chess

Chess

Classify and you will gain!
Funny name, real hobbies
[ Sign up | Log in | Guest ] (beta)
tim_b 65 ( +1 | -1 )
There are correct sacrifices and.... Hi folks. I enjoyed Craig’s thread on miniature games and so I thought it would be nice to have something similar, except this time people get to shamelessly show off some of their favourite games where a nice sacrifice or two paid off.

Normally I get the opportunity to give up a minor piece for some clear advantage, but this recent game got my heart thumping. The sacrifices begin on move 20. I think it would be interesting to not only share games, but see if anyone else can come up with surprise rebuttals to sacrifices.

Thanks.

board #7762861
sf115 15 ( +1 | -1 )
The game you showed wasn't really a sacrifice because there was a immediate combination to regain the piece. We need some proper sacrifices
heinzkat 53 ( +1 | -1 )
A sacrifice is a sacrifice.

Might be worthwhile:
board #3616421 (Rxg6)
board #6654426 (Bxh6)
board #7009507 (Qh6+)
board #7547421 (Rb1+)

And more recent:
board #7780484 (Re8+)
ionadowman 244 ( +1 | -1 )
I tend towards a more liberal view... ... nice combination tim_b! There are sacrifices that lead to some more or less immdiate gain, and those that are speculative: the gains are other than material. Even temporary sacrifices to gain material can be interesting or exciting owing, possibly, to the difficulty of finding the idea.
Here's a game from my early days in which the closing sacrifices look pretty good, but they weren't at all difficult to find...
White: A. McL. Black: I.A.D. Latvian Gambit Auckland University 1973.
1.e4 e5 2.Nf3 f5 3.exf5 e4 4.Qe2 d5 5.d4 Nf6
6.Ne5 Bxf5 7.Bg5 Nbd7 8.Nxd7 Qxd7 9.Bxf6 gxf6 10.f3 0-0-0
11.Nd2 Bh3 12.fxe4 Rhe8 13.0-0-0? Bg4 14.Qb5 c6 15.Qa4 Bxd1
Having won the exchange, Black ought to win fairly easily from here...
16.Kxd1 Qg4ch 17.Nf3 Qf4 18.Ke2 Qe3ch?!
Prolongs the game by 2 moves. The right move is played on move 20.
19.Kd1 Qc1ch 20.Ke2 Rxe4ch 21.Kf2 Qc3ch 22.Kg3 Rg8ch 23.Kh3...
b
Now comes the final combination:
23...Rh4ch 24.Kxh4 Qf4ch 25.Kh5 Qg4ch 26.Kxh6 Qg6 mate.
The whole sequence from move 17 was forced...

From a correspondence game of the same year, I found myself sacrificing the exchange in what I thought was sheer desperation. Turns out it was probably correct:
White: I.A.D. Black: G.K. CC 1973 Ruy Lopez/Marshall Attack
1.e4 e5 2.Nf3 Nc6 3.Bb5 ... I very rarely play the Lopez..
3...a6 4.Ba4 Nf6 5.0-0 Be7
6.Re1 b5 7.Bb3 0-0 8.c3 d5!?
The Marshall. I like it as Black, but now I have to defend against it...
9.exd5 Nxd5 10.Nxe5 Nxe5
11.Rxe5 c6 12.d4 Bd6 13.Re1 Qh4 14.g3 Qh3 15.Be3 Bg4
16.Qd3 Nxe3 17.Rxe3 c5 18.Bd5 Rad8
So far, so book. Fischer-Donner a few years before continued 19.Nd2 Bb8?! 20.Bg2 Qh6 21.d5 c4?! 22.Qd4 etc. My opponent here improves on Donner's play:
19.Bg2 Qh6 20.d5 f5!
21.Nd2 c4! 22.Qd4 f4! 23.Ree1 Bc7!
A better square than b8, for sure: heading for b6 of course.
24.Re7!
Having the next move in mind... Frankly, I thought Black had a superb game at this point...
24...Bb6
w
25.Rxg7ch! Qxg7 26.Qxb6 Qf6?!
During the game I thouht this was a mistake, but it's hard to find an improvement...
27.Qxf6 Rxf6 28.Ne4 Rff8 29.f3 Bc8 30.Rd1 Bb7
31.d6 Bxe4 32.fxe4 f3 33.Bh3 Rde8 34.Bf5 a5 35.Kf2 b4
36.g4 bxc3 37.bxc3 Rb8 38.Kxf3 ...
At this point, 13 months having elapsed, came the time for adjudication. White has a straightforward win (3 pawns, two central passers at that, for the exchange), so 1-0.
At the time I was quite proud of the sac on g7, but was surprised how quickly the tables were turned...

