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: -- 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.

On 2014-Nov-30, at 11:07 AM, Ian S. King wrote: 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.

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