Relay noughts and crosses machine

[bobsterkent]

There is a picture of one in the Hulton Getty picture library:

http://www.gettyimages.co.uk/license/HB8162-001

[Ed_Dinning]

Hi Bob, thanks for that. That is similar to the one in the Melbourne museum, possibly with fewer relays.

Ed

[ParcGwyn]

There is a noughts and crosses machine in Practical Electronics Sept 1965, available here: http://www.americanradiohistory.com/...cs-1965-09.pdf that uses just switches and light bulbs.

Dave

GW7ONS

[Lucien Nunes]

It's fascinating to see the difference in complexity between devices designed to play the same game.

Schemes using only combinational logic, such as cascaded switching, can (with enough contacts) theoretically play a 'perfect' game but have the following inevitable limitations:
Human must execute the machine moves (deliberately or incidentally).
Cannot change strategy mid-play for a given board position without input from the human operating a 'randomising' element.
Given an identical input sequence from the human, including any adjustment of randomising, strategy-obfuscating or defeating controls, must always make the same moves.

Paradoxically, some of the extra complexity in a design like Relay Moe is devoted to reducing playing performance below ideal, to improve entertainment value. The remainder provides a more engaging man-machine interface, where the human need only press one button to indicate a move, the machine then has a separate response cycle of non-zero time, the game concludes with a win/lose/draw indication etc.

I think, like Ed, we might be building a few of these contraptions for Electrokinetica. Camswitch-based and branching relay-based flow control are quite easy for the non-technical person to understand, especially if the sequence and conditions are indicated on a row of lights. Pinball is another example, where if you play strategically (e.g. light all 5 lights in sequence for a bonus) you can follow the gameplay progress on the camswitch and relays.

[Karen O]

I wonder if this is the machine I saw at Birmingham Science Museum all those years ago. The bulb-illuminated masked display certainly fits the memory I have of it. I believe this machine has been acquired by the Museum of Computing, which apparently is in Swindon (why did I think it was in Cambridge )

[Viewmaster]

Originally Posted by ParcGwyn

There is a noughts and crosses machine in Practical Electronics Sept 1965, available here: http://www.americanradiohistory.com/...cs-1965-09.pdf

I plan to build that using 18 miniature 4PDT toggle switches instead of rotary.

Also using nine 2 colour LEDs.
Red illuminated for crosses and green for noughts.

All this will make a much more compact machine.