Commentary for white move 45:

MAURICE ASHLEY: -- Deep Blue would have been kicking it out. So Deep Blue realized quickly there was no mate, there was a perpetual check. Let's avoid that, get back to the positional advantages, like you said, and like it seems to us now, this is -- it's playing wonderful positional moves and ideas, and maybe now it's pushing Kasparov over the top. Finish your story.

YASSER SEIRAWAN: Finishing the story, I'm playing this computer HIARC, I'm a pawn up, life is good. Then comes this devastating counter-attack, and sure enough, it's perpetual check. Well, we were playing a time control where I get a bonus for every move a make. So I make a couple repetitions, and the operator says, "

YASSER, would you like a draw?" And I saw this opportunity with my extra pawn to skedaddle out of the way, and I said, "No," and I moved my king and the operator said, "I'm sorry, but it's checkmate in seven." About that -- "About that draw..." So yes, you're absolutely right, had Deep Blue perceived checkmate that would have been instantaneous.

MAURICE ASHLEY: Yeah, it would have taken about two seconds to see that, because 200 million positions is -- mate is so simple. All you have to do is cut off the king's running squares and the moves are so direct, that I wouldn't want to guess but it's got to be less than a thousand positions you'd have to look at in order to confirm a mate, or certainly a lot less than 200 million. If it could do that in a second it would have gotten the mate like that.

PATRICK WOLFF: Let's see if we can understand this position. I think we agree black has to capture the queen. The only possibility I see is some sort of counter play like this but after queen takes bishop there is not going to be any sort of perpetual check and the king is in very bad shape, so instead it looks like black is going to have to capture his queen and then white will capture with the pawn.

YASSER SEIRAWAN: d5xc6, absolutely.

PATRICK WOLFF: Now what has happened? The pawn on c6 -- well, if

YASSER was using the phrase a bone in the throat, this is a very, very powerful --

YASSER SEIRAWAN: A large bone. Part part large and well placed bone.

YASSER SEIRAWAN: I think that the ending, when the pawn is on d5, the ending with the pawn on d5, black has excellent holding chances. With the pawn on c6, I think he's gone. This is a very bad pawn. Because with a pawn on c6, white's bishop will be able to go to the d5 square check. The e6 square, and then when he moves Ra7 and c7 he will just win a piece and the game. So after this what appears to be a forced exchange of queens, I don't see how Garry is going to defend.

PATRICK WOLFF: Exactly. Let's take a look at Rc8. Now, the idea of Rc8 is white was threatening this check. Now, if that check should happen, black can block it with the rook. But I think we'll see that there's still a lot of trouble for black here. Ra7+ I think is still very strong. Now, Bc7 is very bad because white plays Rb7. Black cannot defend like that, because that guy goes. So after rook b7 this b5 pawn is in bad shape. Therefore, after Ra7+, Rc7 is pretty much forced. But now we play Ra8.

YASSER SEIRAWAN: Ra5.

PATRICK WOLFF: Also Ra5, good point. Itself going to -- I was going to suggest that black would be tied up with Ra8, but that would be bad enough, but

YASSER is correct.

MAURICE ASHLEY: And looking at Garry Kasparov now, he does not look like a happy man. He he's all these variations --

MAURICE ASHLEY: Garry Kasparov has resigned the position. (Audience applause.) (Long and sustained audience applause.)

MAURICE ASHLEY: We were indeed impressed by the way the computer handled this position. We hope, we'll cross our fingers, but don't count on it that Garry will come on stage and explain why he lost. I don't think so.

PATRICK WOLFF: I think the computer guys will.

YASSER SEIRAWAN: I think he missed the move Ra6, Ra6 forcing the exchange of queens and that endgame is gone.

MAURICE ASHLEY: And I tell you again, I saw the theme in this game, it was the computer playing control chess all the way, no counter play, no chances, moves like Be4 looked a little bit like maybe white would get aggressive, it just mounted and mounted.

PATRICK WOLFF: I'm sure you'll agree. This was a very, very impressive game by the computer. The computer really played a very high-level game.

MAURICE ASHLEY: We'll ask them to come on stage and explain why their baby won. If you want to stick around for that, I advise you to stay because in a few minutes I'm sure they will be down on the stage.

