jbn@glacier.STANFORD.EDU (John B. Nagle) (01/17/89)
In article <8486@ihlpl.ATT.COM> knudsen@ihlpl.ATT.COM (Knudsen) writes: >Neat idea. I just happen to own a pair of what must be the >oldest type of DAC in existence. Maybe. In the London Transport Museum at Covent Garden, there is a device used to control a station display indicating the destination of the next train. It consists of a manual control box with a rotary dial and lever, on which an operator indicated the train destination when the train passed a signal box some distance up-line from the station, a 4-bit shaft encoder in the box, a 4-bit by 32 shift register used to store the train identification data, 1 of 16 decoder used to decode the output of the shift register, and a sign with illuminated sections for each possible destination. All very straightforward digital logic. But it's all implemented with 1890s technology. The shaft encoder has huge brass cams. The shift register is an electromechanical device about four feet high, using a drum with pins depressed by solenoids as the storage medium. The drum is powered by a large weight on a rope, as in a clock tower, driving a gear train and an escapement released by a solenoid. The 1 out of 16 decoder is a beautiful device of green solenoids nearly six inches long driving big brass switches, all mounted on a polished wooden panel; the four stages of decoding are laid out clearly for all to see. The elegance of this device lies in its clear enunciation of the basic principles of binary representation. One doesn't usually see explicit binary representation in devices of that era. The Duo-Art is one of the few others of which I am aware; other attempts to perform the same function in player pianos were usually analog devices. John Nagle