The Courier-Journal from Louisville, Kentucky • Page 209

Input your move" gw ever it put the bishop. In any case I continue on my merry way up the center of the board with pawn to queen four and the machine duly takes my pawn with one of its own and then I do knigh't-takes-pawn, triumphantly, and wait for the machine to catch on to the fact that it is going to lose a knight on the next move no matter what it does. And what it does is what I hoped it would do: it brings its king up to queen two in a desperate effort to protect the knight. Only the protection doesn't work; I rip off the knight with mine, and now its queen is pinned by that rook I've had waiting for long time. I am beginning to exult: I've got you, you transistorized s.o.b.! Now if this were a fable for the instruction of the young and prideful I would at this point proceed to get thoroughly whipped, since I have hubris practically coming out my ears.

If this were Greek tragedy the machine would actually be wired in to the oracle at Delphi and would now come up with some impossible set of moves, right out of left field, and humble me. But I am here to tell you that is not what happened. Not at all. Our contemporary deities can be embarrassed. On move 19 I ripped off its queen and the screen lit up with: AM IN CHECK." And how sweet it was! To have a er to waste a move.

And a small light is beginning to dawn in my head; I am beginning to penetrate a possible combination of moves in the future, if I can just take the time to put a rook on the queen file. On my 1 2th move I do just that. A couple of pieces one of my pawns and one of its bishops are between my rook and a possible pin of its queen. I wonder if the machine notices this. And then, on move 15, I push my pawn to king four, starting my tricky little combination and attacking the bishop that is in my way.

I wait impatiently to see what the computer will do. This is my best move of the game, innocent though it may look, and I study the board, wondering what the computer will come up with. And what it eventually does come up with is a total surprise. It must have panicked or something, because it flees with its bishop all the way back to the first rank, which is pretty dumb. It looks as if it has tried to hide the piece I was attacking.

I look over at Ken. "What happened?" "Damn if I know," he says. "Sometimes it does things like that." Actually the move, although bad, is not a disastrous one. In fact I am a little peeved that the computer made it, since the combination coming up for me would still work wher computer on the run a 20th-century circumstance all too rare. And I kept it in check, ignoring a rook that I had hurdled and going for the jugular vein.

And sure enough, after five checking moves I pushed a humble pawn to king knight three and delivered the death blow. And the screen lit up with: "I AM CHECKMATED." I sighed and leaned back in my chair. About five computer experts were analysing the game, talking about the game trees and the search procedures and inputs. I didn't listen. Not since the time when, as a UK student, I had hit the pinball at the Paddock Club in Lexington for 260 free games had I felt so good.

The following day was, for me, an anticlimax. There were a good many luminaries in the room John Breckinridge, U.S. representative from the Sixth District; Dr. William Ekstrom, vice president of the University of Louisville; several other important people to help in thejudgingand to watch while U-KILL silently played K-CHES 5. My main interest in the game was, I blush to admit, that I wanted to see K-CHES 5 beat the UK machine so that I could, in a sense, feel myself master of both.

I settled back in one of the black theater seats and watched the moves as a judge made them from the computers' dictates, with big cardboard pieces on a vertical board the sizes of a home movie screen. The game took more than four hours. There was trouble with the telephone lines again men were still working on them in Jett and elsewhere and both machines seemed to have embarked on a kind of attack-first-and-think-later brand of chess that led them to chasing one another all over the board until finally, on the 50th move, Ken Presley, with K-CHES 5 down by a rook and with no chance of winning, resigned the game for his machine. Applause broke out. A silver trophy bowl was awarded to the UK team, to be engraved, WINNER, FIRST ANNUAL COMPUTERIZED CHESS TOURNAMENT, STATE OF KENTUCKY.

Nobody offered me a trophy, but' I felt fine anyway. rj out that it will be almost the other way around. By the time the game is over I am to have spent only about a fifth as much time doping out my moves as the computer has spent on its. So we watch the screen as it begins to list possible moves before the machine will decide, on the basis of a complex arithmetic, which is the best one for it to make. Eventually the machine finishes its cognitions and prints: "THE COMPUTER MOVES N-KB3." (Actually it doesn't say "N-KB3," but gives a string of figures that Ken translates into that move.) That is precisely the orthodox move for the situation, and I reply with equal orthodoxy, wondering if the computer can independently reason out White's responses to the Sicilian the way it has taken grandmasters years to do.

I move my knight to KB3 and wait. And, sure enough, four minutes later the screen lights up with: "THE COMPUTER MOVES P-Q4. INPUT YOUR MOVE." RlGHT ON the button! I am getting a little apprehensive. Having given up on Durkin's Attack a few minutes before the game started, I really have no plan now except to wait and hope the computer makes a dumb move. But the computer isn't looking very dumb yet.

I look up at Ken. "What do you call your program?" "K-CHES 5," he says, impassively. I look over at Dana. "What about the UK program?" "It's called U-KILL," he says. That sounds more like it, I think.

Maybe Dana's mild manners conceal a Bobby Fischer killer mentality. I turn my attention back to the board, and take the pawn. The game continues, with those damnable four-minute waits, in wholly orthodox fashion, with the machine making no bad moves at all, until its eighth move, when it advances a knight too far, to QN5. I push a pawn up to chase the knight away. The knight is duly drawn back to where it had come from and I finally begin to feel good; I haven't yet got any real advantage, but I've caused the comput For chess players, a move-by-move accounf of writer Tevis' duel with the K-CHES 5 computer: K-CHES 5 Tevis K-CHES 5 Tevis 1 P-K4 P-QB4 P-QR3 Q-R4 2 N-KB3 N-QB3 15 K2 P-K4! 3 P-Q4 PxP 16 B-Nl P-Q4! 4 NxP N-KB3 17 PxP NxP 5 N-QB3 P-KN3 18 K-Q2? NxN 6B-K3 B-N2 19 QXR QxO 7 P-KB3 O-O 20 KxN -K5 8 N(3)-N5? P-QR3 21 B-Q4? qb 9 N-QB3 P-Q3 22 K-N3 B-K3 10 P-QN3 Q-R4 23 P-B4 Q-N7 11 Q-Q3 NxN 24 K-R4 Q2 12 BxN R-Ql 25 K-R5 -N3 Mote 13 P-QN4 OxP 5 'tat THE COURIER-JOURNAL TIMES MAGAZINE.