Monday, March 20, 2017

Tomb Raider

The Tomb Raider franchise began in 1996 when game studio Core Design released under its parent company Eidos Interactive, a third-person 3D action adventure featuring acrobatic (and shapely) Lara Croft as the Tomb Raider navigating through traps. Since that time, the franchise has gone on to publish about fifteen game titles and inspire two movies starring Angelina Jolie. A reboot of the film franchise is in development with Alicia Vikander (from Ex Machina) in the title role.

I won a game last month that reminded me of the Tomb Raider. My opponent offered material for the possibility of trapping my queen. I stayed one step ahead of the traps and managed to escape with the material advantage intact. I'm playing Black. White's 20th move was 20.Qc2-e2, loosening the protection of the Nb3, to which I responded 20...Qc7-d7 threatening to capture the white pawn on a4 and having a follow-up threat against Nb3 for the next move. Notice that Black has a loose a7 pawn.

My opponent and I were both a little short on time, about 15 minutes to make 10 moves and reach move 30. Ideally, both sides could sink into 30-45 minute thinks and try to work out the next 8-12 plies with some level of certainty before making such committal moves. Barring that, I fell back on intuition that I could use the pawn exchanges at e4 and b4 and my well-placed rooks at c8 and d8 to aid my queen's escape. If the two white rooks came to a1 and b1, the bishop would have trouble discovering the attack of the b-rook because a1 was occupied and Bc1 allows simply Qxc3. If the minor pieces could not trap my queen, then I might be able to sac my queen for a rook, thereby gaining at least material parity. I also used a rough risk-reward calculation: if I took on a4, worst case scenario was that Ra1 forces my queen to retreat to d7 and then I would have to contend with Rxa7. So, sometimes, I used shorter variations than I really should have to substitute for calculated certainty. My opponent sweetened the reward by cutting off the protection of Bb2. 21.Nc2??

I knew that I was getting the a-pawn and at least positional play against a pair of awkward knights (if either knight goes to a1). If he tried to trap my queen with Ra1, I would get the Nb3 and tempo against the loose Bb2. So I took the bait and entered the tomb. 21...Qxa4 22.Ra1

Last chance to bail out with Qd7 Rxa7. No guts no glory. 22...Qxb3 23.Reb1

White is one move away from trapping the Black Queen with Ra3, but Black has several resources to delay that outcome. The most immediate is cxb4. This has the possibility of opening the c-file for the Black rook and it also temporarily prevents Ra3. However, White would almost certainly play Nxb4, keeping the c-file semi-closed. Another resource is dxe4, temporarily giving Black the potential of Qd5, but fxe4 leaves the queen trapped. I already began to see the possibility of a bailout sacrifice to get my queen out of trouble, but I wasn't sure it was going to work. So I went for the pawn exchanges to improve my rooks. 23...dxe4 24.fxe4

24...cxb4 25.Nxb4

A pause now that we're five moves closer to time control. I was annoyed at the weakness of my a7 pawn and the tempo that Rxa7 might get on my Bb7. Since I was already up a knight and a pawn, I calculated that 25...a5 26.Ra3 Qxa3 27.Bxa3 axb4 might be a bailout strategy. Do you see the other possibility of bailout yet? 25...a5 26.Ra3. White moved in to kill my queen.

Even though 26...Qxa3 27.Bxa3 axb4 28.Bxb4 carries the material advantage of queen (9) for rook and two knights (11), White's position seemed annoyingly consolidated. I couldn't see how I was going to organize my pieces for the next round of battle. With White's Nb4 under attack and both our time troubles becoming serious, I decided to play my ace in the hole. 26...Nd4!.

If he captures my queen with Rxb3, I capture his queen with check Nxe2+ with more even exchanges helping Black's endgame. If he captures my knight with cxd4, I capture his knight Qxb4 and get my queen away safely with the knight advantage. If he saves his queen, I save my queen with Qe6. My opponent let his time tick down to about 1 minute for 4 moves as he tried to work out this mess. He finally settled on 27.Qd2 Qe6.

