mersenneforum.org White 33
 Register FAQ Search Today's Posts Mark Forums Read

 2014-06-20, 20:45 #12 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22×3×641 Posts We have plenty of time to discuss this important decision on how to proceed at move 33. :-)
 2014-06-20, 21:36 #13 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22·3·641 Posts Let's look at the position after [B]33 Nb5 Be5 34 Nc5[/B] [B]Kd8 [/B][B]35 Ra8 Nc4 36 Bb7: [/B]At this point, The B/c8 is attacked twice and defended twice, so there's no immediate need for ... Nb6 (which stops defending the a-pawn). Suppose Black plays something other than 36 ... Nb6, like 36 ... g5 (I'm not saying this is the best alternative). Then we can't add an attacker (of the B/c8) with 37 Na7 because of 37 ... Nd4+ winning our N/c5, or with 37 Nd6 because of 37 ... Bxd6. and it would be difficult to maneuver our N/c5 onto square a7, b6 or d6 from which it could attack c8. So Black doesn't _lose_ the B/c8 unless he moves his king to e8 (in order to escape to the king-side, for instance). We could take the B/c8 with 37 Bxc8, forcing Black to put his knight into the pin with 37 ... Nxc8, but unlike the situation after 33 Bxc8 where we can _force_ Black to pull his bishop away from guarding a5 in order to add a defender to the pinned knight, here we don't have as much leverage. We could play 38 Nb7+, which would allow 38 ... Kd7 (continu9ing to guard the N/c8), and then add another attacker with 39 Na7, but now Black could just escape with 39 ... N8b6 or (perhaps better) 39 ... N8d6 or (perhaps best - see below) 39 ... Ne7. In that last position, we've crowded all our pieces into the ab78 corner, whereas Black's pieces are much freer, so we didn't accomplish anything. 40 Nb5 would attack the a5-pawn a second time, but 40 ... Nc6 (after 39 ... Ne7) would defend it a second time, too. The position looks like Black has herded our R+2N into the corner with his B+2N. It's true that we could just bring our rook out with Rf8 or Rg8 or Rh8 so as to use it from a different angle ... but that was true in the lines after 33 Nxc8, too. [B][/B]
 2014-06-20, 22:04 #14 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22·3·641 Posts [QUOTE=LaurV;376311]After i spent one more hour in front of the board (1:05 AM here), it seems we also win with 33 Nxc8 Kxe6 (other moves seems worse for black) 34 Bh3 Kf7 35 Rb7, Bb4 etc, so it seems you are right, in this point does not really matter.[/QUOTE]Yes, it does matter.[quote]However, this seems a bit more complicate for me, it is not "straight by force". [/quote]Wrong. It _is_ "straight by force"! We _force_ the Black king's moves toward f7 because [I]if he moves his king anywhere else, he loses his N/e7[/I]. Then we _force_ Black to pull his bishop away from defending a5 to defend the N/e7 because we attack his N/e7 twice while it's defended only once and it can't move away from our attack on it. Then we _force_ Black to put his bishop into the pin by taking the N/e7 with our N/c8, so that the only way Black can avoid losing a piece outright is to retake with his bishop. Then, by moving our rook to attack the a5-pwan while maintaining the pin of the bishop, we _force_ Black to use his only remaining piece, his knight, to defend the a5-pawn. Then we use our bishop to attack Black's knight so that we _force_ Black either (1) to lose that knight when we take it with our bishop or (2) to move that knight away from defending the a5-pawn, so that we can take the a5-pawn with our rook. It's force, force, force ... _all the way_ after 33 Nxc8 Kxe6, until we increase our advantage to ensure a win. (It's also force, force all way after 33 Nxc8 Nxc8, to an ensured winning advantage.) [quote]Also 33 Rb6 is winning move.[/quote](It almost seems that you want to avoid capturing Black's B/c8. :-) Both Nb5 and Rb6 have in common that they remove an attacker from the B/c8.) [quote]They can't do too much now,[/quote]But they can't do [I]anything at all[/I] after we finish forcing them for move after move after move starting with 33 Nxc8. [quote]When I posted today (the "long" branch) I didn't have a board in front of me (job, lunch break), beside of Serge's board, and I may (intentionally) missed to post "branches",[/quote]Take your time when you have a board later. :-) [quote](i keep telling you the time so you won't blame me hard if I say something stupid, I should be sound sleeping at this "early" hour)[/quote]Get your rest. There are five more days for looking at this, and it's an important decision on which way we proceed with our attack. :-) Last fiddled with by cheesehead on 2014-06-20 at 22:09
 2014-06-21, 02:25 #15 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 1E0C16 Posts [QUOTE=LaurV;376311]Also 33 Rb6 is winning move.[/QUOTE]It could transpose right into the 33 Nxc8 sequence ... 33 Rb6 Kxe6 34 Bh3+ Kf6 35 Nxc8+ Kf7 35 Rb7 ... except that 34 ... Kf6 is not forced, as it would be if the N/e7 were already being attacked as in the 33 Nxc8 line. Black could instead play 34 ... Kd5 or ... Ke5, so that we can't easily arrange to pin some piece to the king so as to distract the defender of the a5-pawn and then capture the a5-pawn. Also, if 34 ... Kd5 or Ke5 35 Nxc8, Black can skewer our rook and king with 35 Bd4+, and we lose. Last fiddled with by cheesehead on 2014-06-21 at 02:32
 2014-06-21, 02:50 #16 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22·3·641 Posts [QUOTE=cheesehead;376273][QUOTE=cheesehead;376214]< snip > If [B]33 Nxc8[/B] If [B]33 ... Nxc8 34 Bh3[/B] If [B]34 ... [/B] (THIS PORTION IS NOT YET DOUBLE-CHECKED) < snip >[/QUOTE] There's a different move 34 for us that's superior to 34 Bh3 here: If [B]33 ... Nxc8[/B][/quote]There's yet another move for us here: [b]34 Nf8+[/b] gets a head start on cleaning out Black's king-side pawns. If [b]34 ... Kc7 35 Rb7+ Kd6 36 Nxg6[/b] and we continue being careful not to have pieces on dark squares that could be skewered by Black's bishop.
 2014-06-21, 03:04 #17 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 170148 Posts Reviewing the analysis, I see that I made too broad a claim for [B]33 Nxc8[/B]. I think I have discovered and documented a forced win there _only_ when Black plays [B]33 ... Kxe6[/B]. If Black instead plays [B] ... Nxc8[/B], we can force at least one swap-down of pieces, which increases the effect of our Exchange material advantage, but that alone is not yet quite enough to say that we definitely have a win. Now, in the lines after [B]33 Nxc8 Nxc8[/B], we do force enough clearing of the position so that we can reasonably foresee using our rook to eliminate Black's king-side pawns (which Black will have too few pieces left to prevent), which will result in our win. So, a question in comparing move 33 alternatives is: After [B]33 Nb5 Be5[/B], can we show that we can force clearing the board position enough so that we also can reasonably foresee eliminating Black's king-side pawns? If yes, then I'll consider the [B]33 Nb5 Be5[/B] line to be about equal to the [B]33 Nxc8[/B] line in win-forcing value. Then my choice would be a matter of which position seems more comfortable to me. That doesn't mean I'd necessarily choose the line whose position seems most comfortable. I've already seen earlier in this game how I can learn from going through positions that I'd normally never see when employing my usual individual playing style. So _if_ I can be convinced that 33 Nxc8 and 33 Nb5 are about objectively equal in win-forcing value, I may well vote for 33 Nb5 in order to have the further learning from a type of position I'd normally not encounter. Last fiddled with by cheesehead on 2014-06-21 at 03:16
 2014-06-21, 20:45 #18 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 1E0C16 Posts Just a quick note: One big reason for my preference for 33 Nxc8 is the elimination of Black's light-squared bishop before it can ever challenge our light-square bishop's influence on the long a8-h1 and c8-h3 diagonals. You can see in the analyses I've posted how well our bishop supports our more-forward pieces (and so often restrains the Black king's movements) from its g2 or h3 post. Sure, the B/c8 is pinned if the Black king retreats to the back rank, but I don't want it to get any chance to get out to exercise its long-range powers (or contest ours) while it's unpinned. Of the alternatives suggested for our move 33, I'm struck by the common factor that both 33 Nb5 and 33 Rb6 take an attacker off Black's B/c8. To me, either is a mistake because of that alone. One active Black bishop is hard to handle; I don't want us to have to face two. Last fiddled with by cheesehead on 2014-06-21 at 21:01
 2014-06-22, 00:31 #19 WMHalsdorf     Feb 2005 Bristol, CT 33·19 Posts 33 Nxc8 Nxc8 34. bC6+ results in a favorable exchange down to a won position.