YASSER SEIRAWAN: We're in the midst of asking somebody from Deep Blue, perhaps C. J. Tan. Right now there seems to be a little celebration going on up on the 35th floor. I do believe that we will get somebody. I don't know what they can say other than the fact that they played a great game.

PATRICK WOLFF: Absolutely.

MAURICE ASHLEY: I'm sure they'll have lots to say. What was interesting to me -- well, first of all, history has indeed repeated itself. Last year, the player with white won the game. This year, the player with white won the game, the first game, and in the second game, the loser bounced back, had white and won, and we've seen it happen again. This time Kasparov was in the reverse position, the enviable position of having lost a game when he thought he was in control of the match after the first game, it looked like he had worked out the kinks. But this was masterful for the computer. What is it going to take from this step to say, man, this thing can play some chess?

PATRICK WOLFF: I think this game serves to highlight that the advantage of having white is going to be a very big advantage in this match. If things keep going like this, we can expect a very seesaw battle all the way to the end. This is going to be a very close one.

MAURICE ASHLEY: Well, a tremendous development over the weekend. I don't think we would have dreamt of it like this. The story went that if Kasparov won game one, no one would show up for the rest of the games because we thought that Garry was just going to rule, and that streak of victories on Kasparov's part has been broken, and now Deep Blue has won a second game, a second game against the world champion, and not many human players can say that. We are anticipating -- This is like -- what an exciting moment for the Deep Blue team.

YASSER SEIRAWAN: And again, I think we're about to be joined by Mike Valvo -- no, we're not?

MAURICE ASHLEY: We would like to thank Grandmaster Patrick Wolff for his commentary. (Audience applause.)

MAURICE ASHLEY: Mike Valvo is back. Mike, what's it been like upstairs? Have you got a sense of what's going on especially after the stunning reversal, Kasparov resigning?

MIKE VALVO: It seems like people are very excited that we really have a match now, and most are astounded that the computer could play this position so well and not do well yesterday. It just amazes them. Actually some of the Grandmasters seem to be in fear that he could play like this.

AUDIENCE MEMBER: Would you think this is the best game the computer has ever played.

YASSER SEIRAWAN: The question is, do you think this is the best game the computer has ever played?

MIKE VALVO: Several Grandmasters were astounded by this game. It's clearly the best game ever played against Kasparov by a computer.

MAURICE ASHLEY: It's a gorgeous game. When you look at it, it's power chess, control chess, all the way down. We may have critiqued a couple of moves --

YASSER SEIRAWAN: I would be proud to have this one.

MAURICE ASHLEY: Absolutely. I don't think if you put this game -- if you gave this game to a Grandmaster who was asleep and he woke up and said okay, take a look at this game and we're not going to tell you the players, I don't think any Grandmaster would say, "It's obvious that this is a computer playing the white pieces."

YASSER SEIRAWAN: No.

MAURICE ASHLEY: He would say something like "Who is playing white, Kasparov, Karpov, Portisch, Smyslov, maybe

YASSER Seirawan."

YASSER SEIRAWAN: Yeah, sure.

MAURICE ASHLEY: But certainly you could not guess, you could not dream, with moves like Rfc1 we saw earlier, ultrapositional ideas, the doubling of the rooks, the play on both fronts, thinking so globally, getting a weakness, getting the passed pawn and then making exchanges and then just the suppression of the counter play. Instead of going for the fiery tactical variations, it just said, "No, no, no, you get no chances. We'll keep it nice and good and we're just going to win." I can't tell the difference between the computer and human in this game.

YASSER SEIRAWAN: That was a grade game.

MIKE VALVO: What's going to happen in the future when humans think "Well, just get the computer in a positional game and we'll have an easy time of it?" Boy, I wouldn't want to play the black side of this, would you?

YASSER SEIRAWAN: No, no, no, no.

MIKE VALVO: An anaconda type of variation.

MAURICE ASHLEY: Anaconda. That's a great word. I'm sure Kasparov felt like that, being squeezed to death in this position.