At this point, I figured that White should trade knights and gain a protected passer at d5, e.g. 28.cxd4 axb4 29.d5. Another variation could have gone 28.Nc2 Nxc2 29.Rxc2. However, with his time trouble, White tried too hard to avoid exchanges and came up with a move that made his knight, bishop, and rook awkward. 28.Na2?

With the remaining time on my clock, I retreated while picking on the awkward rook. 28...Nb5 29.Ra4 Nd6 30.Qf2 Bc6. With a piece and a pawn down and the possibility of losing the rook, my opponent resigned. Black's most straightforward win goes 31.Ra3 Nb5 32.Ra4 Qb3 going back to the tomb to loot more treasure.

Monday, March 13, 2017

Brazil

Terry Gilliam's "Brazil" is a movie about a dystopian, Kafkaesque world of bureaucracy, superficiality, and terrorism. The main character descends into criminality and madness when he tries to swim against the current using his conscience, passion, and resourcefulness. One of the images toward the end of the movie involves a heroic figure becoming enveloped and then consumed by flying scraps of paper, a metaphor for the triumph of bureaucracy.

One of my recent games involved enveloping an enemy piece in pins and cross-pins in order to eventually win. Here is the position after Black played 16...h5:

Sam Lowry (Jonathan Pryce) wakes up one night to find the central heating in his apartment has gone on the fritz and is now too hot. He leaves a message with the Central Services answering machine and goes to sleep with his head in the refrigerator.

Black seizes the initiative with a flank attack on White's g4-h3 pawn chain. Note that if g4-g5, Black has the fork Bf4+. 17.Nh4?!.

Sam is awakened by freelance outlaw repairman Archibald "Harry" Tuttle (Robert De Niro represented by the White knight) who tries to resolve the heating problem.

I think that my opponent wanted to try to exploit the hole at g6, but there is no time for that because the Black rooks are going to use the tempo Rh6 to assist in doubling on the h-file. 17...hxg4 18.hxg4 Rh6.

The call to Central Services goes through and two repairmen, Spoor (Bob Hoskins) and Dowser (Derrick O'Connor) are dispatched to Sam's apartment on a collision course with Harry Tuttle.
With some quick thinking, Sam rebuffs the attempt by Spoor and Dowser to bully their way into his apartment by requesting official paperwork, thereby preventing an armed confrontation with Tuttle. Tuttle zips away via a zipline.

White regroups with an awkward sequence: 19.Ng2 Reh8 20.Rg1?!. I mentioned after the game that White should have probably exchanged one pair of rooks so that I wouldn't get so much activity with the second rook on White's third rank.

Far from being defeated, Spoor and Dowser (Black's rook pair) keep showing up at Sam's apartment, once to pull all the ducting out of the walls, and then once more when the system has turned the apartment into a freezer. During the daytime, Sam spends all his efforts at work trying to track down the girl of his dreams, Jill Layton (Kim Greist).

Black's rooks soon lodge in White's position, first at h2 where it causes immobility in the Ng2 because of the looseness of f2. White tries to stabilize his weaknesses by moving his king from c1 to e2. This is double-edged in that his king becomes the target of pins and skewers. 20...Rh2 21.Kc2?! Be8 22.Kd2 Bg6 23.Ke2 R8h3 24.f3. With the third rank pressure preventing White from moving his Rd1 for fear of Bd3+, White blocks this coordination by advancing his f-pawn. But now the knight is pinned to the king.

With the aid of Sam's erratic behavior, Sam (White King) and Jill (King Bishop Pawn) run afoul of the law and are soon labeled as terrorists.

I'm somewhat proud of this next sequence which was not easy to find and advance the attack. The light-squared bishop is hindered by the f3 and g4 pawns. It would really like to participate in the attack on g2 and therefore belongs on e4 or h3, but there is currently no path. Luckily, a pawn break 24...f5! was handy, weakening the f3-g4 structure enough to become porous. Then followed 25.gxf5 Bxf5 26.Rdf1 Rg3 27.Rf2 Bh3.

