
thousands of tubes, as the audiophiles would have no need for those,
but if used off the shelf 100 or 200 Watt or so - ouch.
--
Will
3673
days inactive
3674
days old
6 comments
5 participants
participants (5)
-
dave.g4ugm＠gmail.com -
evan＠snarc.net -
hilpert＠cs.ubc.ca -
isking＠uw.edu -
wdonzelli＠gmail.com