1. How do I run it under Windows?
2. What strength shall I expect?
3. Have you abandoned Embla and Micah for good?
CaffeinatedPawn questions
Moderator: Ras
-
Gabor Szots
- Posts: 1494
- Joined: Sat Jul 21, 2018 7:43 am
- Location: Budapest, Hungary
- Full name: Gabor Szots
CaffeinatedPawn questions
Gabor Szots
CCRL testing group
CCRL testing group
-
flok
- Posts: 611
- Joined: Tue Jul 03, 2018 10:19 am
- Full name: Folkert van Heusden
Re: CaffeinatedPawn questions
1. you need java for that. then you need to invoke something like:Gabor Szots wrote: ↑Wed Oct 27, 2021 12:53 pm 1. How do I run it under Windows?
2. What strength shall I expect?
3. Have you abandoned Embla and Micah for good?
java -cp CaffeinatedPawn-<code>.jar;chesslib-1.3.2.jar com.vanheusden.CaffeinatedPawn.CaffeinatedPawn
Can't verify as I don't have windows.
2. according to the cutechess results at home (which were different from the one forum3/viewtopic.php?f=2&p=909955 thread) 1775.3 elo
3. nah. while working on caffeinatedpawn I found some possible issues in embla/micah and I might fix & test those. but don't hold your breath.
-
Brunetti
- Posts: 424
- Joined: Tue Dec 08, 2009 1:37 pm
- Location: Milan, Italy
- Full name: Alex Brunetti
Re: CaffeinatedPawn questions
Previous version was around 1750, I'm testing right now the new one.Gabor Szots wrote: ↑Wed Oct 27, 2021 12:53 pm 1. How do I run it under Windows?
2. What strength shall I expect?
If you need them, I made Windows executables for all versions.
Alex
-
flok
- Posts: 611
- Joined: Tue Jul 03, 2018 10:19 am
- Full name: Folkert van Heusden
Re: CaffeinatedPawn questions
Thanks for testing.Brunetti wrote: ↑Wed Oct 27, 2021 1:44 pmPrevious version was around 1750, I'm testing right now the new one.Gabor Szots wrote: ↑Wed Oct 27, 2021 12:53 pm 1. How do I run it under Windows?
2. What strength shall I expect?
If you need them, I made Windows executables for all versions.
Alex
I'm puzzled why it plays worse in a real testing setup and not in mine.
How can I make sure I've got the same testing conditions?
-
Brunetti
- Posts: 424
- Joined: Tue Dec 08, 2009 1:37 pm
- Location: Milan, Italy
- Full name: Alex Brunetti
Re: CaffeinatedPawn questions
To have the same results we should play the same opponents pool at the same conditions. And the CPU? And what about the opening books? Is yours more drawish or more unbalanced than mine? But most important, the two rating scales how are anchored each other?