Cheers,
Ion
heinzkat 51 ( +1 | -1 )
Good straight to the point and sound attacking, by the way, tim_b.

I forgot to include the ultimate sacrificing game that I played here on GameKnot:

-> gameknot.com
or
-> gameknot.com
or
board #6679109
ionadowman 172 ( +1 | -1 )
The first of the games... ... in heinzkat's last posting I highly recommend: a hugely entertaining fighting game, with amazing sacrifices for attack. I must check out the other games...

I especially enjoy sacrifices in the endgame, where material considerations start to loom very large. The diagram position is from a game I played against a German traveller who happened to be visiting this part of the world a few years ago.
w
Sure, White had an extra bishop, but Black's two extra pawns and active position was beginning to make the win look problematical. It's too early to call this an endgame, but it seemed that a major readjustment of the material situation would improve White's chances:
1.Bxh7ch! Kxh7
...else White's last becomes a smash and grab raid...
2.Qxf7ch Kh6
... 2...Qg7 is no better after 3.Qh5ch Kg8 4.Rd6 etc.
3.Rd7 Rh8
4.Rd5!? ...
This will do, but 4.Rd3 g4 5.Rd5 is more accurate.
4... Qg7
5.Rd6ch Kh7
6.Qf5ch Kg8
7.Rg6 Rh7
8.Rxg7ch!? ...
Still good enough, but the g5-pawn was available gratis...
8...Rxg7
9.Qxc5 Rf8
Black could equally well have resigned, but hoped that maybe something could be achieved on the king-side with those advanced pawns.
w
10.Qxf8ch ...
Now this really is hard to call a sacrifice. This is one of those situations that you kind of see at a glance, without calculating specific moves: the heavy pieces are forced off and the BK finds itself stranded too far away to stop the a-pawn.
10...Kxf8 11.Rc8ch Kf7 12.Rc7ch Kf6 13.Rxg7 Kxg7 14.a4
Black played on few more moves but the game is already decided.


tim_b 10 ( +1 | -1 )
Cheers Thanks for all the feedback and games, I'm having fun going through them. Such a violent game ;)
himu 29 ( +1 | -1 )
Do you think this one is any good ? Waht motivated me to that sac is that white was playing too aloof not paying attension to black is doing, and that castle also was too early.

