November 18, 2019 6 min to read
The Chess Machine
Category : Games
Modern computers are multi-purpose devices: your cellphone can do everything from telling the time to watching live video from across the globe, depending on what app you run. It wasn’t always that way, though: some of the first home computers were built and programmed to perform only one task, such as playing chess. These chess machines were amongst the first computerized gadgets to find their way into many people’s homes.
Invisible Chess Players
The basic idea is simple: a small chess board and set of pieces that you could move, and an invisible opponent programmed into the device. You moved the pieces and the computerized opponent indicated how it wanted to move by blinking LEDs on the board.
The hidden opponent was a computer dedicated to the task of playing chess, programmed with the rules of the game and a way to search for possible moves and evaluate each to decide which is the best.
I remember playing with these in the late 1970s and 80s in electronics shops, sneaking in a quick game before the sales assistant would shoo me away in favor of someone who might be able to afford one of these pricey devices.
The algorithms that made these choices had been the subject of computer science research since computers had first begun. Alan Turing and David Champernowne had created one of the first chess programs called Turochamp in 1948, although he had to wait until 1951 for there to be a computer fast enough to run it: the Ferranti Mark 1. Unfortunately, the original code for Turochamp was not preserved, but a recreation was beaten by Grandmaster Gary Kasparov in 16 moves at a conference in 2012.
Over the years, computer scientists developed and refined the algorithms behind chess programs, either using algorithms to pick and evaluate future moves or applying brute force to examine every possible future move. By the late 1970s, these approaches were sophisticated enough that they could beat many human players, and computers were fast enough that they could be run on a relatively small, cheap computer.
These two factors came together in 1977 with the launch of the Chess Challenger by Fidelity Electronics, the first dedicated chess computer. This device was built around a 2MHz 8080 CPU, an enhanced version of the original 8008 CPU produced by a small company called Intel. This was combined with 2Kb of ROM and 256 bytes of RAM.
To play, you entered a move by using the letters and numbers on the chessboard to identify the origin and destination of the move. So, to move your king’s pawn as an opening move, you would press 5, then b, then 5 and d, then the EN key to set the computer working. It would respond with its move in the same format.
For some reason, this way of indicating moves was reversed from the format used by most chess players, which caused some confusion. The chessboard itself was for the convenience of the human player: there were no sensors or other parts under it.
The Chess Challenger 1 didn’t play a great game and was mainly designed for beginners or young players. Fidelity did offer an upgrade, though: if you sent the chess machine back to them with a check for $75 after June 1977, they would upgrade the machine to play a better game by replacing the motherboard. They also included a set of stickers to reverse the notation to the more standard algebraic notation for moves.
The Chess Challenger was the first of many chess machines launched over the next few years. As these devices became both more sophisticated and more mainstream, the manufacturers found new ways to make them more friendly for users. To appeal to more users, some models used blinking lights to show the moves of the game. Some even moved the pieces themselves, like the 1983 Milton Bradley Phantom, which was sold in the US as the Milton Bradley Grandmaster.
Another robot chess machine was the Mephisto Phantom, which used magnets under the board to drag the pieces around in a way that made the board look rather like it was possessed.
One of the strangest chess machines was the Novag Robot Adversary, which used an articulated arm to pick up and move the pieces around the board. Magnetic sensors in the board detected when you picked up and moved the pieces, so there was no need to enter moves into a keypad.
A huge number of chess machines were launched over the next few years, from portable devices for playing on the bus to fancy models that used the latest chess programs to challenge everyone from beginner to grandmaster. These were big sellers: the Fidelity Electronics Chess Challenger 7 sold over 600,000 units in the US.
The Battle of the Bits
These machines would face each other in the annual World Computer Chess Championship. Held since 1974, this championship was dominated by chess programs running on high-end mainframe systems. To test the mettle of chess machines, the championship added a class for chess machines in 1980. This was dominated by Fidelity Electronics Chess Challenger models for the first few years, but the European Mephisto chess machine took the top spot from 1985 to 1990. However, after 1990, the winners were all chess programs running on personal computers.
That shift was because chess machines were facing this new enemy: the home computer. The new breed of computers built around fast chips like the Intel 486 had more computing power than chess machines, and they were not that much more expensive. They could do other things too, like play other games. As the cost of these computers continued to fall, it became harder to justify buying a dedicated chess machine: why buy one when you could spend just a little more on a PC that could do so much more?
This meant that chess machines had to go one of two ways: very cheap or very expensive. At the high end, the last chess machine released by Fidelity Electronics was the 1994 Fidelity Elite V11, which used a 75Mhz Motorola 60860 processor, a similar processor to the ones used in Apple Macs of the time.
At the cheap end, chess machines became educational, cheap or portable devices. The pages of gadget catalogs that had been filled with chess machines were now filled with computers.
In the end, chess machines were the victim of the thing that created them: the revolution in cheap computing of the 1980s and 1990s. The very same cheap processors that made them possible killed them off when they started appearing in home computers that were far more flexible.
The Turk: The Life and Times of the Famous Eighteenth-Century Chess-Playing Machine – Tom Standage. An excellent book on the history of the famous fake mechanical chess player of the 18th century.
My Chess Computers – A great resource for finding details of old chess computers. Not updated anymore, but it has a lot of great information.
Chessprogramming.org – Has lots of info on the different models of chess machine available, and the algorithms that they use.