I call cutechess with the following parameters:
Code: Select all
@cutechess-cli -engine conf="engine1" -engine conf="engine2" -each tc=60+1 option.Threads=1 option."Max CPUs"=1 option."Max Threads"=1 option."Max threads"=1 option.Hash=256 -pgnout "file" -site "Intel i7-3770@3.40GHz" -event "event" -openings file="file" order="random" -recover -srand 6730291 -draw movenumber=30 movecount=12 score=15 -resign movecount=12 score=600 twosided=true
The opponents pool is another question: I don't play gauntlets with a few opponents and an high number of games. My approach is to play a Swiss-like tournament with a few games vs. many opponents. I found this system to be more efficient, I get quite accurate (whatever it means) and stable ratings with far less games than the standard big gauntlets way. These are the results of the last CP:
Code: Select all
-------------------------------------------------------------------------------------------------------------
Results of CaffeinatedPawn 0346f64 64-bit
-------------------------------------------------------------------------------------------------------------
Rank Elo Engine Score Games Wins Draws Losses
-------------------------------------------------------------------------------------------------------------
1. 2221 Hagrid 0.7.47 0% 1 0 0 1
2. 2217 Feuerstein 0.4.6.1 0% 2 0 0 2
3. 2211 FireFly 2.7.0 64-bit 0% 1 0 0 1
4. 2204 Kurt 0.9.2.2 beta 64-bit 0% 1 0 0 1
5. 2196 Blitzkrieg 1.0 rev337 64-bit 0% 1 0 0 1
6. 2191 DrunkenMaster 1.2 0% 1 0 0 1
7. 2185 BlackBishop 1.0u 0% 1 0 0 1
8. 2177 Embla 2.07 64-bit 0% 1 0 0 1
9. 2071 Bad Chess Engine 0.4.4-ucifix 64-bit 0% 1 0 0 1
10. 2032 Etude 0.11 0% 1 0 0 1
11. 2031 Celes 0.78 RWBC 100% 1 1 0 0
12. 2028 Goldfish 1.13.0 64-bit 0% 1 0 0 1
13. 2021 Asterisk 0.4b 0% 2 0 0 2
14. 2018 CPW-Engine build 05-09-08 0% 2 0 0 2
15. 2006 Rival 1.18 25% 2 0 1 1
16. 2001 Taktix 2.23x 0% 2 0 0 2
17. 2000 Skiull 0.5 64-bit 25% 2 0 1 1
18. 1995 ALChess 1.84 25% 2 0 1 1
19. 1991 Presbyter 1.3.0 64-bit 50% 2 1 0 1
20. 1990 Big Lion 2.23x 50% 2 1 0 1
21. 1988 Sissa 2.0.0.0 0% 2 0 0 2
22. 1982 Pleco 0.1.6 64-bit 33% 3 1 0 2
23. 1981 Jabba 13032012 33% 3 1 0 2
24. 1980 Hopper 1.3 64-bit 25% 2 0 1 1
25. 1979 SoFCheck alpha 20210918 64-bit 50% 3 1 1 1
26. 1970 Claudia 0.51 64-bit 0% 3 0 0 3
27. 1969 Dumber 1.1 64-bit 17% 3 0 1 2
28. 1966 Gunborg 1.0 64-bit 0% 3 0 0 3
29. 1963 Weiawaga 1.3 64-bit 33% 3 0 2 1
30. 1957 Hopper 1.2 64-bit 0% 3 0 0 3
31. 1940 Lila 3.0.1 64-bit 0% 3 0 0 3
32. 1927 FreeChess 2.17b 50% 4 2 0 2
33. 1927 Roce 0.039 25% 4 1 0 3
34. 1924 Odonata 0.3.26 64-bit 38% 4 1 1 2
35. 1915 Clueless 1.4 25% 4 1 0 3
36. 1910 Rustic Alpha 3.0.0 64-bit 25% 4 1 0 3