-> gameknot.com

himu
ionadowman 16 ( +1 | -1 )
Nice, himu: ... without going into it in depth, it looks like a well motivated speculative exchange sac that gave Black a lasting attack.
Cheers,
Ion
sf115 24 ( +1 | -1 )
Very good sacrifice and a very good follow up with the mini pawn storm and attack. But I'm sure that there must be a defense to the attack. For example I'm not sure about 25. h4. Can anyone else comment on this?
himu 54 ( +1 | -1 )
I also thought that 25.h4 was not a good move. But white is in a zugwang the queen is immobile and also both rooks have no job. for eaxample any other plausible move except pxp will be followed by Ba6 which will seal the whites bishops fate. At 30th move probably giving the exchange back would have prolonged the game but the two connected passed pawns after mass cancellation will be vital. (in the game I was worried about this because my endgame is not so good :) ) Thanks for the comments Ion and sf115
ionadowman 150 ( +1 | -1 )
Missed opportunities... Haven't we all wanted at some time or other to have played a queen sacrifice? Something to brag about. I have given up a queen on move 7, but that was in exchange for scattered material that was worth at least a queen. I have found my way out of difficult situations by a timely exchange of queen for rook, pawn and minor piece: a trade, really. I did once in a correspondence game sac minor piece and follow it up with a double rook sacrifice (or it would have been, except that the enemy resigned on the first rook - I wish I still had the score for that game....). But queen sacrifice? Nada.
Then just a few years ago, opportunity knocked... and went away unanswered. I even saw the early part of the combination, but stopped calculating too soon.
Here's the position:
w
Black is threatening mate in 2, but White has a strong attack.
What I ought to have played was;
23.Qxf7ch! Kxf7 24.Rxe7ch Kg6 (or ...Kf8 25.Re8ch 26.Kf7 R1e7ch 27.Kg6 etc)
25.Rg7ch Kxf5 (My original calculation stopped here, failing to notice the real implication of White's next..)
26.g4ch! ... (Black has just one legal move!)
26...Qxg4ch 27. hxg4ch Kf4 28.Re4ch Kf6 29.Rf7! (Putting the screws in and winning the second rook)
29... Rxf2 (Only move to avoid mate)
30.Rxf6ch wins the second rook.
Failing to calculate far enough, I played 23.f3 and won in much more prosaic fashion...
Cheers,
Ion

ccmcacollister 141 ( +1 | -1 )
Queen Sac? Ok, this is a Fav of mine... Here is the only one I can recall making at GK, right now. But I really enjoyed making it, and planning it as the reply if he should play Bxg4 sac, as he did. Thanks to my opponent for pushing on into that interesting sac, which was probably the only way to try to make headway vs my Kingside, to sac thusly at some point. And having done do, the position really does become very dangerous and critical for my king, and should at least provide him with enough material won back, if I try to play thru it conventionally after accepting his B-sac.
The think I really enjoy about this Queen sac is that i think a Computer would Never Ever Ever do such a thing as Sac a Queen to go into to lengthy, pawn down endgame...
which I judged to be a nearly forced win for me at the time. However, later I discovered he could have played 37.Bd3 for a better position than in the game, and perhaps put the outcome into doubt then. The way it goes however, with 37.e5 still leaves a won ending I believe, tho it can go into several very delicate varations, EG such as having to nursemaid in a single pawn to Queen without allowing it to be sac'd for. Or another of being left with N+B to mate with ... which I practiced on for a week to get up to speed again on those! But finally got there ...
I really like this thread ... well done everyone~!
board #4260172
ccmcacollister 16 ( +1 | -1 )
forgot the diagram ... This is just before ...Qxb2


kbaumen 68 ( +1 | -1 )
I know it's not a very complicated combination, but I still feel good for finding it.



1. ... Nd4! gives two threats, one of them harder to see. A fork Nxc2 and a strong attack with Rf3. I think that the best move for white would've been 2. Qg5 but even that loses a pawn to 2. ... Qxh3+! (if 3. gxh then 3. ... Nf3+). However, my opponent played 2. Rc1 unaware of the combination that was to follow. 2. ... Rf3!!. Now it's too late to play 3. Qg5 due to 3. ... Rxh3+ 4. gxh (4. Kg1 Rh1#) Qxh3+ 5. Kg1 Nxf3#. Here white should've played 3. Qxf3 though even that move loses, just not so fast.