 2014-06-22, 20:54 #20 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22×3×641 Posts [QUOTE=WMHalsdorf;376407]33 Nxc8 Nxc8 34. bC6+ results in a favorable exchange down to a won position.[/QUOTE]Thank you. 34 Bc6+ was the unposted move I had in mind for: [QUOTE=cheesehead;376339]< snip > If Black instead plays [B] ... Nxc8[/B], we can force at least one swap-down of pieces,[/QUOTE]though I'm a bit more conservative in evaluating the result ==> Not a win yet, with bishops of opposite color on the board (after 34 ... Kxe6). - - I just found that in my favorite [B]33 Nxc8 Kxe6 34 Bh3+[/B] line, Black does have at least one defensive sequence that avoids losing his a5-pawn, so I retract my claim that we have a forced win of material there. [B]33 Nxc8 Kxe6 34 Bh3+ Kf6 35 Rb6+ Kf7 36 Rb7 Bb4 37 Nxe7 Bxe7 38 Ra7[/B] If [B]38 ... Nb3[/B] unlike the 38 ... Nc4 line, this choice does allow Black a non-material-losing defense (* sigh *) [B]39 Bg2[/B] threatens Bd5+ winning the N/b3. (Here, Bf1 doesn't work because after Bc4 the knight can retreat to Nd2 threatening the bishop so that we don't have the tempo to capture on a5 before Black can do Bb4.) But because our king is still on the a7-g1 diagonal, Black will be able to gain a tempo here: If [B]39 ... Kf6[/B] we have to move our K or R to avoid the ... Bc5+ fork, so we don't have the tempo for 40 Bd5. If [B]40 Kf1 Bb4[/B] relieves the N/b3 as guard of a5. If [B]40 Ra6+ Ke5[/B] positions the king to guard d5 so that we can't play 41 Bd5 to drive the knight away from guarding a5. 41 Bf1 doesn't work, as explained above, so next Black can play ... Bb4. In this position after 40 ... Ke5, our rook could [I]start[/I] cleaning up Black's king-side pawns (but if 41 Rxg6 Bg5 42 h4 Be2 slows that down) so that we have connected passed g-h pawns, which will eventually prevail. But my favorite [B]33 Nxc8 Kxe6 34 Bh3+[/B] line is not as invulnerable as I'd previously thought. - - While I was finding the Black defensive line shown above, my analysis tree "mushroomed" toward enormous size. I hate to post it in its current ragged state, but it's necessary for us to all look at the branches. I might not have time for cleaning it up. Last fiddled with by cheesehead on 2014-06-22 at 21:48
 2014-06-22, 21:23 #21 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22×3×641 Posts This "consolidated analysis" is messier than I'd prefer to post ... but our clock is running. Basically, it's just a dump of my off-line analysis file. A) This may not have incorporated every posted analysis. I hope it has, but not 100% sure. B) I've tried to put ???? at places that need going-back-to, but several such places are unmarked with that. ======================= White 33 analysis