Once 27.Rf2 appeared on the board, I had to calculate the possibility of 27...Bh3 being answered by Ng2 jumping. The trickiest jump is 28.Nf4 because it cuts the rooks off from their protection by the bishop on d6. At first I thought I had to trade both pairs of rooks 28...Rxf2+ 29.Kxf2 Rxg1, but I worried about 30.Nxh3. I couldn't see clearly enough to find 30...Rh1 which reopens 2 threats of Rxh3 and Rh2+ skewering the Bb2. But I was reassured when I found 28...Rxg1 29.Rxh2 Bxf4 30.Rxh3 Rg2+ 31.Kd3 Rxb2. Unfortunately, the next pair of moves were both blunders. White bluffed and Black blinked with 28.cxd5? exd5?. 28...Bxg2! would have been completely winning as Bxf3+ is difficult to meet. But finally, the White king walks into the trap that Black initiated with 20...Rh2. 29.Kf1. The knight is pinned again.

Sam hatches a daring plan to erase Jill from the Ministry of Information's databases. Unfortunately, Big Brother is more than a one-trick pony. Information Retrieval, e.g. police forces and interrogators, find Sam's hideout, break down the doors, and arrest him.

With the rooks and knight and king largely immobilized, the plan of Bd6-Bf4-Be3 seemed decisive. 29...Bf4 30.Re2. Here I missed the clever zugzwang 30...g5!. 30...Rxf3+ is enough to win, but my advantage falls from 5.9 to 3.0. 31.Rf2 Rxf2+ 32.Kxf2.

Sam finds himself about to be tortured. Suddenly a shot rings out and he is rescued by some commandos, including Harry Tuttle. There is seeming triumph over bureaucracy when Tuttle blows up the Ministry of Information buildings.

A semi-crucial move appears at this point of the game. If I had felt confident with all the pin pressure I had put on the Ng2, I still had to find my way to a winning endgame. The extra pawn at g7 could still win, but the ending should be bishops of same colors to win it. With that in mind, White must not be allowed to unpin the knight and capture Kxf4. Therefore, 32...g5! was necessary. My opponent said he thought he had a chance at this point, but the advance of the g-pawn shut down his last hope. The cocoon around the knight unravels, but the knight is also gone. 33.Kf3 Rxg2 34.Rxg2 Bxg2+ 35.Kxg2 g4

But the falling paper debris from the explosion envelops Tuttle and mummifies him like a spider's prey. Sam rushes to help, but by the time he unravels the paper, Tuttle has seemingly evaporated.
The authorities chase Sam until he finds Jill again and they escape to the idyllic countryside.

The game concluded with fairly simple plans of centralizing the Black king to f5, possibly e4, and trying to queen the g-pawn. The exploitation of two weaknesses is a common endgame principle. 36.Kf2 Kd7 37.Bf3 Ke6 38.Ke2 Kf5 39.Kd3 g3 40.Be1 g2 41.Bf2 Bh2 42.a4 g1=Q 43.Bxg1 Bxg1 44.b4 Kf4 45.Kc3 Ke4 and White resigned.

The unharmed faces of Sam's tormentors, Jack Lint (the Black King Bishop, played by Michael Palin) and Deputy Minister of Information Mr. Helpmann (the Black King, played by Peter Vaughan) interrupt Sam's fantasyscape, revealing that Sam has only escaped his torture by becoming completely delusional.

Sunday, March 12, 2017

Rosetta Stone

The Rosetta Stone was found in Egypt in 1799 having parallel texts of ancient Greek, Demotic, and Egyptian hieroglyphics. The Rosetta Stone proved key to deciphering Egyptian hieroglyphics. I wanted to review one of my wins at the club last month with some attention to the language used to in analysis, from what I was thinking at the time to what postmortem analysis has shown, and try to translate it into Temposchluckerese.

