Test position for you and your engines

[Arpad Rusz]

The following position is reached after move 5 by black in one of my studies (study no.34 in my blog):
[d]8/2r1q2p/3P3P/1p6/kP6/Pp6/1P6/KB6 w

[Ovyron]

Is it that the pawn has to capture the rook? My engines are completely lost...

[Mike S.]

8/2r1q2p/3P3P/1p6/kP6/Pp6/1P6/KB6 w - - 0 6

I found the study on your blog, in the entry of March 9th, 2008. - The only thing I saw quickly is the stalemate after 6.dxc7 Qxc7 7.Bxh7 Qxh7.

In the variations after 6.dxe7, engines have quickly shown me that a rook sac on c2 could be 'interesting', but the rest of the study (in the main line with bQ) after the h-pawns are gone, is very difficult especially for engines. I think engines will never play 10...Qc2 in the main line, maybe with the theoretical exception that the have a kind of progress detection.

There is a logical problem though: That move doesn't improve the result anyway. So, even if engines would correctly evaluate both the continuations with, or without that sac, they would be identical, or maybe 0.01 better for the alternative continuation which keeps the large material advantage. In other words, the choice there is only between 1. draw with large material advantage, or 2. draw with a small material advantage. So, we cannot really blame the engines for choosing (1.)...

[George Tsavdaris]

8/2r1q2p/3P3P/1p6/kP6/Pp6/1P6/KB6 w - - 0 6

I found the study on your blog, in the entry of March 9th, 2008. - The only thing I saw quickly is the stalemate after 6.dxc7 Qxc7 7.Bxh7 Qxh7.

In the variations after 6.dxe7, engines have quickly shown me that a rook sac on c2 could be 'interesting', but the rest of the study (in the main line with bQ) after the h-pawns are gone, is very difficult expecially for engines. I think engines will never play 10...Qc2 in the main line, maybe with the theoretical exception that the have a kind of progress detection.

There is a logical problem though: That move doesn't improve the result anyway. So, even if engines would correctly evaluate both the continuations with, or without that sac, they would be identical, or maybe 0.01 better for the alternative continuation which keeps the large material advantage. In other words, the choice there is only between 1. draw with large material advantage, or 2. draw with a small material advantage. So, we cannot really blame the engines for choosing (1.)...

Yes but we can blame engines for not being able to find 6.dxc7!! that draws and instead prefer 6.dxe7? that loses.

To blame those engines that don't prefer dxc7 i meant.