Yes, perhaps it’s a time management issue. I ran the match overnight and did not watch the games. I simply verified in the morning there were no abnormal terminations.
When I get home tonight I’ll test at 40/4 and update this thread with my results, Minic’s NPS, etc.
A complete 2000 lines of code engine
Moderators: hgm, Rebel, chrisw
-
- Posts: 434
- Joined: Thu Apr 26, 2012 1:51 am
- Location: Oak Park, IL, USA
- Full name: Erik Madsen
Re: A complete 2000 lines of code engine
My C# chess engine: https://www.madchess.net
-
- Posts: 1871
- Joined: Sat Nov 25, 2017 2:28 pm
- Location: France
Re: A complete 2000 lines of code engine
Ok, so here's last night results of Minic 0.28 (not really converged ... +/-60, ) at 40/0:20 :
Myrddin 0.87 is 2487, so 130 weaker is 2350. Sungorus is 2312 and zevra 1.8.6 is 2302.
The test at 40/1 against Isa 2.0.64 gives 137-113-69 for Isa so around +26elo for Isa that is 2329.
Minic seems to be between 2300 and 2400 (more 2300 than 2400). This is more or less coherent with Minic results versus Weini (or the stupid test against Fruit2.1).
I'll run an incremental TC test soon.
I guess Minic is weaker than Madchess. Madchess 3 has some good passer eval (taking king square, free, unstoppable into account). I often see Weini and Minic loss end game, I guess I should work on end game evaluation. Also Minic only see FIDE draws not "near" draw (like KRRK, KNKR, ...) so it might go to a drawish end-game thinking it can win...
Code: Select all
Rank Name Elo +/- Games Score Draws
1 myrddin 131 67 100 68.0% 16.0%
2 sungorus64 14 61 100 52.0% 22.0%
3 zevra -14 60 100 48.0% 24.0%
The test at 40/1 against Isa 2.0.64 gives 137-113-69 for Isa so around +26elo for Isa that is 2329.
Minic seems to be between 2300 and 2400 (more 2300 than 2400). This is more or less coherent with Minic results versus Weini (or the stupid test against Fruit2.1).
I'll run an incremental TC test soon.
I guess Minic is weaker than Madchess. Madchess 3 has some good passer eval (taking king square, free, unstoppable into account). I often see Weini and Minic loss end game, I guess I should work on end game evaluation. Also Minic only see FIDE draws not "near" draw (like KRRK, KNKR, ...) so it might go to a drawish end-game thinking it can win...
-
- Posts: 434
- Joined: Thu Apr 26, 2012 1:51 am
- Location: Oak Park, IL, USA
- Full name: Erik Madsen
Re: A complete 2000 lines of code engine
Looks like Minic has time management issues for games with incremental time controls. I ran another match between MadChess 3.0 Beta and Minic 0.28 using a time control of 40 moves in 4 minutes. Minic won the match.
The absolute ELO values are not relevant. I don't have a database of 40/4m games involving MadChess and Minic playing a pool of other engines (I test using a time control of 2m+1s). The relative ELO is what matters. Minic is 78 ELO stronger than MadChess 3.0 beta at this time control.
Vivien, to answer your question from an earlier post: Minic on my PC (AMD Ryzen Threadripper 1950X) runs at ~1.1 million NPS in the middlegame.
Code: Select all
# PLAYER : RATING ERROR CFS(%) PLAYED W D L (%) OppAvg
1 Minic 0.28 : 2407.0 19.7 100 1000 511 197 292 61 2329.0
2 MadChess 3.0 : 2329.0 ---- --- 1000 292 197 511 39 2407.0
Depends on the time control (until you improve Minic's time management with incremental time controls) and the version of MadChess. The 3.0 Beta version (a rewrite using bitboards still in early development) is weaker than the 2.2 release version (mailbox). So yes, Minic is weaker than MadChess 2.2 but likely stronger than 3.0 Beta.I guess Minic is weaker than Madchess.
Vivien, to answer your question from an earlier post: Minic on my PC (AMD Ryzen Threadripper 1950X) runs at ~1.1 million NPS in the middlegame.
My C# chess engine: https://www.madchess.net
-
- Posts: 1871
- Joined: Sat Nov 25, 2017 2:28 pm
- Location: France
Re: A complete 2000 lines of code engine
Thanks lot for those tests.
-
- Posts: 1871
- Joined: Sat Nov 25, 2017 2:28 pm
- Location: France
Re: A complete 2000 lines of code engine
Minic 0.29 is released. Nearly +100elo.
-
- Posts: 1142
- Joined: Thu Dec 28, 2017 4:06 pm
- Location: Argentina
Re: A complete 2000 lines of code engine
Wow, progress is so fast... "Etherealesque"
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls
-
- Posts: 1871
- Joined: Sat Nov 25, 2017 2:28 pm
- Location: France
Re: A complete 2000 lines of code engine
Thanks for the compliment, but Minic will probably never reach Ethereal strength ... and I am probably far less skilled than Andrew.
