Patricia 2.0 - likely the most aggressive chess engine ever made

[pohl4711]

What we definitly need is a skill-level parameter in Patricia for making the engine playing weaker. Then this engine would be very cool for playing as a human against it (perhaps on a electronic chessboard, using the Android-Binary on a Smartphone). Right now, Patricia is way to strong for using it as an opponent for humans.

That is planned as well! I think a good way to go about doing it is instead of just having it make random blunders, try and determine what sort of blunder a "human" would make - this requires having Patricia make "non obvious" mistakes (for example, it won't hang a piece, but it might hang a tactic), which will be fun to try and implement

Thats the "hard way" to do this... As a quick-shot, you could just make a node-limit with different settings. Example: The TheKing-Chesscomputer offers levels with a fixed number of nodes (this number is increased automatically in the endgame, to prevent a huge strength-loss in the endgame (clever idea, IMHO)). From 125 nodes (lowest skill) to 32000 nodes (highest skill).

For me, this makes much more fun to play, compared to an engine, making mistakes on purpose. And a node-limit does not affect the engine's playing-style. And it is easy to implement. Just a node limit in the calculation, increasing it, when the endgame begins or so.

[pohl4711]

I made a small site on my website about Patricia...
Since the 1950s, the only goal of computerchess was gaining Elo. But in these days of superstrong engines beyond 3700+ Elo, IMHO it makes a lot of sense, to make engines playing more spectacular, aggressive and interesting, instead of just gaining more and more Elo... Since my EAS-Tool was made, it is possible for the first time, to measure the aggressiveness of engines. And so, using the EAS-Tool, to make an engine playing more aggressive, is the next logical step of development in computerchess.

https://www.sp-cc.de/patricia_eas_engine.htm

[BrendanJNorman]

In case anyone is interested in an Android version.

Patricia 2.0 for Android:
https://sourceforge.net/projects/chess- ... p/download

Patricia 2.0 for Android (OEX):
https://sourceforge.net/projects/chess- ... k/download

Absolutely love your work, my friend.

Do you have an OEX version of Pedone 3.0 for Android?

Here:
http://kalab.com/pedone.apk

Thank you very much!

[BrendanJNorman]

What an incredibly fun engine!

I'm not sure about this statement though as it got into a lot of trouble against a 2300 Elo engine:
"It's worth noting that Patricia is still well into superhuman territory, with an estimated CCRL elo of 3100-3150; she'll lose badly to top engines, but will still crush any human."

Thanks for the kind words!
She's absolutely superhuman, I can guarantee you that, however I wouldn't be surprised if she sometimes had a propensity to mess up with some ridiculous sacrifice against a weaker engine.

Do you think you’ll make it multi threaded in the future? (I can only set to 1 thread)

Yes, multithreading is definitely planned.

What we definitly need is a skill-level parameter in Patricia for making the engine playing weaker. Then this engine would be very cool for playing as a human against it (perhaps on a electronic chessboard, using the Android-Binary on a Smartphone). Right now, Patricia is way to strong for using it as an opponent for humans.

That is planned as well! I think a good way to go about doing it is instead of just having it make random blunders, try and determine what sort of blunder a "human" would make - this requires having Patricia make "non obvious" mistakes (for example, it won't hang a piece, but it might hang a tactic), which will be fun to try and implement

This engine has me more excited than I've been for a while about computer chess.

The way it plays is really like an aggressive human grandmaster (with the same over-zealous flaws, at times too).

Does this look like a computer is playing?

[pgn][Event "Pat 2"]
[Site "DESKTOP-Q1QR5FU"]
[Date "2024.04.03"]
[Round "1"]
[White "Patricia 2"]
[Black "Bright-0.4a"]
[Result "1-0"]
[ECO "B01"]
[WhiteElo "2200"]
[BlackElo "2200"]
[PlyCount "59"]
[EventDate "2024.??.??"]

1. e4 d5 2. exd5 Qxd5 3. Nc3 Qa5 4. d4 Nf6 5. Bd2 Qb6 6. Nf3 e6 7. g4 h6 8. Bc4
Be7 9. Qe2 Bd7 10. O-O-O Nc6 11. h4 Nxg4 12. Rde1 O-O 13. Rhg1 h5 14. d5 Na5
15. Bd3 Qxf2 16. Rxg4 Qxe2 17. Rxg7+ Kxg7 18. Nxe2 b6 19. Nf4 Rh8 20. Bc3+ f6
21. dxe6 Be8 22. Nd5 Kf8 23. Nxf6 Bxf6 24. Bxf6 Rh6 25. Bg5 Kg7 26. Nd4 Rg6 27.
Bxg6 Kxg6 28. Rf1 Nc6 29. Rf6+ Kg7 30. Nf5+ 1-0

[/pgn]

Not at all! It looks like a Kasparov simul game.

As for weakening it, yes this would be amazing.

Imagine dumbing it down to 2200 (approx) but keeping the same style? The perfect training opponent for most people.

[BrendanJNorman]

I made a small site on my website about Patricia...
Since the 1950s, the only goal of computerchess was gaining Elo. But in these days of superstrong engines beyond 3700+ Elo, IMHO it makes a lot of sense, to make engines playing more spectacular, aggressive and interesting, instead of just gaining more and more Elo... Since my EAS-Tool was made, it is possible for the first time, to measure the aggressiveness of engines. And so, using the EAS-Tool, to make an engine playing more aggressive, is the next logical step of development in computerchess.

https://www.sp-cc.de/patricia_eas_engine.htm

Another interesting challenge would be to make a completely positional engine with a very shallow search (enough to be as weak as humans tactically), but have a ton of positional knowledge.

Kind of like Korchnoi or Karpov in their older days.

This would be cool.

[pohl4711]

Look at this game of Patricia 2.0 vs Rybka 4.1 from my ongoing testrun (Patricia 2.0 scores around 50% vs Rybka 4.1)...
Balanced openings. Engines started thinking from move 9 of white.
Just madness. I love it!

[pgn]
[Event "Patricia 2.0 testing"]
[Site "?"]
[Date "2024.04.04"]
[Round "6"]
[White "Rybka 4.1"]
[Black "Patricia 2.0"]
[Result "0-1"]
[ECO "A25"]
[Opening "English"]
[Variation "Closed system"]
[TimeControl "180+1"]
[PlyCount "96"]
[GameDuration "00:06:25"]
[GameEndTime "2024-04-04T09:55:24.957 Mitteleuropäische Sommerzeit"]
[GameStartTime "2024-04-04T09:48:59.262 Mitteleuropäische Sommerzeit"]

1. c4 e5 2. Nc3 Nc6 3. g3 g6 4. Bg2 Bg7 5. e3 d6 6. Nge2 f5 7. d4 e4 8. b4
Nf6 9. O-O O-O 10. a3 Ne7 11. Bb2 d5 12. c5 f4 13. Nxf4 c6 14. Kh1 Qe8 15.
Nh3 g5 16. Nxg5 Bh6 17. Nh3 Bg4 18. Qc2 Qh5 19. Ng1 Nf5 20. Rae1 Bf3 21. h3
Nh4 22. gxh4 Bxg2+ 23. Kxg2 Kh8 24. Kh1 Rg8 25. Bc1 Raf8 26. b5 Rg7 27.
bxc6 bxc6 28. Qd2 Rfg8 29. Rd1 Rg2 30. Nce2 Qxh4 31. Ng3 Rxg1+ 32. Kxg1
Qxh3 33. Qe2 Ng4 34. Qxg4 Rxg4 35. Bd2 Bg7 36. Ba5 h5 37. Bd8 h4 38. Bxh4
Rxh4 39. Rfe1 Bf6 40. Rd2 Bd8 41. Rb1 Bc7 42. Rdb2 Bxg3 43. Rb8+ Bxb8 44.
Rxb8+ Kg7 45. Rb7+ Kh6 46. Rh7+ Kxh7 47. f4 Qf3 48. a4 Rh1# 0-1
[/pgn]

[BrendanJNorman]

Look at this game of Patricia 2.0 vs Rybka 4.1 from my ongoing testrun (Patricia 2.0 scores around 50% vs Rybka 4.1)...
Balanced openings. Engines started thinking from move 9 of white.
Just madness. I love it!

