LEO (1950) - test-drb@ccmp.vtda.org

18 Jul 2007

...
  Date: Tue, 17 Jul 2007 20:43:50 +0100
 From: "Rod Smallwood" <RodSmallwood at mail.ediconsulting.co.uk>

 I happened on the program by chance. It might have been called
 Disappearing Britain. A lot of it was contemporaneous and could have
 been a Lyons publicity film. (or at least bits of one). I suspect our
 friends at the National Film Archive might know. 
I suspect it is a contemporary promotional film on the Leo II dated  
1957 and mentions that it has been working since 1953 - referring to  
Leo 1 I think, it is black and white and if that's it, I have it on a  
DVD published by Buzz KnowledgeWorks which I bought on eBay.

...

 I inadvertently made an ambiguous statement. By commercial I meant its
 use, not its availability for sale. 
I knew what you meant. Most early designs were one or two off  
scientific machines, price no object.

...
  Where would you start to design such a thing? Valves
yes.. 12AT7
 Bistables as binary counters. Neon devices such as dekatrons as  
 decimal
 counters. RVL (Resistor Valve Logic). Storage = Ferrite Cores, Tape,
 Drum possibly. 
I'm not sure many commercial machines were made with valves logic and  
a main memory of core. Transistors and core main memory often came in  
together. The ICT1200 series (AKA Hollerrith Electronic Computer) was  
drum main memory and valves but its replacement the ICT1300 series  
used "high speed switching transistors" and core memory with backing  
store of drum and optionally tape.

...
  I once saw a Univac FAST RAN Drum memory. What a
lump!!! 