It is interesting that while Turing is considered a creative genius at a level I am not sure how to characterize but he was certainly very important and successful in the code-breaking efforts during WW2, he worked with some of the best British chess players among them Golombek and Alexander. He would play with Golombek who I think could allow Turing to change color at a point of his choosing during the game (I am not sure of the constraints -- maybe it was up to a certain move) and Golombek could then beat Turing from Turing's presumably losing position. I have never played at those odds but I can't imagine, assuming that Golombek was playing to win that Turing had much of a position -- on the other hand, perhaps Golombek would try to keep a more or less even game so that he had chances. Anyway, I really wonder how weak of a player Turing was and if perhaps some of his games survive.

  • 2
    Only one game found from 1952 and Turing lost. – TheSimpliFire Dec 30 '19 at 8:29
  • 2
    Unclear that this a real game or one using the "program" Turing had written which, as I understand it, was just a series of rules that a player would follow simulating being a computer. It is interesting that as late as the 1970s some good players who also were computer scientists could not imagine a world champ computer chess player. – releseabe Dec 30 '19 at 8:33
  • 2
    The text indicates that it was not Turing, who played that game. I have a strong suspicion that you will not be able to get an answer for this question as I could also find no games that he played, and thus, cannot make any reasonable guess. I will say that if he were able to hang with Golombek, even for a while, he must have been a fair bit stronger than just a pure beginner. Of course, I also suspect that had he not had more important things to do, and devoted himself to the game, he would have easily been a Master, if not a lot stronger. – PhishMaster Dec 30 '19 at 11:56
  • 3
    It is shameful how the British government treated Turing, which ultimately drove him to suicide. Even if this is inflated, he, alone, is estimated to have shortened the war by two years, which saved as many as 14 million lives. He is one of the greatest men in history. What an utterly amazing man he was. – PhishMaster Dec 30 '19 at 11:58
  • 1
    I also just found something in Golombek's obituary in the "Independent" saying that he used to play Turing giving him a "queen start", aka queen odds, and "still beating him". So the question is was that early on, and did Turing improve? I suspect that many GMs could give 1200-1400's queen odds, and still win most of the time, so this does not really narrow it down. – PhishMaster Dec 30 '19 at 12:25

A similar question was asked on another forum a few years ago, and the answer that was given was "Alan Turing wasn’t a very strong chess player at all. He was probably at the level of a beginner. He certainly knew the rules of the game, but not much more."

Wikipedia notes, regarding the chess program that Turing and a colleague designed, "... the algorithm is primarily designed around the decision to take a piece or not; according to Turing, the resulting gameplay produces a low level game of chess, which he considered commensurate with his self-described average skill level at the game." According to the article, Turing's algorithm (it's really difficult to call it a "program", as it never ran on any actual computer - Turning only ever "ran" it on pencil and paper) only looked two half-moves ahead, applied a static evaluation function to each resulting position, and selected as the next move whichever one led to the highest average evaluation after the following half-move.

  • I am not sure if what is being said is that Turing was able to produce a "program" about as good as he was but that would be remarkable -- I think what might be meant is that the "program" was very weak but Turing himself was also very weak for a human -- I doubt if such a procedure for generating moves would be any better than, say, 500 rating which is what preschool kids just learning moves have. Maybe lower. – releseabe Jan 1 '20 at 9:16
  • I agree with your interpretation. The details are probably more clear in the footnoted references attached to that comment, but these refer to printed materials, not online, that I don't have ready access to (I do have Levy's book around, somewhere, but not with me at the moment...) – patbarron Jan 1 '20 at 9:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.