37. 1904 Deep Chess r113 64-bit 25% 4 1 0 3
38. 1902 Minace 1.0 13% 4 0 1 3
39. 1900 Snowy 0.1 64-bit 25% 4 1 0 3
40. 1896 GearHeart 0.16 TP 64-bit 38% 4 1 1 2
41. 1894 CDrill 1800 Build 4 38% 4 1 1 2
42. 1893 Deepov 0.4 64-bit 50% 4 2 0 2
43. 1890 Predateur2.2.1 0% 4 0 0 4
44. 1888 HeavyChess 0.13 beta 13% 4 0 1 3
45. 1884 Heracles 0.5.9 25% 4 0 2 2
46. 1882 Chareth 2020-06-28 64-bit 20% 5 0 2 3
47. 1881 Sayuri 2018-05-23 64-bit 13% 4 0 1 3
48. 1877 Ziggy 0.7 64-bit 25% 4 1 0 3
49. 1867 Hopper 20211004 64-bit 10% 5 0 1 4
50. 1862 Pigeon 1.5.1 64-bit 40% 5 1 2 2
51. 1858 Apollo 1.2.1 64-bit 20% 5 1 0 4
52. 1853 Zurichess Appenzeller 64-bit 50% 5 2 1 2
53. 1836 Silke Chess 1.2.1209 50% 5 2 1 2
54. 1823 BeRoChess 1.00.2012.04.10.0112 64-bit 17% 6 0 2 4
55. 1814 Hopper 20211003 64-bit 58% 6 3 1 2
56. 1794 Hedwig rev25 64-bit 25% 6 1 1 4
57. 1785 RumbleMinze 1.0.33 64-bit 0% 7 0 0 7
58. 1778 Rocinante 2.0 64-bit 43% 7 3 0 4
59. 1776 Celestial 1.0 64-bit 43% 7 3 0 4
60. 1776 Vapor 0.01r12 64-bit 43% 7 2 2 3
61. 1774 Testina 3c 29% 7 0 4 3
62. 1764 CaffeinatedPawn 95458ab 64-bit 57% 7 3 2 2
63. 1760 CaffeinatedPawn 6d214df 64-bit 50% 7 2 3 2
64. 1759 RapChessCs 2021-07-22 64-bit 43% 7 3 0 4
65. 1753 Shallow Blue 2.0.0 64-bit 57% 7 3 2 2
66. 1740 Small-C 90.3 31% 8 2 1 5
67. 1740 Pwned 1.3 64-bit 43% 7 3 0 4
68. 1712 Crabby 1.00 64-bit 88% 8 7 0 1
69. 1706 FracTal 1.0 64-bit 50% 7 2 3 2
70. 1700 CaSpEr rev4 64-bit 50% 7 3 1 3
71. 1699 Walleye 1.5.0 64-bit 43% 7 2 2 3
72. 1670 Walleye 1.4 64-bit 50% 7 1 5 1
73. 1668 Chancellor 1.00d 50% 7 3 1 3
74. 1656 ArcBishop 1.00d 75% 6 4 1 1
75. 1634 Cicada 0.1 64-bit 50% 6 2 2 2
76. 1629 Pigeon 1.37 64-bit 50% 6 2 2 2
77. 1624 Pulse 1.7.2 64-bit 58% 6 2 3 1
78. 1596 Koedem 1.1 64-bit 100% 5 5 0 0
79. 1591 Conqueror 1.2 70% 5 3 1 1
80. 1588 CaffeinatedPawn 54311af 64-bit 80% 5 4 0 1
81. 1584 Marginal 0.1.051 60% 5 3 0 2
82. 1581 SebLague Chess-AI 0.1 .Net5 70% 5 3 1 1
83. 1562 Blank 1.0 64-bit 75% 4 2 2 0
84. 1557 ChessEngine (SeminarEngine) 0.1-alpha 64-bit 75% 4 2 2 0
85. 1557 DoctorB 1.2.1 64-bit 75% 4 3 0 1
86. 1553 Piranha 0.5 100% 4 4 0 0
87. 1528 Jacky Chess 0.9.14 64-bit 88% 4 3 1 0
88. 1525 Irina 0.15 75% 4 3 0 1
89. 1504 Superpawn build 110 64-bit 100% 3 3 0 0
90. 1485 Walleye 1.3.0 64-bit 100% 3 3 0 0
91. 1472 Quokka 2.1 64-bit 100% 3 3 0 0
92. 1458 Madlenka 2015.02.13 DC 64-bit 75% 2 1 1 0
93. 1457 Lynx 5.2 64-bit 100% 2 2 0 0
94. 1451 Lynx 5.3 64-bit 50% 2 1 0 1
95. 1447 Lynx 6.0 64-bit 100% 2 2 0 0
96. 1444 Hactar 0.9.0 64-bit 100% 2 2 0 0
97. 1443 Eden 0.0.13 100% 2 2 0 0
98. 1438 Lynx 5.1 64-bit 100% 2 2 0 0
99. 1421 Endamat 20210217 64-bit 75% 2 1 1 0