consolidation, in progress ======================= [B]33 Nxc8[/B] If [B]33 ... Nxc8[/B] [B]34 Bh3[/B] threatens to win material. If [B]34 ... K-moves 35 Rxc8[/B] we're a whole rook up If [B]34 ... Na7 35 Rb7+ and 36 Rxa7[/B] we're a whole rook up If [B]34 ... Nd6[/B] If [B]35 Nf8+[/B] If [B]35 ... Ke7 36 Nxg6+[/B] If [B]36 ... Kf7 37 Nf4[/B] If [B]36 ... Kf6 37 Nf4 Ke5 38 Rf8+[/B] If [B]38 ... Kd4 39 Ne2+ and 40 Nxc3[/B] If [B]38 ... Kd5 39 Be6+[/B] If [B]39 ... Kc5 40 ????[/B] If [B]39 ... Kc6 40 ????[/B] If [B]39 ... Kd4 40 Ne2+ and 41 Nxc3[/B] If [B]39 ... Ke4 40 Kg2[/B] If [B]40 ... Ke5 41 Rf8[/B] If [B]41 ... Ne4 42 ????[/B] If [B]39 ... Ke5 40 ????[/B] If [B]35 Nf4+[/B] If [B]35 .... Kc7[/B] If [B]35 Ra8[/B] If [B]35 ... Kb6 36 Nd5+[/B] If [B]35 ????[/B] If [B]34 ... Ne7 [/B](big mistake for Black)[B] 35 Rd8+ Kc6 (forced) 36 Bg2+ (NOT 36 Nd4+ Bxd4+ and if 37 Rxd4 Nf3+)[/B] If [B]36 ... Kb6 (not ... Nd5 or Ne4 or Nf3+) 37 Rd7 checkmates or wins a piece[/B] If [B]37 ... Bb4 or Bf6 or Nc8 or Nf5 or Ng8 38 Rb7+ Ka6 (forced) 39 Nc7++[/B] mate If [B]37 ... Nc6 38 Rd6 [/B] threatens Rxc6+ If [B]38 ... Bb4 39 Rxc6+[/B] we're a whole rook up If [B]38 ... Be5 39 Rxc6+[/B] we're a whole rook up If [B]38 ... Bd4+ 39 Nxd4[/B] we're a whole rook up If [B]38 ... Ka6 39 Rxc6+ and 40 Rxc3[/B] we're a rook and bishop up If [B]38 ... Ka7 39 Rxc6[/B] we're a whole rook up If [B]38 ... Kb7 39 Rxc6[/B] we're a whole rook up If [B]38 ... Ne4(Nf3+) 39 Bxe4(Bxf3)[/B] we're a whole rook up If [B]37 ... ????[/B] If [B]36 ... ???? (not ... Nd5 or Ne4 or Nf3+)[/B] If [B]35 Rb7+ K???? [/B] [B]34 Bc6+[/B] Very tempting to play, because of its beauty in offering the opponent a choice of pieces to capture. But it results only in trades, not win of material. If [B]34 ... Kxc6 35 Rxc8+ Kd5/6/7 36 Rxc3[/B] we're up (only) a rook for knight, but should win in this simplified position If [B]34 ... Kxe6 35 Rxc8 Bb4[/B] we're up (only) a rook for knight [B]34 Nf8+[/B] ** the following analysis needs cleaning up ** If [B]34 ... Kc7 35 Rb7+[/B] If [B]35 ... Kd8 36 Ne6+ Ke8 37 Bc6++[/B] If [B]35 ... Kd6[/B] If [B]36 Rd7+[/B] If [B]36 ... Kc5 37 Rc7+[/B] If [B]37 ... Kb4 or Kd4 38 Rxc8[/B] we're a rook up If [B]37 ... Kb6 or Kd6 38 Rxc3[/B] we're a rook up If [B]36 ... Ke5[/B] If [B]37 Nxg6+[/B] If [B]37 ... Ke6 38 Re3 Bb4 39 Nf4+ K-moves 40 Nd5[/B] threatens Rxd2 or Nxb4 axb4 and then our rook cleans up the b-pawn If [B]39 ... Bc5+ 40 Kh1[/B] If [B]40 ... Nb1 41 Rd1 Na3 42 Rc1[/B] ???? If [B]40 ... Nc4 41 Rc3[/B] ???? If [B]37 ... Kf6[/B] If [B]37 [/B] If [B]37 ????[/B] If [B]34 ... Kd8[/B] loses a piece to [B]35 Bb7[/B] If [B]35 ... Kc7 36 Rxc8+ Kxb7 37 Rxc3[/B] we're a rook up If [B]34 ... Ke8 or Ke7 35 Rxc8[/B] we're a rook up [B]34 ????