FEN:3r3r/pk3p1b/1pn1p2p/2p5/Q3P3/2PqB3/PP1N1P2/2KR2R1 w - - 0 22
I reached this position in a 5 minute blitz game. After 1:14 of thinking time I decided on 22. Bxc5 and went on to win the game. In post game analysis, engines (Komodo 10.3 and Houdini 5.01) seem to prefer 22. Bxh6. From a strategic point of view, it seems the engines are off track as whites chances of winning the game would appear to be on queenside.
What are your thoughts and opinions? Thanks.
PS: Play continued ...
22. Bxc5 bxc5
23. Nb3 Qf3
24. Nxc5
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
AdminX wrote: FEN:3r3r/pk3p1b/1pn1p2p/2p5/Q3P3/2PqB3/PP1N1P2/2KR2R1 w - - 0 22
I reached this position in a 5 minute blitz game. After 1:14 of thinking time I decided on 22. Bxc5 and went on to win the game. In post game analysis, engines (Komodo 10.3 and Houdini 5.01) seem to prefer 22. Bxh6. From a strategic point of view, it seems the engines are off track as whites chances of winning the game would appear to be on queenside.
What are your thoughts and opinions? Thanks.
PS: Play continued ...
22. Bxc5 bxc5
23. Nb3 Qf3
24. Nxc5
Hi Ted, please try to post a normal diagram, it is easier to make sense what is going on there.
first glance at the position on a brown background (and I am much better on a green background, at least 500 elo), draw is the natural outcome.
Lyudmil Tsvetkov wrote:to what leads Bc5 b5? presumably a draw?
on a brown background I would play Rg7 as white; on a green background, I would probably find some better moves.
Hi Lyudmil,
Yes, exactly the position is still equal. However now blacks king is a little bit more exposed, thus giving white more opportunity to threaten black's king. I really don't see why engines prefer taking black's pawn on h6 as the bishop seems better placed on the g1-a7 diagonal. Strategically taking on h6 feels like white is going off course. To me taking on h6 appears to only help black's cause in the long run.
PS: 500 elo, man I wish changing the board would give me 500 elo in OTB chess.
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Agreed, I don't think black should accept the sac. The position would still be even, but I like black's chances better.
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
Lyudmil Tsvetkov wrote:to what leads Bc5 b5? presumably a draw?
on a brown background I would play Rg7 as white; on a green background, I would probably find some better moves.
Hi Lyudmil,
Yes, exactly the position is still equal. However now blacks king is a little bit more exposed, thus giving white more opportunity to threaten black's king. I really don't see why engines prefer taking black's pawn on h6 as the bishop seems better placed on the g1-a7 diagonal. Strategically taking on h6 feels like white is going off course. To me taking on h6 appears to only help black's cause in the long run.
PS: 500 elo, man I wish changing the board would give me 500 elo in OTB chess.
man, in OTB it is more like 1000 elo.
I guess at least 2 factors make above diagram more drawish, so it almost does not make a distinction what one plays:
- opposite bishops
- both sides have exactly the same number and type of pieces
to me it seems an issue of completely clinical analysis, which is, at least by default, what engines give...if winning chances are taken into account (ie if white is to play practically) Bxc5 is superior. but if its technically drawn, it doesnt matter.
so imo this would be a good test of an engines contempt setting; it'd be nice to see contempt enabled would = preferring Bxc5.
yanquis1972 wrote:to me it seems an issue of completely clinical analysis, which is, at least by default, what engines give...if winning chances are taken into account (ie if white is to play practically) Bxc5 is superior. but if its technically drawn, it doesnt matter.
so imo this would be a good test of an engines contempt setting; it'd be nice to see contempt enabled would = preferring Bxc5.
Hi John,
Your answer is probably best from a technical point view and one which I had not considered. It goes a long way to explaining why they (engines) prefer 22. Bxh6.
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
i think it highlights an issue that will only get worse as engines get stronger. sort of the chinook effect, where a hypothetical machine thats solved chess as a draw will evaluate 1.e4 and 1.h4 as equally good moves, until its given contempt parameters...i hope we'll see programmers investing a lot of time in contempt over the next few years, but it might take longer, since we're still chasing elo.