The improvement between 0.28 and 0.29 is mainly linked to a bug in TT (the previous stored score was total crap), and a better use of SEE I guess.
Evaluation is still very poor, but I'm working on it, maybe the next release will have a more complex eval.
Minic is now near 3000sloc, so my plan to stick near 2000sloc seems to vanish. Some part of the code needs quite some lines (SMP, material table, evaluation, ...). I'll try to compact things a little to get back to 2500sloc maybe ...
-
- Posts: 1871
- Joined: Sat Nov 25, 2017 2:28 pm
- Location: France
Re: A complete 2000 lines of code engine
Minic has a lichess bot now : https://lichess.org/@/Minic-chess_engine
It won't be playing 24/7.
It won't be playing 24/7.
-
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: A complete 2000 lines of code engine
Keep up the good work!xr_a_y wrote: ↑Sun Jan 06, 2019 4:21 pm Minic has a lichess bot now : https://lichess.org/@/Minic-chess_engine
It won't be playing 24/7.
-
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: A complete 2000 lines of code engine
I played a two quick knights odds games vs McCain 5 sec game with 5 sec increment. single core/thread 256 hash. They split the set.MikeB wrote: ↑Mon Jan 07, 2019 3:51 pmKeep up the good work!xr_a_y wrote: ↑Sun Jan 06, 2019 4:21 pm Minic has a lichess bot now : https://lichess.org/@/Minic-chess_engine
It won't be playing 24/7.
[pgn] [Event "Mac Pro x5690 3.46 Ghz"] [Site "Mac-Pro.local"] [Date "2019.01.07"] [Round "-"] [White "Minic 0.29"] [Black "McCain X1"] [Result "1-0"] [TimeControl "1+1"] [FEN "r1bqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"] [SetUp "1"] 1. e4 {+3.54/14 +1} e6 {-3.41/19 +1} 2. Nf3 {+3.47/13 +1} d5 {-3.40/17 +1} 3. e5 {+3.56/12 +1} c5 {-3.44/21 +1} 4. Nc3 {+3.45/12 +1} Ne7 {-3.24/18 +1} 5. Bb5+ {+3.29/11 +2} Bd7 {-3.11/17 +2} 6. a4 {+3.28/13 +2} a6 {-3.08/19 +2} 7. Bxd7+ {+3.13/13 +2} Qxd7 {-3.01/20 +2} 8. O-O {+3.06/12 +2} Qc7 {-2.97/21 +2} 9. d4 {+3.17/13 +2} cxd4 {-3.09/20 +2} 10. Qxd4 {+3.31/12 +2} Nc6 {-2.98/21 +1} 11. Qd1 {+3.04/10 +3} Be7 {-2.81/19 +1} 12. Bf4 {+3.11/12 +3} h6 {-2.77/17 +1} 13. Bg3 {+3.13/11 +3} O-O-O {-2.92/19 +1} 14. Ne2 {+3.27/12 +3} g5 {-2.74/16 +1} 15. h3 {+3.41/13 +3} h5 {-2.88/20 +1} 16. c3 {+3.26/12 +3} Rdg8 {-2.33/19 +1} 17. Ned4 {+3.58/12 +3} g4 {-2.46/19 +1} 18. hxg4 {+3.77/11 +4} hxg4 {-2.58/19 +1} 19. Nxc6 {+4.02/12 +4} Qxc6 {-2.81/19 +1} 20. Nh2 {+4.24/13 +4} Rh5 {-2.54/20 +1} 21. Nxg4 {+4.41/13 +4} Rgh8 {-2.44/16 +1} 22. Bh2 {+4.41/12 +4} Kb8 {-2.30/18 +1} 23. Re1 {+4.78/11 +4} Qc8 {-2.70/20 +1} 24. Re3 {+4.58/10 +4} Qg8 {-2.56/21 +1} 25. Rh3 {+3.46/11 +4} Rxh3 {-3.07/23 +1} 26. gxh3 {+3.52/14 +4} Rxh3 {-3.11/21 +2} 27. Bg3 {+3.65/14 +4} Rh8 {-3.24/24 +1} 28. Nf6 {+3.59/12 +4} Qg6 {-3.29/22 +1} 29. Qf3 {+3.64/13 +5} Ka7 {-3.13/18 +1} 30. Rc1 {+3.49/11 +5} Bxf6 {-3.12/21 +1} 31. exf6 {+3.75/15 +5} Qg5 {-3.17/20 +1} 32. Qf4 {+3.76/15 +5} Qxf4 {-3.43/19 +1} 33. Bxf4 {+3.82/15 +5} b6 {-3.60/21 +1} 34. Be5 {+3.84/14 +5} a5 {-3.75/21 +1} 35. Kg2 {+3.87/16 +5} Ka6 {-3.89/21 +1} 36. b3 {+3.86/13 +5} b5 {-4.15/22 +1} 37. Kg3 {+3.80/14 +5} Kb6 {-4.29/23 +1} 38. Ra1 {+3.83/14 +5} Rh5 {-4.63/22 +1} 39. Bd4+ {+4.07/15 +5} Ka6 {-5.13/22 +1} 40. axb5+ {+4.27/15 +5} Kxb5 {-5.12/24 +1} 41. Kf4 {+4.27/15 +5} Rh8 {-5.33/23 +1} 42. Ke5 {+4.30/16 +5} Kc6 {-5.52/21 +1} 43. Rxa5 {+5.92/14 +5} Rh5+ {-5.70/21 +1} 44. Kf4 {+6.42/13 +5} e5+ {-5.94/21 +1} 45. Bxe5 {+7.02/12 +5} d4 {-6.76/21 +1} 46. cxd4 {+7.89/12 +5} Rh4+ {-7.51/21 +1} 47. Kg5 {+8.83/11 +5} Rh1 {-8.68/22 +1} 48. Ra7 {+9.41/13 +5} Kb6 {-38.06/17 +1} 49. Rxf7 {+10.18/12 +5} Rg1+ {-48.08/18 +1} 50. Bg3 {+10.45/10 +6} Rd1 {-54.48/20 +1} 51. Re7 {+15.19/12 +3} {User adjudication} 1-0 [Event "Mac Pro x5690 3.46 Ghz"] [Site "Mac-Pro.local"] [Date "2019.01.07"] [Round "-"] [White "McCain X1"] [Black "Minic 0.29"] [Result "1-0"] [TimeControl "5+5"] [FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/R1BQKBNR w KQkq - 0 1"] [SetUp "1"] 1. d4 {-2.90/23 +8} Nf6 {+3.07/16 +6} 2. Bg5 {-2.99/25 +5} g6 {+3.30/16 +7} 3. e3 {-2.82/22 +7} Bg7 {+3.24/16 +8} 4. Bd3 {-2.99/23 +5} O-O {+3.26/17 +9} 5. Nf3 {-2.93/21 +5} h6 {+3.28/16 +9} 6. Bh4 {-2.93/22 +5} d5 {+3.29/16 +10} 7. O-O {-2.92/19 +8} Be6 {+3.30/15 +11} 8. Nd2 {-2.81/22 +9} Nc6 {+3.48/12 +12} 9. c3 {-2.57/18 +12} Na5 {+3.29/13 +13} 10. f4 {-2.82/22 +5} Nc4 {+3.39/13 +13} 11. Bxc4 {-2.49/19 +8} dxc4 {+3.39/14 +14} 12. e4 {-2.20/20 +9} b5 {+3.44/15 +15} 13. h3 {-2.23/20 +11} Rc8 {+3.40/15 +15} 14. Qf3 {-2.13/21 +5} c6 {+3.45/15 +16} 15. g4 {-2.13/20 +8} Nh7 {+3.38/14 +17} 16. f5 {-2.24/21 +5} Bd7 {+3.24/14 +17} 17. Qe3 {-2.31/19 +6} Nf6 {+3.15/13 +18} 18. e5 {-2.05/19 +9} Nd5 {+3.03/14 +18} 19. Qf3 {-1.88/21 +12} a6 {+2.92/13 +19} 20. Ne4 {-1.83/19 +15} a5 {+2.76/13 +19} 21. Rae1 {-1.87/23 +6} Bh8 {+2.80/15 +20} 22. Nc5 {-1.56/19 +9} g5 {+2.65/16 +20} 23. Bg3 {-1.21/19 +12} Bg7 {+2.59/17 +21} 24. h4 {-1.11/22 +13} a4 {+2.53/15 +21} 25. hxg5 {+0.00/21 +16} hxg5 {+2.60/15 +21} 26. a3 {+0.65/22 +18} Be8 {+2.56/15 +22} 27. Qh1 {+2.62/21 +18} Qb6 {+2.25/17 +22} 28. Qh3 {+4.92/21 +20} Ra8 {+2.28/14 +22} 29. Rf2 {+11.76/23 +23} Rd8 {+2.60/11 +23} 30. Rh2 {+1000.04/58 +26} Nf6 {-99.93/58 +27} 31. exf6 {+1000.03/99 +28} Bh6 {-99.95/58 +32} 32. Qxh6 {+1000.02/125 +29} exf6 {-99.97/58 +37} 33. Qh8# {+1000.01/127 +34} {White mates} 1-0[/pgn]
Here's macOS Minic: