exactly ... this solved the problem of the day.
A third test is necessary ... after Fritz 16.
Dog 4.10.2 can be play the third test.
I will add the new rank-system later this day.
Best
Frank
A test idea without Elo, I think I start middle of Jan.26!
Moderator: Ras
-
Frank Quisinsky
- Posts: 7185
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: A test idea without Elo, I think I start middle of Jan.26!
Hi there,
exactly ... this solved the problem of the day.
A third test is necessary ... after Fritz 16.
Dog 4.10.2 can be play the third test.
I will add the new rank-system later this day.
Best
Frank
exactly ... this solved the problem of the day.
A third test is necessary ... after Fritz 16.
Dog 4.10.2 can be play the third test.
I will add the new rank-system later this day.
Best
Frank
-
Frank Quisinsky
- Posts: 7185
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: A test idea without Elo, I think I start middle of Jan.26!
Code: Select all
[b]Standard Engines:[/b]
01. Uralochka 3.42a JA 3550
02. Revenge 4.0 3500
03. CSTal 2.0 3475
04. Velvet 8.1.1 3475
05. Igel 3.6.0 JA 3475
06. SlowChess Blitz 2.9 3400
07. Texel 1.12 3400
08. Stockfish 200731 HCE 3375
09. Wasp 7.00 3350
10. Patricia 5.0 JA 3350
11. Monty 251209 MCTS dev 3300
12. Leorik 3.1.3 JA 3275
13. Tcheran 9.0 3275
14. Nemorino 6.11 JA 3250
15. Booot 6.50 HCE 3200
16. Xiphos 0.6.1 HCE JA 3175
17. Laser 1.7 HCE JA 3175
18. Senpai 3.0.1 HCE 3125
19. DanaSah 9.1 JA 3075
20. Fizbo 2.0 JA 3075
21. Petrel 3.1 JA 3050
22. Vajolet2 2.8.0 HCE 2975
23. Critter 1.6a HCE 2975
24. Deep iCE 4.0.853 HCE 2950
25. Hakkapelitta TCEC v2 HCE 2875
26. Spark 1.0 HCE 2775
-----
83875 : 26 = 3225,97 Elo
Code: Select all
Rank = 20, 3-4-3 system
1.300,0 (100,00%) - 1.144,0 ( 88,00%) points = 01. ***** General Field Marshal
1.143,5 ( 87,97%) - 1.105,0 ( 85,00%) points = 02. **** General
1.104,5 ( 84,97%) - 1.066,0 ( 82,00%) points = 03. *** Lieutenant General
1.065,5 ( 81,97%) - 1.027,0 ( 79,00%) points = 04. ** Major General
1.026,5 ( 78,97%) - 988,0 ( 76,00%) points = 05. * Brigadier General
--
987,5 ( 75,97%) - 936,0 ( 72,00%) points = 06. Colonel
935,5 ( 71,97%) - 884,0 ( 68,00%) points = 07. Lieutenant Colonel
883,5 ( 67,97%) - 832,0 ( 64,00%) points = 08. Major
831,5 ( 63,97%) - 780,0 ( 60,00%) points = 09. Captain
779,5 ( 59,97%) - 728,0 ( 56,00%) points = 10. First Lieutenant
727,5 ( 55,97%) - 676,0 ( 52,00%) points = 11. Second Lieutenant
--
675,5 ( 51,97%) - 637,0 ( 49,00%) points = 12. Sergeant Major
636,5 ( 48,97%) - 598,0 ( 46,00%) points = 13. Master Sergeant
597,5 ( 45,97%) - 559,0 ( 43,00%) points = 14. Sergeant First Class
558,5 ( 42,97%) - 520,0 ( 40,00%) points = 15. Staff Sergeant
519,5 ( 39,97%) - 481,0 ( 37,00%) points = 16. Sergeant
480,5 ( 36,97%) - 442,0 ( 34,00%) points = 17. Corporal
441,5 ( 33,97%) - 403,0 ( 31,00%) points = 18. Specialist
402,5 ( 30,97%) - 364,0 ( 28,00%) points = 19. Private First Class
363,5 ( 27,97%) - 0,0 ( 25,00%) points = 20. Private
New is the "Staff Sergeant" and "Sergeant First Class".
Problem solved!
Well, they're not new.
They've always been around, but they're new to my rank system.
For 19. Private First Class ... Fritz 16 (Rybka) should make 28,00% - 30,97%.
Live mode:
shredder *.sto file (tournament configuration, game plan)
https://www.amateurschach.de/fling/etoc-g_test-02.sto
shredder *.html file (current results)
https://www.amateurschach.de/fling/etoc-g_test-02.html
shredder *.pgn file (the games)
https://www.amateurschach.de/fling/etoc-g_test-02.pgn
Updates each 2 minutes with FTP-Software Fling Plus 5.04.
Best
Frank
-
Frank Quisinsky
- Posts: 7185
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: A test idea without Elo, I think I start middle of Jan.26!
Hi there,
Now, what should I display later the detail site?
In the group of "Standard Engines" there are many of the "King attacker", engines can win many games very shortly.
The following situation:
Engine A plays 1.300 games vs. the 26 Standard Engines, result = 745,5 points, rank: First Lieutenant
Engine B plays 1,300 games vs. the 26 Standard Engines, result = 745,5 points, rank: First Lieutenant
Is it possible if I test around 250 or more engines next year.
I have interest in displaying on a detail site only 4 stats.
- draw quote
- move average
- short won games below 50 moves
- short lost games below 50 moves
After all the years computerchess ...
