Commentary for white move 31: The gentleman's question. I tend to see chess as an evolving game and I tend to say Garry Kasparov as a world champion that stands on the shoulders of the previous world champions. And I see from Wilhelm Steinitz and you might even say Paul Morphy before him, that chess evolved and got better and better and better and better. And I think that each world champion has brought to the game new information, new knowledge, new styles, new possibilities. And that Garry is, in a sense, if you will, the ultimate achievement of this line of world champions. And what Garry has done in the world of chess is phenomenal. He's literally won every challenge that he's ever faced both from a human and, so far, from the computers. From Deep Blue's perspective, and its historical achievement, I think -- personally I think it's as breath-taking as what Garry does. As an International Grandmaster, I know how really, really difficult the game of chess is to master. It's a very simple game to understand and learn, but to master requires an enormous amount of work and dedication. And to get the computer to do it to me is just a great achievement, and Mike, as somewhat of a computer expert yourself, maybe you can talk to Deep Blue's achievement. I think it's phenomenal.

MIKE VALVO: Oh, I think so, too. It's the most wonderful achievement by computers to date by far, and it's introduced the parallel processor approach, which has been the next step above some of the things that have occurred before. There's been incremental improvements with software thoughts like the null-move concept, or extensions --

YASSER SEIRAWAN: What's the null-move concept, by the way? I heard Joel mention it. What does that mean?

MIKE VALVO: Well, that assumes that if you could make two moves in a row, what would your position look like, if you ignoreed your opponent's reply? And they actually have that algorithm actually built into every program. Although I'm not sure Deep Blue has it today, but most micro -- computer programs do have that.

YASSER SEIRAWAN: Well, to be clear, in this particular position, if you say Yaz, give me two moves, I take your king. I'm just that kind of person. (Audience laughter.)

MIKE VALVO: Well, they do think of it that way.

YASSER SEIRAWAN: Really?

MIKE VALVO: Yeah, they try to see if I had two moves in a row -- in other words, what is my threat.

YASSER SEIRAWAN: Okay, what is my threat.

MIKE VALVO: That's another way of saying what is my threat. And they actually find that they can get a more accurate assessment of the position by using that algorithm. And there's extensions, too, we mentioned that yesterday where you just follow all the captures to their logical conclusion. If I take this and he takes back, can I take this and he takes back in well, there's no more captures, what happens? And that gives the computers an advantage. But Deep Blue has taken parallel processing, and they're not the first. There's been a number of efforts, but they've taken it to the farthest extension, and they're clearly the strongest machine in the world today.

YASSER SEIRAWAN: Well, okay, we sit here and say Deep Blue -- and I understand parallel processing and 200 million moves a second, and that's quite awesome, and I say to myself, "Well, Garry is the world champion, and it may happen sooner than I think it will, that the world champion is a computer. Why -- will it be because of this incredible calculating speed just overrides the human intuition, or do you believe that the computer will ever bet the world's best? Full stop. Why would the computer beat the world's best human?

MIKE VALVO: I think it's inevitable that eventually a computer will beat the world's best on a regular basis. Will it be Deep Blue? That I'm not sure. Will it be the next generation? I'm not sure. I think eventually it's going to happen. Will it be from butte force? In other words, just faster and faster hardware. I think it's going to take more than just faster and faster hardware, although hardware alone will do it, but I think chess knowledge is an ingredient. And the faster machines increase the possibility of including the other Shannon type of programs. Shannon invented computer chess. He had two types. One was brute force, and the other was selective search. And selective search is, we look at all the move possibilities and we exclude some and go with the best moves. That really hasn't been used too much in these days. Microcomputers use it a lot, but the bigger machines don't. I think we're going to see a lot more of that coming back because with faster machines it becomes more possible to use that approach.

YASSER SEIRAWAN: All right. We have a question that we'll take from the audience. We'll get a microphone to the gentleman in the room there.

AUDIENCE MEMBER: Thank you. Trying to continue along the same line of your commentary, I think there's been a move.

YASSER SEIRAWAN: Yes. Thank you for -- and we have two moves, in fact. Garry Kasparov has activated his bishop with the move Bf8-e7.

MIKE VALVO: And Deep Blue guessed it.

YASSER SEIRAWAN: And Deep Blue guessed it, with the move Bb6-c5. And Kasparov has set to think. And I haven't been following the times too well, but we just reached our 32nd move, and the time in the game --

MIKE VALVO: Garry has about 50 minutes left, I think.

YASSER SEIRAWAN: All right. And we'll go from there. Sir?

AUDIENCE MEMBER: Thank you. I just wanted to continue along the lines of commentary, both thinking about the future of computer chess rising to the front. The question I would have is do you have a sense for how many Elo rating points stronger Deeper Blue is than Deep Blue was a year ago?

YASSER SEIRAWAN: We do have a time by the way, 48 minutes for Garry. An hour and 14 minutes for the computer.

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