1c4+0.2e5+0.32Nc3+0.1d6+0.43e3+0.4g6+0.54d4+0.3Nd7+0.45Nge2+0.1...An unintuitive move, blocking in the f1 bishop. White's plan is to play g3 and develop his bishop to g2. In that structure the knight is often better on e2 than f3 as it does not block the g2 bishop.5...Bg7+0.56g3+0.6Ngf6+0.47Bg2+0.2O-O+0.58b3+0.4...double fianchetto games can be fun as most of the pieces stay on the board and the game often becomes full of subtle maneuvers. If you are going to play these kind of positions it is worth studying some master games though, as the positional maneuvers to try to gain advantage can be a bit different than in more traditional setups.8...Re8+0.49a4!?+0.2...Castling looks more intuitive here. Yasser says he wanted to leave the option open to develop the c1 bishop to a3 and castle queenside should should black play ...e4. This kind of convuleted thinking shows up in my own games all the time which is probably why I enjoy Seirawan game's so much. Most people would probably be well served to just castle kingside and not worry about such subtleties.9...c6+0.2Black almost always plays this move at some point when white develops his bishop to g2, as it blunts the g2 bishop, controls the center, and takes pressure off of b7 enabling black to develop his c8 bishop.10h3!?−0.3...Again delaying the commitment of the king and the c1 bishop to see what black does. So subtle!10...exd4+0.1black finally commits to a center strategy.11exd4+0.1...11.Nxd4a5secures the c5 square for the d7 knight and gives black a good position.11...d5+0.112a5+0.2...Don't try this at home folks! White continues to delay castling and development to try to keep the black pieces pinned in, particularly the d7 knight that is blocking the c8 bishop.12.O-Odxc413.bxc4Nb6gives an active, pleasant game.12...Ne4+0.2This kind of move was very unintuitive to me in my youth, as I always liked fixed pawn structures, while black is allowing a lot of confusing dynamic transpositions for the structure. After seeing enough games like this, I have an appreciation for this kind of play by black and even do it myself. Black activates his g7 bishop, and if white captures on e4 he opens the d file to pressure the d pawn. He also frees the f6 square for the d7 knight.13Nxe4+0.1...13.O-ONdf6allow black to free his c8 bishop and get a good game13...dxe4+0.414O-O−0.5c5?!+0.4A very human move, disrupting the white center before he completes his development with Be3, Qd2, and Nc3.14...Nc515.Be3Nd3which I guess is fine, but looks very confusing to me. The knight is very strong on d3, but how stable is it? While I am sure the computer is right, in a game I would be hesitant to play this way for black.15Be3+0.1cxd4+0.416Bxd40.0Ne50.0Black finally gets the knight out of the way of the c8 bishop, and onto a nice square. It is probably objectively about equal now.17Bxe5!?0.0...Giving up the two bishops to speed up the advanced of the queenside, hoping to create a strong passed pawn over there. Seirawan really likes his knights, and is willing to do these kind of trades much more often than modern players.17...Bxe5+0.118Qxd80.0Rxd80.019Rad10.0Bc7+0.419...Be620.Bxe4Bxh321.Rfe1is probably fine but does give white some annoying pressure.Rxd122.Rxd1Rb8⩲20Rxd8++0.3Bxd8+0.221Ra1+0.2Rb8+0.4Bxe4 was threatened not that ...Bxh3 no longer threatens the white rook leaving white free to take the pawn on b7.22Nf4+0.1...again threatening Bxe422...f50.023b40.0...Seirawan's note on this position is illuminating: "This is the position for which I was aiming. If I can slip in Ra1-d1 and f2-f3, Black's bishops can easily be dominated." These positions cam be more double edged than they look though.Black's pieces look passive now, but if the position somehow opens up I can imagine the black rook and bishops running wild.23...Bd7+0.124Nd5+0.1Be60.025Rd1+0.1Bxd5?!+0.1I can only assume black was aiming for a draw, as the opposite color bishops can be drawish in many positions. Objectively I don't think this exchange makes sense though.25...Kf726.f3exf327.Bxf3b628.a6Rc8seems more reasonable, giving black some pressure down the c file. Note that black can trade his e6 bishop for the knight at any point, so there is no need to commit to it right away.26Rxd50.0Be7+0.127c5!+0.1...an instructive move, reducing the mobility of the e7 bishop.27...b6?+2.727...Rd8was black's best bet. Seirawan was planning28.Bf1Rxd529.Bc4Kf830.Bxd5winning the b pawn, but these opposite bishop endgames can be hard to win.28cxb6+2.9axb6+2.829Rb5+2.2...29.a6is even stronger. Seirawn says he had calculatedBxb430.a7Ra831.Rd7b5but missed that black does not have time to win the a pawn32.Rb7Bc533.Rb8++−This line does not seem that hard to see for me, which just shows that even world class grandmaster are still human.29...Bd6+2.630a6+2.8...30.axb6?Kf7would give black good drawing chances as his king, bishop, and rook are all active and coordinated, while white's bishop and king take a bit of time to get in the action.30...Bc7+2.831Bf1+3.0Kf8?!+4.031...e332.Rd5b533.Rd7Bb634.fxe3Bxe3+35.Kh2Ra836.Bxb5Ra737.Bc4+Kf838.Rd332Rd5+3.8b5+3.9a weird looking move, giving away a pawn for free. It is intentional though, trying to blockade the white pawns on the dark squares, where black's bishop can give him an advantage.. Other moves are not any better.. For example:32...Ke733.Bb5Ke634.Bc6and the a pawn runs33Bxb5+3.9Bb6+3.634Rd6+4.3Ba7+4.134...Ke7?35.Rxb6Rxb636.a735Rd7+4.3Ra8+4.736Rxh7+4.8Bd4+4.837Rd7+4.7Bb6+4.838Rd6+4.8Ba7+4.839Bc6+4.9Ke7+5.040Rxg6+5.3Rf8+5.941Rg7++6.3Rf7+6.542Rxf7++6.3Kxf7+6.043Kf1+6.1...if there were no pawns on the inside black could draw this by using his king and is bishop to guard against any potential advance of the a or b pawn. That is clearly no the case here, as whatever side the black pieces try to stop advancing allow white to advance on the other side of the board.43...Kf6+6.544Ke2+6.2Ke5+6.645f4++6.6exf3++7.146Kxf3+7.0Kf6+7.547Bd7+7.9...Seirawan says "After the game, Vassily told me, 'I don't know how I lost this ending.' I must confess it felt awfully good to defeat one of the most renowned endgame players in history in an ending."