Latest Updates:
Normal Topic Schlechter - Walbrodt, Vienna 1898 (Read 753 times)
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 1727
Location: Columbus, OH (USA)
Joined: 01/02/15
Re: Schlechter - Walbrodt, Vienna 1898
Reply #2 - 02/23/24 at 07:29:38
Post Tools
Haha, humans are not supposed to calculate this to mate. Smiley Just finding a way to get in via one of the threats Fine mentioned (Nxc6 or Kg5) is good enough. I mentioned the engine only because 20 years ago it was no help at all and back then I had to figure it out on my own. Different today though.

If you looked at the pgn, it's worth asking yourself why Fine gave 1.Kf3 Ke7! -- why the exclam? Let's see.
1.Kf3 Ke6 "?" 
2.Nd3 Kf6 (2...Kf5 3.Nf4 and takes h5. 2...Ke7 3.Kf4 Kf6 4.Ne5) 
3.Ke3 and we have the position where Walbrodt resigned. See the chessgames.com analysis for how to win that. 

Of course this means 
1. Ke3 Ke7 is also "?"
2. Kf3 Ke6 
3. Nd3 as above.

So we have three related squares: (1) f4 vs f6, (2) f3 vs e7, (3) e3 vs e6. But white has maneuvering room on the second rank, whereas black does not, because the knight controls f7. Ergo, play the white king to the second rank and lose the needed tempo there (triangulation). We just have to do it in a way that black does not have a helpful check with the Bishop. 

Anyway, here is a short winning variation given by Roberto Grau (1940) Tratado General de Ajedrez, Tome IV. (I don't have this book, but I have a PGN from Christian Sánchez, Argentina.)
1. Kf3 Ke7 
2. Kf2 Kf6 
3. Ke2 Bg4+ (3...Kf5 4.Kf3 is simple enough, although the engine likes 4.Nf7. Pick one.)
4. Ke3 Bc8 
5. Kf4 +-

I doubt Grau used Stockfish....
  
Back to top
 
IP Logged
 
cathexis
God Member
*****
Offline


No matter where you go,
there you are.

Posts: 645
Location: Stafford, Virginia USA
Joined: 03/03/20
Gender: Male
Re: Schlechter - Walbrodt, Vienna 1898
Reply #1 - 02/22/24 at 23:34:34
Post Tools
Well,
I started to take a wack at it but felt absolutely wiped out. I could hardly sit at the PC. Guess who the latest COVID-19 patient is? No idea from where. But...at least this gave me a good excuse to weasel out and turn on SF. After 5 hrs it said White +250! but still hadn't found a mate. That was yesterday. Then I saw today that somehow SF15 was running.  Roll Eyes So I loaded SF16 this morning and let it run. 3 hours later it had the solution:

[Stockfish 16:]

1.Kf4–e3 Kf6–e6 2.Ke3–e2 Ke6–f6 3.Ke2–f2 Kf6–e6 4.Kf2–g3 Ke6–e7 5.Kg3–f3 Ke7–e6 6.Ne5–d3 Ke6–f6 7.Kf3–e3 Kf6–f5 8.Nd3–f4 Kf5–g4 9.Nf4xh5 Kg4xh5 10.Ke3–f4 Kh5–g6 11.Kf4–e5 Bc8–g4 12.Ke5–d6 Kg6–f5 13.Kd6–c7 Kf5–e4 14.Kc7xb7 Ke4xd4 15.Kb7xc6 Bg4–f3 16.b6–b7 Kd4–e3 17.b7–b8Q d5–d4+ 18.Kc6–d6 d4–d3 19.Qb8–b3 Ke3–e2 20.c5–c6 d3–d2 21.c6–c7 Bf3–b7 22.Qb3xb7 d2–d1Q+ 23.Qb7–d5 Qd1–c1 24.Qd5–g2+ Ke2–d3 25.Qg2–f3+ Kd3–d2 26.Qf3–f4+ Kd2–d1 27.Qf4xc1+ Kd1xc1 28.c7–c8Q+ Kc1–d2 29.Qc8xa6 Kd2–c3 30.Qa6–e2 Kc3–b3 31.Qe2–d3+ Kb3–a2 32.a5–a6 Ka2–b2 33.a6–a7 Kb2–c1 34.a7–a8Q Kc1–b2 35.Qa8–a3#[#35/103]

Ran it a couple times and it typically promotes b7-b8Q on move 17 or 18. When I ran it on SF 15 it was at move 41, level 71 and still hadn't found a way to finish. Thank you for sharing this. I can't imagine how a human player would figure all this out at the board though! Gui was Fritz 17/ 14CPUs. Some PC info from CPUID, FWIW: 
Number of cores            16 (max 16)
     Number of threads      24 (max 24)
     Hybrid                  yes, 2 coresets
     Core Set 0            P-Cores, 8 cores, 16 threads
     Core Set 1            E-Cores, 8 cores, 8 threads
     Manufacturer            GenuineIntel
     Name                  Intel Core i9 12900KF
     Codename            Alder Lake
     Specification            12th Gen Intel(R) Core(TM) i9-12900KF
Core Speed            5100.0 MHz

If you use SCID, etc., or if you ponied up for CB I'd be curious what you or others got. Again, thx!
  
Back to top
 
IP Logged
 
an ordinary chessplayer
God Member
*****
Offline


I used to be not bad.

Posts: 1727
Location: Columbus, OH (USA)
Joined: 01/02/15
Schlechter - Walbrodt, Vienna 1898
02/20/24 at 03:44:29
Post Tools
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
White to move and win.

Quote:
The double attack is the key to the ending in weak color complexes. Where there are no reserve Pawn moves a tempo must be gained by the King and if this is not possible the defender will be able to hold the game. Thus in No. 252 there are two threats: Ne5xc6 (if the B moves) and Kf4-g5 (if the King moves). If it were Black's turn he could therefore just as well resign. ... But White to move cannot lose a tempo.
--Reuben Fine (1941) Basic Chess Endings, pg.241

Significantly, the same ending is omitted from Benko's 2003 revision. I'm reasonably certain that means Benko found Fine's conclusion incorrect.

Give it a go, see if you can find the win for white. It's not easy, I remember wrestling with this 20 years ago, but I can't find my old analysis. Attached is a PGN with Fine's white-to-move analysis (see notes at move 60), plus the chessgames.com kibitzing, and a couple of additions by me ("aoc"). Schlechter won, but Walbrodt made a mistake with 66...Kf7?, instead 66...Kf6 was tougher.

When I switched on the engine, at first it gave white +1, this grew to +3, then sank back to +0.5, before growing unbounded to +16 or so, when I stopped it. The machines are so strong these days, if I had let it run overnight it might have announced a forced mate.

https://www.chessgames.com/perl/chessgame?gid=1120991

Later if needed I'll post the solution.
  

Schlechter-Walbrodt-1898.pgn ( 2 KB | 38 Downloads )
Back to top
 
IP Logged
 
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo