Piket vs Smirin 1993 : CB claims 31.Qxd7 is better than Qa4

[syzygy]

Sorry for an offtopic post. Maybe my question would merit a different thread on its own or maybe even already discussed on old threads previously and heavily debated. If I understood engines correctly, the eval score was taken from the last leaf of any PV. If the whole PV cannot be trusted up to the end, then logically the eval score of that PV cannot be trusted too, or is this understanding of mine incorrect?

Serious and honest question since I am only a newbie and not a serious engine programmer and only understood some programming fundamentals with few actual programming experience.

Maybe it helps to look at it in the following way.

If an engine searches N plies deep, the score and best move returned are a bit more reliable than when the engine searches N-1 plies deep. This is because the engine looked 1 ply (half a move) deeper into what could happen.

However, an N-ply search of the position on the board will search the position immediately after the best move only N-1 plies deep. So the second move in the PV has the same quality as the best move returned by an (N-1)-ply search of the board position.

In the same way we see that the third move of the PV will have the quality expected of an (N-2)-ply search. And the fourth move will have the quality expected of an (N-3)-ply search, etc.

The last move of the PV will have the quality of an 1-ply search. That's pretty bad. But the quality of the best move (first move of the PV) will still be very good if N is large.

[MikeGL]

Sorry for an offtopic post. Maybe my question would merit a different thread on its own or maybe even already discussed on old threads previously and heavily debated. If I understood engines correctly, the eval score was taken from the last leaf of any PV. If the whole PV cannot be trusted up to the end, then logically the eval score of that PV cannot be trusted too, or is this understanding of mine incorrect?

Serious and honest question since I am only a newbie and not a serious engine programmer and only understood some programming fundamentals with few actual programming experience.

Maybe it helps to look at it in the following way.

If an engine searches N plies deep, the score and best move returned are a bit more reliable than when the engine searches N-1 plies deep. This is because the engine looked 1 ply (half a move) deeper into what could happen.

However, an N-ply search of the position on the board will search the position immediately after the best move only N-1 plies deep. So the second move in the PV has the same quality as the best move returned by an (N-1)-ply search of the board position.

In the same way we see that the third move of the PV will have the quality expected of an (N-2)-ply search. And the fourth move will have the quality expected of an (N-3)-ply search, etc.

The last move of the PV will have the quality of an 1-ply search. That's pretty bad. But the quality of the best move (first move of the PV) will still be very good if N is large.

Wow that was precise, terse and very clear. Thanks for your time in explaining this.

.