[pgn]
[Event "Patricia 2.0 testing"]
[Site "?"]
[Date "2024.04.04"]
[Round "6"]
[White "Rybka 4.1"]
[Black "Patricia 2.0"]
[Result "0-1"]
[ECO "A25"]
[Opening "English"]
[Variation "Closed system"]
[TimeControl "180+1"]
[PlyCount "96"]
[GameDuration "00:06:25"]
[GameEndTime "2024-04-04T09:55:24.957 Mitteleuropäische Sommerzeit"]
[GameStartTime "2024-04-04T09:48:59.262 Mitteleuropäische Sommerzeit"]

1. c4 e5 2. Nc3 Nc6 3. g3 g6 4. Bg2 Bg7 5. e3 d6 6. Nge2 f5 7. d4 e4 8. b4
Nf6 9. O-O O-O 10. a3 Ne7 11. Bb2 d5 12. c5 f4 13. Nxf4 c6 14. Kh1 Qe8 15.
Nh3 g5 16. Nxg5 Bh6 17. Nh3 Bg4 18. Qc2 Qh5 19. Ng1 Nf5 20. Rae1 Bf3 21. h3
Nh4 22. gxh4 Bxg2+ 23. Kxg2 Kh8 24. Kh1 Rg8 25. Bc1 Raf8 26. b5 Rg7 27.
bxc6 bxc6 28. Qd2 Rfg8 29. Rd1 Rg2 30. Nce2 Qxh4 31. Ng3 Rxg1+ 32. Kxg1
Qxh3 33. Qe2 Ng4 34. Qxg4 Rxg4 35. Bd2 Bg7 36. Ba5 h5 37. Bd8 h4 38. Bxh4
Rxh4 39. Rfe1 Bf6 40. Rd2 Bd8 41. Rb1 Bc7 42. Rdb2 Bxg3 43. Rb8+ Bxb8 44.
Rxb8+ Kg7 45. Rb7+ Kh6 46. Rh7+ Kxh7 47. f4 Qf3 48. a4 Rh1# 0-1
[/pgn]

This game is absolutely unbelievable. Looks like one of those old games where a GM beats the engine with "anti-computer" play.

And I can remember a time when Rybka 4 was considered absolutely the elite of chess engines.

[pohl4711]

This game is absolutely unbelievable. Looks like one of those old games where a GM beats the engine with "anti-computer" play.

And I can remember a time when Rybka 4 was considered absolutely the elite of chess engines.

Yep, Patricia 2 is the resurrection of M.Tal. But a 3100-Elo-Tal... This is just awesome, but - oh boy - I need a skill-level system in Patricia...
And mention, Patricias development has just started some months ago. Cant wait to see the things, that are coming up in the future!!!
And all that started with my little EAS-Tool. So cool.

[BrendanJNorman]

This game is absolutely unbelievable. Looks like one of those old games where a GM beats the engine with "anti-computer" play.

And I can remember a time when Rybka 4 was considered absolutely the elite of chess engines.

Yep, Patricia 2 is the resurrection of M.Tal. But a 3100-Elo-Tal... This is just awesome, but - oh boy - I need a skill-level system in Patricia...
And mention, Patricias development has just started some months ago. Cant wait to see the things, that are coming up in the future!!!
And all that started with my little EAS-Tool. So cool.

Would it be possible to just "borrow" the limit-strength code from Pedone 3.0?

Or is Pedone 3.0 not open source?

[reflectionofpower]

Any plans on making a SMP version?