Mark Wells playing a game of chess with a young opponent. Courtesy photo

An example of Mark Well’s hand-written algorithm for ‘Knight’ moves of his 6×6 chess game (1957). Well’s early computer chess algorithms, including the one shown here, are now housed in the Computer History Museum in Palo Alto, Calif. Courtesy photo

By MARCELLA WELLS
Fort Collins, Colo.

Mark Wells, long-time resident of Los Alamos will receive a posthumous award from the U.S. Chess Federation based on his pioneering efforts in computer chess (c.1950s) and his lifelong love of the game.

His surviving daughters were notified of the award May 21. He will be honored in the 2020 Annual Report of the U.S. Chess Federation expected to be released online in August 2020.

BACKGROUND and CONTEXT

Mark Brimhall Wells was born the second of two children June 10, 1929 to Madelen and Arwid Robinson (“Jube”) Wells. He learned to play chess from his father as a youth.

Wells spent his career as a mathematician and computer scientist at Los Alamos National Laboratory (1951-1980) and Department Head of the Computer Science Department at New Mexico State University (1980-1989).

He played chess throughout is life winning the New Mexico State Championship in 1971 and playing in numerous other chess tournaments including the 1968 U.S. Open Chess Championship in Snowmass, Colo., where he placed 46 of 172 participants. In the 1970-80s he taught chess to inmates at the State Penitentiary in Santa Fe.

However, Well’s contribution to chess has had far greater reach than his personal chess skills. As a mathematician, he pioneered computer chess using mathematics to program the early MANIAC 1 I and MANIAC II computers in the 1940’s and 1950s. Much of that early work was solely numerical but he also worked with combinatorial problems. He recounts the story of writing a code for the queen’s problem in chess … and calculating the 92 solutions for the 8×8 chess board.

“I remember spending an afternoon in my office with the 92 machine-produced solution at my chessboard grinding out the 12 inequivalent solutions by hand, Wells said. “I believe it was my independent discovery of backtracking in the early days of Maniac that I nurtured the interest in combinatorial algorithms”.

“We had a chess-playing program on MANIAC I, however because of the slow speed (about 10,000 instructions per second) we had to restrict play to a 6×6 board, removing the bishops and their pawns.

“Even then, moves averaged about 10 minutes for a two-move look-ahead strategy.

“Perhaps the most exciting game was one played with Martin Kruskal, a Princeton physicist. Kruskal gave MANIAC queen odds. The game was a stand-off for some time; once after a surprising move by MANIAC, Kruskal even murmured, “I wonder why he did that?” In the end, however, Kriskal did win; but when he checkmated the machine at move 38, it responded with one more move, illegal of course. We were dumbfounded for a while, until we retraced the trouble and realized that the program had never been taught to resign.”

–Recounted by Mark Wells in Lazarus, R.B., Voorhees, E.A., Wells, M.B., and Worlton, W.J. 1978. Computing at LASL in the 1940s-50s. Los Alamos Scientific Laboratory, Los Alamos, NM LA-6943-H.

Well’s early work in early combinatorial computing and his knowledge of chess together helped launch algorithms that were subsequently used in computer programming. Computing is what it is today, due in large part to Well’s early work with algorithms. His love of chess lasted the rest of his life. Wells died Oct. 7, 2018.