YASSER SEIRAWAN: Let's go back to what I said earlier, and that was that this particular opening choice by Garry, in the words of Capablanca, held the seeds of defeat, I don't really know where Garry went wrong, okay? Let's just think about it from that perspective for a moment. Okay, Deep Blue is playing great moves, but you're not -- you're supposed to beat a person because of your great and forcing will. The idea is that Garry should have made a mistake somewhere, but I can't put my finger on it and say, "Oh, Garry erred right here" or was it the move f6, was it the move Bxf5? Where was the error?

MIKE VALVO: And what was the winning idea? Nf5?

YASSER SEIRAWAN: Nf5 was a paradoxical move and it turned out to be a very nice one. So all in all, a masterful game, simply because I don't know where Garry went wrong.

MAURICE ASHLEY: Those of you surprised by Garry's early resignation, one thing -- for one thing usually you resign of course when you perceive it's a lost position. Many might think well, it's lost but maybe he could have played a few moves. Another reason he resigned, though, is out of respect for your opponent's abilities. And when a Grandmaster sees another Grandmaster playing such wonderful chess, playing such accurate moves you don't suppose that somehow they're going to fall asleep and not know how to win some simple technical position. And this is what happened in this game. Kasparov couldn't avoid this passed pawn on c6 and then the subsequent removal of his b-pawns and then he would have suffered what he put the computer through yesterday, those passed pawns on the queen-side. Remember Kasparov had two passed pawns yesterday made him win very easily. And what's interesting, if we go back to yesterday, the material that Kasparov had when he won the game was a bishop and a rook and passed pawns. And this is exactly what the computer has done, reversal from yesterday. It also has a bishop, rook, and passed pawns to win the game. So it's interesting that that has happened also.

YASSER SEIRAWAN: We have some visitors. Garry Kasparov has left. However accident we do have the winner's circle, the Deep Blue team --

MAURICE ASHLEY: The Deep Blue team, please, a round of applause. (Standing ovation by the audience for the six-member Deep Blue team, which has come onto the stage.) DR. C. J. TAN: It was a really exciting game. As I said at the very beginning, this is going to be a long, drawn-out match. And we'll see what happens for the rest of the four games. A while ago I almost asked Ken Thompson to come up on the stage, and Ken didn't want to come up. And as you know, Ken was a pioneer in computer chess, and as well as many areas in computer science. And this victory -- and this result just shows all the work, not only for the scientists up here, but the many computer pioneers that did all the work they've done in the past and we really feel very good. Thank you.

MAURICE ASHLEY: First of all, how does it feel? What are your emotions at the moment? You were in the room when the magic move was being made and Kasparov offered his hand in resignation. Did Kasparov's handshake at all? FENG-HSIUNG HSU: No, it didn't.

MAURICE ASHLEY: I guess you're quite proud of the year's work, 14 months -- FENG-HSIUNG HSU: We believed that the computer -- it's still quite satisfying to see it actually win.

MAURICE ASHLEY: It was stunning for us, and I keep going back to this point, that Deep Blue played such wonderful positional moves in closed positions. Did you guys tweak a bit in closed positions to improve its capabilities here where everyone thought it plays weak chess? FENG-HSIUNG HSU: We did massively rework the software and everything, and as I said, too, what happened is the last year, Grandmaster Joel Benjamin has essentially taken the machine to chess school. The machine was quite talented. It was built to be very fast, had the ability to recognize a lot of chess features, but there was no appropriate way on those features that we had last year. And this year it had a better understanding of chess and some of the subtleties of chess, and that showed up in this game.

MIKE VALVO: What was his mistake?

YASSER SEIRAWAN: The microphone is going down the line there. MR. HOLM: I don't really know what his final mistake was. It thought Kf7 was weak and the king should have gone to h7, but he was suffering no matter what he did.

MAURICE ASHLEY: Were there points earlier in the game where the computer felt Kasparov could have played differently and one move could have changed the evaluation? MR. HOLM: I don't recall a particular mistake that he lost a significant number of points in the evaluation. It just slowly went down here. A few moves that lost a few points here and there.