As Black, I had just played 24...Bd7-c8. The evaluation is approximately equal. White's centralized knights and queenside pawn chain to c5 threaten to create a b-pawn passer. Black's f5-e4 pawn chain and attack against the e3 isolani provide central space and counterplay. It looks like Bc8 overprotects a6 from Qxa6 so that my Nc7 is freed up a bit from defensive duties. This increases not only the mobility of the Bc8 and the Nc7, but also the Rd8 which now faces White's Nd4. The main defect of Bd7-c8 is that the c6 pawn is now held only by my Qh6. I had been keeping my eye on White's weak e3 pawn for the past 10 moves. White had not improved his defense of that pawn for a while. Now that my rook faces Nd4, the Pe3 is that much more vulnerable, especially in a sequence where Be7-g5xe3 lands with check. In my usual language, I was going to try to remove the Pe3 guard of the Nd4.

As I understand Temposchlucker's terminology, e3 is a Point of Pressure (PoP) as is d4. Lines of Attack (LoA) include h6-e3 and d8-d4. The Nd4 is a Barely Adequately Defended (BAD) piece, but I would also like to label the Pe3 and Black's Pc6 and Pf5 as BAD. In addition, Pe3 is already immobilized and blockaded by Black's Pe4, but a functional immobility also exists in that the Nd4 is only defended by Pe3. With these critical items in play, White missed chances to bolster the e3-d4 problems and decided to press his queenside pawn majority with 25.a4. Of course, I played 25...Bg5 and my opponent sank into a long think. I felt somewhat confident at this point and walked around the club. I came back after 26.Kh1.

This shouldn't have surprised me, but it did. Kh1 comes close to solving White's problems. Sometimes if you have to lose material and time, the solution is to quickly concede what you have to concede and move onto getting something in return. Black will capture Bxe3, but White will follow with Nxc6, making the queenside majority scarier and temporarily threatening a Nxd8 win of the exchange. Now that I saw White's plan, it was my turn to concede something and try to gain something in return. I could see that Rd2 might be a devastating blow if I could camp heavy pieces on White's second rank. So I decided to go down this variation. 26...Bxe3 27.Nxc6 Rd2.

At this point, I worried about White's offensive possibilities. 28.Ne7+ is a move. Can Black counterattack with 28...Kf8? I decided, correctly as verified afterward, that 29.Nxf5 threatening Black's queen and a discovery on Black's king were too dangerous, so I was going to have to play 28...Kh8. Because of the balance between pieces en prise at Be3 and Nc6, White probably felt he had to keep his queen in contact with Be3. Also, it was difficult to see past 28.Qc4+ Be6 because you have to look extra ply ahead while the queen is attacked, but 29.Ne7+ Kh8 30.Qxe4 would have smashed Black's center and brought the Nc6 back into the protection of the Bg2 and Qe4.

After Qc4+ Be6 Ne7+ Kh8, the lines of attack (LoA) are c4-g8 and f1-f8. The Rf8 is a BAD piece as is the Pe4 (both are also on Points of Pressure, PoP).

Instead White played the passive 28.Qe1.

Here I thought for a long time on how to proceed. I was fixated by the looseness of my Be3 and his Nc6. I didn't want to trade evenly since my Be3 was quite strong as an unopposed bishop slicing into White's position and supporting eventual passed pawns. I soon noticed that the Nc3 was also loose and tried to limit the mobility of the Nc6 by playing 28...Bg5. Now that the bishop is safe, Qxc6 is back on and Ne5 runs into Bf6 skewering two knights on the diagonal.

But what I failed to appreciate was that Rd2 had given me a significant Line of Attack on PoPs g2 and h2. Plus my queen which had been at h6 for the past 15 moves had another serious Line of Attack against PoP h2. Add this to the already existing LoA against the BAD Nc6, and the limited mobility of White's King because of Be3 and I could have increased my chances of finding the brilliant 28...Rf6!! This move looks like it only increases the pressure on Nc6, but what it really does is threaten checkmate. White staves off mate with 29.h4 Rxc6, but has to soldier on a piece down. If he saves his knight, e.g. 29.Na5 or even 29.Ne7+ Kf7 30.Nxc8, there comes the shocking 29...Qxh2+ 30.Kxh2 Rh6#.

I emerged from the middle game with 2 extra pawns and converted them to victory.