StockFish : best move not found due to MultiPV ?

Discussion of chess software programming and technical issues.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Sesse
Posts: 142
Joined: Mon Apr 30, 2018 9:51 pm
Contact:

Re: StockFish : best move not found due to MultiPV ?

Post by Sesse » Fri Jul 20, 2018 7:38 am

tissatussa wrote:
Thu Jul 19, 2018 11:36 pm
At MultiPV 5, it shows up already at d25… as the fourth best move.
i think b4! should soon (after a small depth) be in the top # MultiPV list when alternative evaluation values are used / developed.
OK, you're free to think that. But I don't think it's worth continuing this further—you can maybe consider yourself which of one is “too rigid” in their thoughts.

jorose
Posts: 155
Joined: Thu Jan 22, 2015 2:21 pm
Location: Zurich, Switzerland
Full name: Jonathan Rosenthal

Re: StockFish : best move not found due to MultiPV ?

Post by jorose » Fri Jul 20, 2018 10:20 am

tissatussa wrote:
Thu Jul 19, 2018 11:36 pm
At MultiPV 5, it shows up already at d25… as the fourth best move.
i think b4! should soon (after a small depth) be in the top # MultiPV list when alternative evaluation values are used / developed. Do all engines show b4! at d25 when MultiPV is 5 ? I guess their evaluation top # will be different. But the move b4! should be considered at a possible best move and so given a high priority for deeper search. Are you a chess player ? What is your ELO ? Mine is 1870 (in Holland). Did you see the videos ? They show that positional play gives certain (and most of the times, a very few) moves to continue the idea of harmony in the piece team - it's a war game of options, the execution comes later ..
They don't “obviously fail”, because b4 is not the best move in the position.
It should be in top # MultiPV from start.

I hope you get my point now. I think you're too rigid in following the existing, known brute force methods.
I'm not sure why you are mentioning rating, but mine is above 2300 FIDE if we are going down that route. It is not clear to me that b4 should be the best move. Sesse has been far more patient with you than I could possibly be myself. Please read the links that all the helpful talkchess users have posted and stop posting all the random videos pointlessly overlayed with music.
-Jonathan

syzygy
Posts: 4310
Joined: Tue Feb 28, 2012 10:56 pm

Re: StockFish : best move not found due to MultiPV ?

Post by syzygy » Fri Jul 20, 2018 6:09 pm

tissatussa wrote:
Thu Jul 19, 2018 11:36 pm
I hope you get my point now. I think you're too rigid in following the existing, known brute force methods.
So you are suggesting to tweak an existing, known brute force method until it does what you like it to do in this one particular position. And you probably think that would measurably increase the engine's strength?

Well, you are faced here with people who actually do have experience in this field. So forget it.

Or... do the tweaking yourself and run it on fishtest (http://tests.stockfishchess.org/tests) and you might be the last one to laugh. Unlikely though...

User avatar
tissatussa
Posts: 24
Joined: Sat Sep 24, 2016 2:13 am
Location: Netherlands
Full name: Roelof Berkepeis
Contact:

Re: StockFish : best move not found due to MultiPV ?

Post by tissatussa » Fri Jul 20, 2018 6:20 pm

you probably think that would measurably increase the engine's strength
No, i think it will not lead to an overall increase of the engine's strength, but such tuning can give more insight in what our game is about : pieces should be on "good squares" and have options, regardless of all possible continuations. Pawns and even pieces can be sacrificed for that. Leela Zero understands such positional play.
-simple is not always best but best is always simple-

Dann Corbit
Posts: 8841
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: StockFish : best move not found due to MultiPV ?

Post by Dann Corbit » Fri Jul 20, 2018 9:17 pm

Straightforward analysis shows that Bd2 or Qg4 are pretty good, giving near equality.
3q1rk1/prppnp2/1p4p1/2b1P1N1/5P2/7P/PP5K/R1BQR3 w - - acd 50; acs 3600; bm Bd2; c0 "Qg4 Nf5 Bd2 is a transposition for Bd2 Nf5 Qg4."; ce -10; pm Bd2; pv Bd2 Nf5 Qg4 Bd4 Rad1 Qe7 Ne4 d6 exd6 cxd6 Ng3 Qd7 Nxf5 Qxf5 Qxf5 gxf5 Bb4 Bxb2 Rg1+ Kh7 Rg5 Re8 Rh5+ Kg6 Rg5+ Kh6 Rxd6+ Re6 Rd5 Re2+ Kg3 Re3+ Kh2 Bg7 Rgxf5 f6 Rd2 Rc7 Rg2 Rd7 h4 Kh7 Bd2 Red3 Bb4 R7d4 Be7 Rd5 Rxd5 Rxd5 Re2 Rd4 f5 Rd5 Re6 Rxf5 Kh3;

This analysis of the forward position after moving to Qg4 from the original position appears to approximately confirm the hypothesis:
3q1rk1/prppnp2/1p4p1/2b1P1N1/5PQ1/7P/PP5K/R1B1R3 b - - acd 47; acs 9960; bm Nf5; ce 15; pm Nf5; pv Nf5 Bd2;
Same for the position after Bd2:
3q1rk1/prppnp2/1p4p1/2b1P1N1/5P2/7P/PP1B3K/R2QR3 b - - acd 47; acs 9960; bm Nf5; ce 15; pm Nf5; pv Nf5 Qg4;

After b4, black appears to have about 1/2 pawn advantage.
3q1rk1/prppnp2/1p4p1/2b1P1N1/1P3P2/7P/P6K/R1BQR3 b - - acd 52; acs 9960; bm Bxb4; ce 46; pm Bxb4; pv Bxb4 Rg1;

Do you have a winning plan for white for the position after b4 has been performed? If so, what is it?
Our intuition can be wrong, on occasion.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Post Reply