Tell me if there are any flaws.
ionadowman 143 ( +1 | -1 )
Craig... ... The sac looks more like a rook sac, rather than a queen sac... But you still had to see quite a long way to make an assessment about thwe ensuing endgame. Sacrifices for such long-term positional considerations are in my view at least as interesting as those motivated by quick temporal or material gains... ;-)
Cheers,
Ion

Speaking of missed opportunities and rook sacrifices, here's one I "saw" and failed to play owing to time-trouble. It was the 3rd round game in a Labour Weekend tournament October 1973. I had the White pieces:
w
Although I did win the game shortly after the time control, I regret not playing
37.Qxe6! Qxg1 38.Qxf5ch etc with a forced mate, which would have been a fitting conclusion to a hard-fought fighting game, e.g.: 38...Ke7 39.Qe5ch Kf8 40.Qd6ch Kf7 41.Qc7ch! (A hard move to find in time trouble!) 41...Kf8 42.Qd8ch Kf7 43.Nd6ch Ke6 44.Bf5 mate. The reason for 41.Qc7ch rather than 41.Qd7ch is found if Black advances the king instead of retreating: 41...Ke6 42.Qe5ch (the square is inaccessible from d7) 42....Kd7 (42...Kf7 43.Nd6ch and mate next) 43.Qd6ch Ke8 44.Nc7ch Kf7 45.Qe6ch Kf8 46.Qe8 mate. The queen and knight make a wonderfully articulated attacking force.
Of course, Black could decline the rook, but with just one pawn for the piece, one imagines he wouldn't have been able to live long...
Cheers,
Ion
tag1153 17 ( +1 | -1 )
ion Wouldn't 37. Qxe6 also deiver check? Maybe I'm not understanding your example......


tag1153
heinzkat 19 ( +1 | -1 )
I think Ion's diagram is incorrect. 37. Qxg7 looks a bit like mate. Also, 37. Qxe6 is a check too, like tag already suggested. So 37. ... Qxg1 is impossible. Maybe Rook and King swapped squares?
ionadowman 21 ( +1 | -1 )
Whoops! I inverted Black's K and R in the diagram (I did check the thing, but it's easy to overlook, when you think about it)! The position was:
w
naamloos 206 ( +1 | -1 )
Queen sacrifices, If giving up material for immediate gains is called a sacrifice, then I made a queen sacrifice in the following game of which I was rather proud (being a mediocre player):



[White "Loos, Rudolf"]
[Black "g2003"]
[Result "1-0"]

1. e4 e5 2. Nc3 Nf6 3. Bc4 Nxe4 4. Qh5 Nd6 5. Bb3 Be7
--Nc6 Would have lead to complications after 6. Nb5. The played move seems good enough for equality. --
6.Nf3 Nc6 7. Nxe5 O-O 8. O-O Nd4 9. Nd5 Nxb3 10. axb3 Ne8 11. Re1 Nf6 12. Nxe7+Qxe7 13. Qd1 Qd8 14. d4 d6 15. Nd3
--(15. Nc4 !?) --
15... Bg4 16. f3 Be6 17. Bg5 h6 18. Bh4 g5 ?!
--Discontent with a draw against a lower rated player, my
opponent makes a rather risky move. --
19. Bf2 a5 20. f4 c6
--(20... Ng4 !? 21. Bg3 Qf6 22. c3 Qf5 23. fxg5 hxg5 ) --
21. fxg5 hxg5
--The position of the black king looks critical, but it is not so easy to break trough. --
22. Qd2 Ng4 23.Bg3
--(23. Be3 !? Nxe3 24. Rxe3)--
23... Kh8
--Re8 Is perhaps an improvement.--
24. h3 Nh6 25. h4 ?!
--This is dubious according to Fritz, but it isn't easy for black to find the right way. --
Nf5 26. Bf2 gxh4 27. Re4 Kg7
--(27... Qf6 !?) --
28. Rg4+ Kh7 29. Qf4
--(29. Rg5 Qf6 30. Rh5+ Kg7 31. Rg5+ This draw is objectively the best for white in this position, but I'm glad I did not take this route.) --
29... Rg8 30. Rxh4+ !?
--Perhaps also dubious, but white has already gone too far to back out.--
Nxh4 31. Bxh4 Qc7 ?
--(31... Qf8 32. Bg5 Rxg5 ! {Black has to return the quality.} 33.Qxg5 Qh6 34. Qf4 Qxf4 35. Nxf4 Kg7 With a light plus for black, altough it should be a draw.) --
32. Bf6 Rg6 ?!
--A second mistake indicating that black has missed the following tactic.
(32...Rg4 33. Qh2+ Kg6 Only move. 34. Rf1 And white has the advantage.) --
33. Kf2 !
--After this black has no choice other than the bleak endgame occuring in a few moves. --
Rh6

