Cyhist Apr 25 1997 B
Date: Fri, 25 Apr 1997 19:21:30 GMT
Reply-To: chris@envex.demon.co.uk
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: "Chris.P.Burton" <chris@ENVEX.DEMON.CO.UK>
Subject: Re: Colossus etc
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Mike O'Brien gave a very nice description of the excitement of being in the Colossus room at BP. Just a couple of details to correct - Tony Sale no doubt would prefer his surname spelled without the 'y'; the other point is a little more technical. The clock signal from the paper tape actually steps on gas triode shift registers which hold the putative lug settings on the Lorenz rotors. The resulting bit-streams are correlated with the paper tape channel bits, and the decade counters are used to count correlation 'hits', i.e. to accumulate the statistics to help the analyst home in on the likely wheel settings.
In answer to Padgett, the valves which Tony is looking for are primarily GT1C gas triodes and EF36 and EF37 pentodes, in hundreds. Also 6J5G, not GT.
--
Chris P. Burton, Wern Ddu, Llansilin, Oswestry, Shropshire, SY10 9BN, UK Tel+fax: +44 1691 791274 A member of The Computer Conservation Society
______________________________________________________________________
Reply-To: chris@envex.demon.co.uk
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@MAELSTROM.STJOHNS.EDU> From: "Chris.P.Burton" <chris@ENVEX.DEMON.CO.UK>
Subject: Re: Colossus etc
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Mike O'Brien gave a very nice description of the excitement of being in the Colossus room at BP. Just a couple of details to correct - Tony Sale no doubt would prefer his surname spelled without the 'y'; the other point is a little more technical. The clock signal from the paper tape actually steps on gas triode shift registers which hold the putative lug settings on the Lorenz rotors. The resulting bit-streams are correlated with the paper tape channel bits, and the decade counters are used to count correlation 'hits', i.e. to accumulate the statistics to help the analyst home in on the likely wheel settings.
In answer to Padgett, the valves which Tony is looking for are primarily GT1C gas triodes and EF36 and EF37 pentodes, in hundreds. Also 6J5G, not GT.
--
Chris P. Burton, Wern Ddu, Llansilin, Oswestry, Shropshire, SY10 9BN, UK Tel+fax: +44 1691 791274 A member of The Computer Conservation Society
______________________________________________________________________