Commentary for black move 26:

YASSER SEIRAWAN: Well, let's take a look at what Garry has done. He's played the move Nh7-f6. So he's inviting Deep Blue to trade pawns in the center with f4xe5, d6xe5. As a result of this exchange the f-file has opened up and it's conceivable that Garry would try to bring a knight to the d6 square or certainly put a bishop on the d6 square. So Deep Blue isn't going to see a lot of advantage from trading the pawns, but you know Deep Blue as well as anybody. Would you expect that Deep Blue would try to play, for example, the move f4-f5, gaining more space on the kings side, or do you think he will keep the tension? JOEL BENJAMIN: Well, I can never be sure what it's going to do. There is some argument to take on e5. But since the rooks are already operating well on the a-file, I don't think that it would go for a plan of switching the rooks back to the f-file, so that opening the file wouldn't get much benefit for it. And it does understand things about block aids, so I think there's a decent chance that it might push f5, because it introduces stuff that it knows about as well.

YASSER SEIRAWAN: More space and so on? JOEL BENJAMIN: Yeah.

YASSER SEIRAWAN: I'm going to take a question from the audience soon. I just have one comment. Garry has now won three games running. Games five and six from Philadelphia, and yesterday's game. The gentleman in the audience earlier had said, have you made some changes as of game one, and I know you don't want to do that because like you said it could affect other things in the program. But three games running gives you pause for thought. Are you on the wrong track, or does the team -- are they having a rethink at the moment? What's going on? JOEL BENJAMIN: Well, first of all, you can separate out the first two games from the sequence because it was already completely --

YASSER SEIRAWAN: Ancient history? JOEL BENJAMIN: -- a different program, situation. So really, Deeper Blue has lost one game running to Kasparov --

YASSER SEIRAWAN: Deeper Blue. JOEL BENJAMIN: And while there are some changes that can always be made, I don't think there's any reason to panic just yet.

YASSER SEIRAWAN: In the audience, the gentleman in the green. Yes.

AUDIENCE MEMBER: Well, after f4 I was looking at maybe Qf2 and Bb6 since the bishop on f8 is kind of weak, to activate that diagonal?

YASSER SEIRAWAN: The gentleman was suggesting that white may play the move Qf2 with the idea that he could introduce his bishop with bishop b6. And I agree with you. The move Qf2 is a very flexible option, not bad at all. Not bad at all. Good suggestion. JOEL BENJAMIN: Yeah, it's certainly possible. We've only talked about so far that white could take on e5 or push on f5 but the other possibility is that it will keep the tension and could very well do that.

YASSER SEIRAWAN: Maybe this is confidential, but prior to the match a lot of people were making their own predictions. I thought that Garry still held a significant edge. We do have a move, by the way, f4xe5. Deep Blue has resolved the tension. We expect that Garry will quickly recapture with the move d6xe5. Which he has now done. What was your prediction -- what was your feeling before the match? How did you expect Deep Blue would do in the match? JOEL BENJAMIN: I expected it to be a very close match. I thought that the final score would probably be 3 1/2-2 1/2 in somebody's favor. And I thought that probably Kasparov should be rated a slight favorite, but that he would actually be considered by most people to be a heavy favorite. But basically outside of me and a few other people, nobody really knows going in how good Deep Blue is. So any kind of predictions are absolute guesswork.

YASSER SEIRAWAN: And what would be, in your opinion, the pivotal points? Would it be the choice of openings that would make all the difference? Would it be tactical? Or how did you imagine a victory occurring for Deep Blue? JOEL BENJAMIN: I think probably the most critical question is getting into the right types of positions. Because when Deep Blue gets into a position where it has a lot of activity, and it's difficult to find moves for the opponent, then it can be devastating. There's no question that there are some positions that Deep Blue plays better than Kasparov. If it gets to them, it will win the game. On the other hand, if Kasparov gets to the positions where -- that he plays much better than Deep Blue, then he will win.

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