34. Qxh6+ !
--A lovely move to play, I was quite content with myself here, having seen this
sacrifice when I calculated what would happen after the rook sacrifice on move
30.--
Kxh6 35. Nf4 !
--This makes everything clear. Black will have give the material back to avoid mate.--
Qd7 36. Rh1+ Bh3 37. Rxh3+ Qxh3 38. gxh3
--The endgame is winning for white. --
a4 39. bxa4 Rxa4 40. Kg3 d5 41. h4 Ra2 42.Bg5+ Kg7 43. Nd3 f6 44. Bf4 Kg6 45. Kg4 f5+ 46. Kh3 Ra1 47. Bg5 Rg1 48. Kh2 Rd1 49. Kg3 Rg1+ 50. Kf4 Rg4+ 51. Ke5 Re4+ 52. Kd6 Rxd4 53. Kc7 c5 54. Kxb7 c4 55.Nf4+ Kf7 56. Kc6 Rd2 57. c3 d4 58. Kc5 Rxb2 59. Kxd4 Rd2+ 60. Kxc4
1-0

I'm sorry for the akward notation and possible mistakes therein, but the game has not been saved on gameknot so I can't give a direct link.
heinzkat 9 ( +1 | -1 )
Nice board #6122682
naamloos 11 ( +1 | -1 )
Oh So, it has been saved. I thought that games of non-paying members where a only saved for a few months or so.
tim_b 23 ( +1 | -1 )
Nice one.. naamloos ! It's very satisfying finding them, isn't it? Just makes me wonder how often I've stared blankly at a cluttered board and completely missed something.
naamloos 86 ( +1 | -1 )
I did I did miss it almost. When I came in my analyses to the position after blacks 33th move the first time, I did try the queen sacrifice immediately (looks natural with the king being so vulnerable), but initially thought that white had nothing (I naturally tried the immedate rookcheck). Only a few days later did I notice that Nf4 creates a mating net, out of which black can't free himself, except when he gives back the material with interest.
Ofcourse the game is for the rest not extremely interesting. g5 Was very weakening, but I did not make the best use of it. The pawn sacrifice was certainly wrong (I missed the intermediate Nf5) and the exchange sacrifice was obvious though not even that good. Still, I like this game more than some of the objectively better games I played. The queensacrifice remains the ultimate thing to do in a chessgame.
heinzkat 20 ( +1 | -1 )
Just for the fun of it: Two pawns, a Rook and a Bishop behind, but a King in return; not too difficult to find but pretty amusing to replay:
board #7853713
tim_b 8 ( +1 | -1 )
Cool. I don't think I've seen such an effective pawn storm for a while ;)
steverand67 37 ( +1 | -1 )
A couple of games: 9185
8080
9250

They're all annotations that I've made public, so If anyone is interested in the annotations they're on my profile.

P.S. Nice games everyone.


steverand67 38 ( +1 | -1 )
Oops... Sorry everyone, looks like I tried to list those games as users. Here are the games:

board #9185
board #9250
board #8080
ionadowman 34 ( +1 | -1 )
Very nice Q-sac by naamloos... ... One of those sacrifices in which the theme is easy to spot, but the correct follow-up his not so easy to find, since one tends to think of the rook check at once instead of a quiet, constricting move such as the knight's. A high-quality sac!.
I rather liked the ending as well: Black wasn't going to lie down easy
Cheers,
Ion!
heinzkat 30 ( +1 | -1 )
steverand67's games -> gameknot.com
-> gameknot.com
-> gameknot.com

kansaspatzer 12 ( +1 | -1 )
I was on the losing end of this one:

-> gameknot.com
kansaspatzer 27 ( +1 | -1 )
Here is an interesting sacrifice I played, from a mini-tournament in which castling is not allowed.

-> gameknot.com

I think 23.Rd3, sacrificing the a1 rook, could have been better met by 25...Qe6.
chasparos 11 ( +1 | -1 )
I'm stupid. Kansas: In the game you resigned what is the continuation? I can only make white win a piece. (24 ... Kb7 25. Nxe5+ Ka6) *blush*
kansaspatzer 2 ( +1 | -1 )
That was why I resigned.
kansaspatzer 3 ( +1 | -1 )
plus the passed pawn
sf115 23 ( +1 | -1 )
I've found a good queen sacrifice. Capablanca-Chase New York 1922

1. e4 e5 2. f4 exf4 3. Qf3 Nc6 4. c3 Nf6 5. d4 d5 6. e5 Ne4
7. Bxf4 g5 8. Be3 h5 9. Nd2 Bg4 10. Nxe4 Bxf3 11. Nf6+ Ke7
12. Nxf3 Bh6 13. Nxg5 Bg7 14. Bd3 Bxf6 15. O-O Qg8 16. Rxf6
Rf8 17. Raf1 Nd8 18. b4 Qg7 19. h4 Ke8 20. b5 b6 21. Be2 Ne6
22. Bf3 Nxg5 23. Bxg5 Qh7 24. Bxd5 Qd3 25. Bc6+ 1-0

a very good game.
heinzkat 20 ( +1 | -1 )
Capablanca-Chase -> gameknot.com
-> gameknot.com
ionadowman 44 ( +1 | -1 )
I knew I'd seen ... ... A Capablanca King's Gambit at some time or other, a Breyer Gambit (3.Qf3) at that. Something of a positional gambit, in which the pawn is given up in order - and here I want you to follow me closely - in order to obtain a better endgame. The follow up is more positional than tactical. Though there's no absence of the latter, Black remains permanently tangled and cramped, however hard he fights... Capablanca at his finest!
Cheers,
Ion
sf115 10 ( +1 | -1 )
without going into into masses of opening theory, how do you get an endgame advantage by moving 3.Qf3
ionadowman 62 ( +1 | -1 )
How... ... do you get a better ending by 3.Qf3? Ask Mr Breyer! ;-)
But seriously, broadly speaking the purpose of the gambit becomes to establish a Q-side pawn majority, with the kings, naturally, castled on the K-side. At least, so I have read somewhere. Possibly it is intended this early either to recover the f-pawn quickly (establishing thereby a central pawn majority), or to force black to overcommit to its defence. At the same time, a more restrained (restrained, forsooth!) development ensures that White's attack doesn't fizzle out too soon, as it does tend to do in the main lines.
Do I believe any of this? Dunno, really...
Cheers,
Ion
jstevens1 214 ( +1 | -1 )
I have some good/not too bad sacs to show you Here are two of my past games in which I was on the receiving end of a rook sacrifice. In the first case, the sacrifice was definitely good, in the second case I could have drawn with correct play.