[/B] If [B]33 ... Kxe6[/B]. We can pin the N/e7 to Black's king, then use that to pull his B away from a5-defending duty long enough to nab the a-pawn. That requires persuading Black to move his king to f7, [I]while always being on guard not to allow Black to skewer our K and R with his B.[/I] (Moving our K off the a7-g1 diagonal would be a good way to avoid the skewer, but it'll be tough to find a place where we can afford to spend that tempo while maintaining the pin.) [B]34 Bh3+[/B] gives our king another flight square in case of (or to avoid) future check from Bd4+ or Bc5+, _and_ reduces the number of squares from which the Black king can defend the N/e7. If [B]34 ... Kf6[/B] (or ... Kf7 35 Rb7) [B]35 Rb6+ Kf7[/B] (or else loses the N/e7) [B]36 Rb7[/B]. [I](Note that we place our rook on the a7-g1 diagonal _only_ when giving check, allowing Black no chance to skewer)[/I] Now the N/e7 is both pinned and double-attacked, forcing Black to use his B to defend it. [B]36 ... Bb4 (or Bf7) 37 Nxe7 Bxe7[/B]. Now we've (1) pulled the B off a5-defending duty and (2) pinned it to the king, allowing us to put our rook on the a7-g1 diagonal without immediate skewer. [B]38 Ra7[/B] threatens 39 Rxa5, [I]but the R's on the same diagonal as our K, so before Black's B can move again, we must move our K or R[/I]. Black can protect a5 with his N from b3 or c4, so we have to devise a way to drive it away from those two positions. If [B]38 ... Nc4 39 Bf1[/B] chases the knight away from a5-guard duty, but not before possible complications if Black un-pins his B. If [B]39 ... Ke8 or Kf6 or Kf8 (not ... Ke6 40 Bxc4+)[/B] threatens the ... Bxc5+ skewer. but we now have just one tempo for moving our king because Black can't defend both of our attack targets (a5 and c4) with one move. [B]40 Kg2[/B]. Next we will soon capture one of our targets - but only after complications in one case. If [B]40 ... any K-move or pawn-move 41 Bxc4[/B] If [B]40 ... Bc5 41 Ra8+ (or Ra6+)[/B] If [B]41 ... K-moves 42 Bxc4[/B] If [B]41 ... (with K on f6) Nb6 or Nd6 42 Rxa5[/B] If [B]41 ... (with K on f6) Bb6 or Bd6 42 Bxc4[/B] If [B]40 ... any B-move except Bc5 41 Bxc4[/B] If [B]40 ... any N-move except Nf3+ 41 Rxa5[/B] If [B]40 ... Ne3+[/B] Black complicates, but as far as I've analyzed none can harm us, and we will eventually succeed in capturing at least one of our targets on a5 and c4. [B]41 Kf2[/B]. We can place our king back on the dark diagonal because it's shielded by the N/e3 from a check via ... Bc5. If [B]41 ... Bc5 42 Rxa5[/B]. We can nab the pawn here because [I](rather amazingly) [U]none[/U] of the eight possible discovered checks from 42 ... N-move can harm us![/I] If [B]42 ... Nc2+ 43 Rxc5[/B] If [B]42 ... Nc4+ 43 Rxc5[/B] If [B]42 ... Nd1+ 43 Kf3[/B] If [B]42 ... Nd5+ 43 Rxc5[/B] If [B]42 ... Nxf1+ 43 Kxf1[/B] (not 43 Rxc5 Nxh2) If [B]42 ... Nf5+ 43 Rxc5[/B] If [B]42 ... Ng2+ 43 Rxc5[/B] If [B]42 ... Ng4+ 43 Ke2[/B] If [B]43 ... Nxh2 44 Rxc5[/B] If [B]41 ... Nxf1 42 Kxf1[/B] If [B]42 ... Bb4[/B] Now Black has made it necessary for our king to travel to assist our rook if we want to capture the a-pawn. But if we were to do exactly that, Black's king might travel down to take our g-h pawns and then march his g-h passers down to victory. (I'm not going to try counting moves _now_ to see whether we could win a king/pawn-race, but we can do that later if this position does arise.) However, by themselves the Black a-pawn and B/b4 pose no offensive threat. Our a-pawn can continue to block theirs by itself (as long as we prevent Black's king from reaching it), while we switch our rook to the king-side. (THE FOLLOWING PORTION (after 38 ... Nb3) _MAY_ NOT YET REFLECT WHAT'S IN MY PRECEDING POST.) If [B]38 ... Nb3 39 Bg2[/B] threatens 40 Bd5+ If [B]39 ... Ke6 40 Bd5+[/B] If [B]40 ... Kxd5 41 Rxe7[/B] If [B]40 ... Kf6 41 Kg2 N-move 42 Rxa5[/B] If [B]39 ... Ke8 40 Ra8+[/B] If [B]40 ... Kf7 41 Bd5+ and 42 Bxb3[/B] If [B]40 ... Kd7 41 Bd5[/B] If [B]41 ... N-move 42 Rxa5[/B] If [B]41 ... Bc5+ 42 Kg2[/B] If [B]42 ... N-move 43 Rxa5[/B] If [B]42 ... else 43 Bxb3[/B] If [B]41 ... Bb4 42 Bxb3[/B] If [B]39 ... Kf6 40 Kh1 (not Bd5 Bc5+) Bb4[/B] If [B]39 ... Kf8 40 ???? - - [/B] THE FOLLOWING ANALYSIS (AFTER 33 Nb5) IS ONLY SKETCHY AND MAY NOT INCLUDE WHAT ELSE HAS BEEN POSTED ABOUT THIS BRANCH. [B]33 Nb5[/B] **=> If [B]33 ... Be5[/B] threatens to take our rook [B]34 Nf8+[/B] If [B]34 ... Kc7 35 Rb7+ Kxc8 36 Rxe7 Bd4+ 37 Kh1 Bc5 38 Re8+ Kc7 39 Ne6+ Kd7 40 Rd8+ Kxe6 41 Rxd2[/B] If [B]34 ... Kd8 35 Nxe7+[/B] If [B]35 ... Kxe7 36 Nxg6+[/B] If [B]34 ... Ke8 35 Nd6+ Nc8 36 Rxc8+ Ke7 37 Nb5[/B] threatens 38 Nxg6+ If [B]37 ... Bd4+ 38 Nxd4[/B] Last fiddled with by cheesehead on 2014-06-22 at 21:48
 2014-06-23, 05:56 #22 cheesehead     "Richard B. Woods" Aug 2002 Wisconsin USA 22·3·641 Posts I'm working to try to analyze [b]33 Nb5[/b] objectively, without the negativity toward it that I had earlier when I thought [b]33 Nxc8[/b] was a forced win.

 Similar Threads Thread Thread Starter Forum Replies Last Post cheesehead Game 2 - ♔♕♙♘♖♙ - Shaolin Pirates 23 2014-02-13 08:04 cheesehead Game 2 - ♔♕♙♘♖♙ - Shaolin Pirates 12 2014-01-31 20:14 cheesehead Game 2 - ♔♕♙♘♖♙ - Shaolin Pirates 16 2014-01-24 22:11 cheesehead Game 2 - ♔♕♙♘♖♙ - Shaolin Pirates 22 2013-12-20 18:11 LaurV Game 2 - ♔♕♙♘♖♙ - Shaolin Pirates 15 2013-12-08 10:49

All times are UTC. The time now is 06:03.

Mon Apr 12 06:03:16 UTC 2021 up 4 days, 44 mins, 1 user, load averages: 1.39, 1.76, 1.75