MAURICE ASHLEY: Joel, you're in a somewhat funny position, because you've been with this teamworking on developing it, the computer, developing Deep Blue to be as strong as it is. At the same time you're a human Grandmaster. You've been beating computers in competition a lot. What are your feelings now that this is happening? JOEL BENJAMIN: I feel great. This is what I've been working towards for eight months, and the gratifying thing about it is that this is a game that any human Grandmaster would be proud to have played for white. This was not a computer-type game. This was real chess. This shows that it's possible to get a computer to play real chess, Grandmaster-style chess, not only Grandmaster level, but the style of play of a Grandmaster, and I -- from the testing we did, I believed that it could do it, but this is actually the proof that it's capable of playing that way. (Audience applause.)

MAURICE ASHLEY: Did you get used to think -- from all the practice you had, did you get used to seeing these sort of ultrapositional ideas that it played in this game? I mean so many ideas were just not at all computer-like, but just textbook, textbook ideas. JOEL BENJAMIN: Well, the first thing I did with chess school was I said, "Okay, Deep Blue is going to have to play all kinds of positions, because we can never guarantee we're going to get what we want out of the opening. We could be in the wrong kind of position, so we have to be able to play everything." So I spent a lot of time with -- forcing it to play closed positions, playing all kinds of openings that led to closed positions. And I tested and I thought it could play this pretty well, and I think really the best thing that I did was I ultimately had faith in Deep Blue to play this position, and I was correct in placing that faith.

MAURICE ASHLEY: Do we have any questions for the team from the audience? I'm sure a couple questions. Here we have a question.

AUDIENCE MEMBER: I guess it's -- I don't know quite who to address this question to, but in light of this game, and even the quality of yesterday's game, would you rate this match at the level of like the Kasparov-Karpov matches? I mean is it going to be that kind of level of play?

MAURICE ASHLEY:

YASSER, you want to talk about that?

YASSER SEIRAWAN: Well, first of all, I would say that both victors played very well. We criticized, correctly so, Deep Blue's play in game one, but we have to remember that Garry faced an enormous challenge to win that game, and he did, and he played a superb game. This game was a superb game. So we're seeing two really, really great games. And we saw a lot of great games in that match, the one that I'm thinking of, the fifth match in particular. But one of the differences that I think the human vs. -- the Karpov-Kasparov match, is we're not seeing the human player -- the human players get nervous. Both sides get nervous, and both sides look at the match score and say "I want a draw." And you'll see a Grandmaster draw or two as everybody takes a deep breath. And I think as we saw in Philadelphia, when the Deep Blue team turned down the draw offer, they're just coming to play. They just want to play, you know, good, hard chess every game. So Garry knows that he's not going to get a respite by simply offering a draw. And so that puts a lot of pressure on him. So it's a different match. I wouldn't want to make a comparison but I would like to say that, yes, we're seeing superb chess by both sides.

AUDIENCE MEMBER: I would just like to say one other thing to the Deep Blue team, I felt I saw history today, because this game was just amazing. (Audience applause.)

YASSER SEIRAWAN: We have another question?

AUDIENCE MEMBER: Yes, I just had a quick question of the Deep Blue team, if they're willing to reveal a little information. I was curious if the evaluation of the current position of the game was ever negative and what it was at the end.

YASSER SEIRAWAN: The question was, during the game, was there ever a negative valuation. In other words, did Deep Blue ever feel itself to have a worse position at any time, and what was the final position. JOEL BENJAMIN: Short answer, no, it always felt that it had a slight advantage, until near the end when it felt it had a larger advantage, and then the score in the last few moves went very high and it was something like two pawns up.

AUDIENCE MEMBER: Excellent achievement by the way, congratulations all you guys. Great.

MIKE VALVO: C. J., I was wondering, do you think we're going to repeat the same results as last year with colors reversed? DR. C. J. TAN: Well, I predicted that yesterday.

YASSER SEIRAWAN: We have a question from David Levy of the international Computer Chess association, I beg your pardon.

AUDIENCE MEMBER: I have a question for C. J.. Yesterday the program played a few dubious moves. Today it played like an absolute genius. What did you guys do to it last night? DR. C. J. TAN: We let it have a couple of cocktails. (Audience laughter.) (Audience applause.)

MAURICE ASHLEY: We'll take one last question.

AUDIENCE MEMBER: If you win match, how will you divide money between you? DR. C. J. TAN: It goes to me, goz my project.

MIKE VALVO: Thank you very much.

Real-time text commentary is made possible by LiveNote, Inc. and Vincent Varallo Associates