[Event "Challenge from jstevens1"]
[Site "-> gameknot.com
[Date "2007.07.27"]
[Round "-"]
[White "jstevens1"]
[Black "NN"]
[Result "0-1"]
[WhiteElo "1590"]
[BlackElo "1717"]
[TimeControl "1/259200"]
[Mode "ICS"]
[Termination "normal"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Qb6 5. Nxc6 bxc6 6. Qd3 a5 7. Qb3 Qc7 8. Bc4 e6 9. Nd2 Nf6 10. Qe3 Bd6 11. Nf3 Be7 12. Bd2 d5 13. exd5 exd5 14. Bd3 O-O 15. O-O Bd6 16. Qe1 Ng4 17. h3 Nf6 18. Qc1 Ne4 19. Re1 Bf5 20. Bxe4 dxe4 21. Nd4 Bg6 22. Ne2 f5 23. Bf4 Rae8 24. Bxd6 Qxd6 25. Rd1 Qc5 26. c3 Bh5 27. Re1 e3 28. f3 f4 29. Kh2 Red8 30. Nd4 Qg5 31. Kh1 Rf6 32. Rg1 Rh6 33. Qc2 Rxd4 34. cxd4 Qg3 35. Qb3+ Kf8 36. Qa3+ Ke8 37. Qxa5 Bf7 38. Qa8+ Ke7 39. Qb7+ Kf6 40. Qxc6+ Be6 41. Rgc1 Kg5 42. Qb5+ Bf5 43. Rc6 Rxh3+ 44. gxh3 Qxh3+ 45. Kg1 Qg3+ 46. Kh1 Qxf3+ 47. Kg1 Qg3+ 48. Kh1 Qh3+ 49. Kg1 f3 50. Rc2 Qg3+ 51. Kh1 f2 0-1


[Event "38th GK tournament"]
[Site "-> gameknot.com
[Date "2007.09.16"]
[Round "-"]
[White "jstevens1"]
[Black "NN"]
[Result "0-1"]
[WhiteElo "1585"]
[BlackElo "1860"]
[TimeControl "1/172800"]
[Mode "ICS"]
[Termination "normal"]

1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 d6 6. Nxc6 bxc6 7. Bd3 Bg4 8. Be2 Qd7 9. Bxg4 Nxg4 10. f3 Ne5 11. Qe2 h5 12. Be3 Rb8 13. b3 e6 14. Nd1 Be7 15. Nf2 Qc7 16. O-O f6 17. Nd3 Nxd3 18. Qxd3 Kf7 19. c4 h4 20. h3 Rh5 21. Qe2 Rbh8 22. Bf2 Bd8 23. Qe3 c5 24. Rfd1 Qc6 25. Qd2 Bc7 26. Qe1 Rh8h6 27. Be3 Rg6 28. Kh1 f5 29. exf5 exf5 30. Qe2 Re6 31. Qf2 g5 32. Bd2 d5 33. cxd5 Qxd5 34. Bxg5 Qc6 35. Bxh4 Reh6 36. Bd8 Rh5xh3+ 37. gxh3 Rxh3+ 38. Kg1 Qg6+ 39. Kf1 Rh1+ 40. Ke2 Qa6+ 41. Kd2 Rh2 42. Qxh2 Bxh2 43. Kc2 Be5 44. Rab1 Qxa2+ 45. Kd3 Qf2 46. Kc4 Bd4 47. Rdf1 Qe2+ 0-1

Ionadowman had the following comment to make about the above game:-

"A closer look seems to indicate that 36.Rc1 would probably have held the draw:
36.Rc1 Rxh4 37.Rxc4 Bb6 38.Rxc6 Bxf2 39.Rxh6 Rxh6. Now, ordinarily this ought to be straightforward: White gives up his 5 pawns for Black's 2 and draws the R B vs R ending. Here it's not quite so simple, owing to the cramped King position - he could find himself imprisoned and mated in the corner.
So: 40.Rd2 Be3 41.Re2 f4 42.Kh2 hoping to break out by 43.g4. However, with 36.Bd8, it seems possible you had a draw in this line as well. 36.Bd8 Rxh3ch 37.gxh3 Rxh3ch 38.Kg1 Qg6ch 39.Kf1 Rh1ch 40.Ke2 Qa6ch 41.Ke3! (instead of 41.Nd2) and after 41.Qe6ch 42.Kd3 and White looks to be wriggling clear. Instead,
41...Rh2 42.Qxh2 Bxh2 43.Rd7ch Ke8 44.Rd2 and White still seems to be in the game.

So in the second case the sacrifice was unsound.

Hope all this helps.

jstevens1 (Joanne)