Nothing is more boring as Elo! ... for such strong engines, today are available!
Engines give me a higher draw quote = several possible reasons!
Engines give me a higher move-average, are fighting more in endgames or fighting for the 0,5-1,0% possibility to make a half point more, waiting for mistakes by the opponent.
Engines give me a higher quantity of won games, are very aggressive in the middle game.
Engines give me a higher quantity of fast-loss games, and are usually stronger in the endgames (in most cases)
Engines give me a higher quantity of won and lost games, are playing more or less a bit speculative
Draw Quote = difficult to evaluate!
Very fast draws are all the time a topic but not for my idea.
More games can be played with a lower move-average, less boring end games will be the result.
On the other hand, a lower move-average because an engine gives to fast a draw isn’t very nice.
Back to Engine A and Engine B, both with 745,5 points.
Engine A: 12 won games below 50 moves, 4 lost games below 50 moves = 12-4 = +8
Engine B: 3 won games below 50 moves, 1 lost game below 50 moves = 3-1 = +2
A battle can be won through surprise attacks.
Very difficult at all, because also the endgame-artist plays nice chess.
I have no solution for an second system for evaluate all this; each game phase is interesting.
What I can do is to display:
1. Engine A = 745,5 P., First Lieutenant, 52,5% draws, 84,4% m-avg, 12-w / 04-l = +008
1. Engine B = 745,5 P., First Lieutenant, 56,8% draws, 89,4% m-avg, 03-w / 01-l = +002
Such information should be enough.
As little information as possible on an overview page with over 250 engines.
I will be thinking about it in the next day’s how I display all that.
Furthermore, url to the test-run result, url to the PGN database, url to the tournament configuration for each test.
Best
Frank
Now, what should I display later the detail site?
In the group of "Standard Engines" there are many of the "King attacker", engines can win many games very shortly.
The following situation:
Engine A plays 1.300 games vs. the 26 Standard Engines, result = 745,5 points, rank: First Lieutenant
Engine B plays 1,300 games vs. the 26 Standard Engines, result = 745,5 points, rank: First Lieutenant
Is it possible if I test around 250 or more engines next year.
I have interest in displaying on a detail site only 4 stats.
- draw quote
- move average
- short won games below 50 moves
- short lost games below 50 moves
After all the years computerchess ...
Nothing is more boring as Elo! ... for such strong engines, today are available!
Engines give me a higher draw quote = several possible reasons!
Engines give me a higher move-average, are fighting more in endgames or fighting for the 0,5-1,0% possibility to make a half point more, waiting for mistakes by the opponent.
Engines give me a higher quantity of won games, are very aggressive in the middle game.
Engines give me a higher quantity of fast-loss games, and are usually stronger in the endgames (in most cases)
Engines give me a higher quantity of won and lost games, are playing more or less a bit speculative
Draw Quote = difficult to evaluate!
Very fast draws are all the time a topic but not for my idea.
More games can be played with a lower move-average, less boring end games will be the result.
On the other hand, a lower move-average because an engine gives to fast a draw isn’t very nice.
Back to Engine A and Engine B, both with 745,5 points.
Engine A: 12 won games below 50 moves, 4 lost games below 50 moves = 12-4 = +8
Engine B: 3 won games below 50 moves, 1 lost game below 50 moves = 3-1 = +2
A battle can be won through surprise attacks.
Very difficult at all, because also the endgame-artist plays nice chess.
I have no solution for an second system for evaluate all this; each game phase is interesting.
What I can do is to display:
1. Engine A = 745,5 P., First Lieutenant, 52,5% draws, 84,4% m-avg, 12-w / 04-l = +008
1. Engine B = 745,5 P., First Lieutenant, 56,8% draws, 89,4% m-avg, 03-w / 01-l = +002
Such information should be enough.
As little information as possible on an overview page with over 250 engines.
I will be thinking about it in the next day’s how I display all that.
Furthermore, url to the test-run result, url to the PGN database, url to the tournament configuration for each test.
Best
Frank
-
Frank Quisinsky
- Posts: 7185
- Joined: Wed Nov 18, 2009 7:16 pm
- Location: Gutweiler, Germany
- Full name: Frank Quisinsky
Re: A test idea without Elo, I think I start middle of Jan.26!
Again, what I saw from Laser is just great. Pure random that I have it in the group of "Standard Engines". Of course, I know the program from all my tourneys, the compiles from Jim give Laser more power.
Monty games = Fun, fun, fun ... what for a fine development.
Still no idea how strong Monty is, but clearly much stronger than I had assumed. We'll find out exactly how strong later, once we've done a lot of test runs.
So far, I am very happy with the group of 26 Standard engines.
This group of 26 engines, can test all 250 TOP-Engines very well with the rank-system.
No changes are necessary here. It must be hard for every engine I test to play vs. the 26-Standard engines.
Also the work from the Russian programmer, the engine Petrel ...
Very aggressive in the earlier mid-games. Often I am thinking the program will win the game after 10 moves.
Monty games = Fun, fun, fun ... what for a fine development.
Still no idea how strong Monty is, but clearly much stronger than I had assumed. We'll find out exactly how strong later, once we've done a lot of test runs.
So far, I am very happy with the group of 26 Standard engines.
This group of 26 engines, can test all 250 TOP-Engines very well with the rank-system.
No changes are necessary here. It must be hard for every engine I test to play vs. the 26-Standard engines.
Also the work from the Russian programmer, the engine Petrel ...
Very aggressive in the earlier mid-games. Often I am thinking the program will win the game after 10 moves.