Round Robin Tournaments

[Ghppn]

Games pgn: https://mega.nz/file/P9Ak2IiS#kABJFuDbW ... OdByzdHro4
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz

4 Cores (8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit,
Graphic board: NVIDIA Geforce GTX 760
Windows 10

Openings: Balsa_v500
All engines had their default settings, except Obsidian which I set to 1 core, as it started every game with 2 cores.

[deeds]

...
And unfortunately there are such guys, who have no clue about what's going..
Actually they may have, but blaming the strong openings, just because they
Are better and stronger..this will be another madness...then may I say simply:
Stockfish or any other super strong engine is not so good.., LOL

And why I say like that? for several reasons, due some fool/s are tried to say
Such as Balsa is not good for determining strength..hehe) anything more else?
And as most strange, they forget to mention that they prefer so weak openings..
...

[Ghppn]

I am very surprised by the expression of Sedat Cambaz, whom I appreciate very much. But as long as he is talking in general, I do not consider that he offended me. But the one who personalizes my text of someone else offends me. I do not even realize what the discussion was about.
I'm just saying: It's not nice to act like that.
If I'm wrong, you can politely point it out to me, and if you're generous, all the better to help me learn.

[deeds]

I doubt that your appreciation of such person changes anything in his attitude.
You are not the first one he has disappointed, you will not be the last !

You seem not to have understood yet that he will have no consideration for any matches or tournaments or experiments or trainings that were not organized by himself.

To get back to the real subject, adding a much weaker participant (Revenge) to hide the fact that the tournament conditions do not allow to differentiate between engines of equivalent strengths, it creates a distortion of the results and it's shocking from a TD who wants to be respectable but it is not the first time nor the last eh !

Long live to tournaments with >90% draw rate

[deeds]

What the TD assumes to show :

What the TD doesn't assume to show :

ERROR MARGIN +/- 87 ELO

Code: Select all

   # PLAYER                               :  RATING  ERROR  POINTS  PLAYED   (%)    W    D    L   D(%)  OppAvg  OppN
   1 Troubadour-1.5-bmi2                  :      25     93    14.0      26  53.8    2   24    0   92.3      -2    13
   2 SpecTral v2_bmi2                     :      13     87    13.5      26  51.9    1   25    0   96.2      -1    13
   3 JudaS ++ 1.01 bmi2                   :      13     89    13.5      26  51.9    1   25    0   96.2      -1    13
   4 CorChess 4.5, dev-20250119-8b9f04    :      13     92    13.5      26  51.9    1   25    0   96.2      -1    13
   5 JudaS ++ 5.0 bmi2                    :      13     93    13.5      26  51.9    1   25    0   96.2      -1    13
   6 Beast M-010125-bmi2                  :       0     92    13.0      26  50.0    0   26    0  100.0      -0    13
   7 TROY 022025_bmi2                     :       0     93    13.0      26  50.0    0   26    0  100.0      -0    13
   8 Obsidian dev-14.01 SE                :       0     93    13.0      26  50.0    0   26    0  100.0      -0    13
   9 Brainlearn 30                        :       0     92    13.0      26  50.0    0   26    0  100.0      -0    13
  10 ShashChess 38                        :       0     95    13.0      26  50.0    0   26    0  100.0      -0    13
  11 Dark SisTer v2_bmi2                  :       0     91    13.0      26  50.0    0   26    0  100.0      -0    13
  12 DarkSeid v5.3_bmi2                   :       0     89    13.0      26  50.0    0   26    0  100.0      -0    13
  13 HypnoS ++ 1.01 bmi2                  :     -13     96    12.5      26  48.1    0   25    1   96.2       1    13
  14 Revenge 4.0 bmi2                     :     -63     93    10.5      26  40.4    0   21    5   80.8       5    13

White advantage = 11.64 +/- 18.88
Draw rate (equal opponents) = 50.00 % +/- 0.00

ENGINES ONLY PLAYED 66 MOVES BY GAME

Code: Select all

014 players                       : avg. D28, 003 623 ms/move, 000 182 games, 00 012 051 moves

beast m-010125-bmi2               : avg. D29, 003 570 ms/move, 000 026 games, 00 000 861 moves
brainlearn 30                     : avg. D30, 003 764 ms/move, 000 026 games, 00 000 768 moves
corchess 4.5, dev-20250119-8b9f04 : avg. D26, 003 474 ms/move, 000 026 games, 00 000 945 moves
dark sister v2_bmi2               : avg. D27, 003 495 ms/move, 000 026 games, 00 001 001 moves
darkseid v5.3_bmi2                : avg. D28, 003 883 ms/move, 000 026 games, 00 000 718 moves
hypnos ++ 1.01 bmi2               : avg. D29, 003 931 ms/move, 000 026 games, 00 000 857 moves
judas ++ 1.01 bmi2                : avg. D28, 003 984 ms/move, 000 026 games, 00 000 836 moves
judas ++ 5.0 bmi2                 : avg. D25, 003 349 ms/move, 000 026 games, 00 000 754 moves
obsidian dev-14.01 se             : avg. D28, 003 756 ms/move, 000 026 games, 00 000 870 moves
revenge 4.0 bmi2                  : avg. D28, 003 053 ms/move, 000 026 games, 00 001 168 moves
shashchess 38                     : avg. D27, 004 043 ms/move, 000 026 games, 00 000 669 moves
spectral v2_bmi2                  : avg. D27, 003 522 ms/move, 000 026 games, 00 000 846 moves
troubadour-1.5-bmi2               : avg. D30, 003 603 ms/move, 000 026 games, 00 000 901 moves
troy 022025_bmi2                  : avg. D28, 003 657 ms/move, 000 026 games, 00 000 857 moves

Averages :
0 240 sec/game (4 min/game)
076 plies/game (066 played plies/game)

Game stages :
- opening @ avg. D26, 005 066 ms/move
- middleg @ avg. D27, 003 350 ms/move
- endgame @ avg. D30, 002 448 ms/move

These tournament conditions do not allow to differentiate between engines of equivalent strength so much that even with a 5 plies difference in average depth, JudaS (avg. D25) holds its own against Troubadour/BrainLearn (avg. D30)

[Ghppn]

https://i.postimg.cc/GpF524yJ/Turneun-Round-Robin-2.jpg

Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz
4 Cores (8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit,
Graphic board: NVIDIA Geforce GTX 760
Windows 10

Openings: Balsa_v500
All engines had their default settings except Obsidian, which I set to 1 core, as I started every game with 2 cores.
I increased the control time. The conclusion is that if I exclude Revenge 4, all games are draws. I also tried Revenge because I know it had an honorable result in one of the TCEC tournaments.

[deeds]

The conclusion is that if I exclude Revenge 4, all games are draws.

We agree so :
- either the tested engines had an equivalent strength you try to confuse people with such distorted results and hiding the huge error margin
- either your tournament conditions are drawish and It's doubtful/suspicious for a TD since years

Reminder :
To differentiate engines of equivalent strength, some make hundreds of thousands of games

[Sedat Canbaz]

I am very surprised by the expression of Sedat Cambaz, whom I appreciate very much. But as long as he is talking in general, I do not consider that he offended me. But the one who personalizes my text of someone else offends me. I do not even realize what the discussion was about.
I'm just saying: It's not nice to act like that.
If I'm wrong, you can politely point it out to me, and if you're generous, all the better to help me learn.

Hello Ghppn

1st of all,
Thanks for using Balsa openings...
And why are you offended here?

2nd.
Strange/Interesting that my quotes/statements are on this thread?!
And it seems someone/s is much affecting that people use Balsa suites )
Plus what I said before regarding all over, I stand totally with my words!

3rd. as a friendly advise,
Do not allow anyone to manipulate you here! because
I said nothing wrong with your tourneyes, work etc.

My mentioned statements about all over (as source available):
viewtopic.php?f=4&t=3600&start=1860

On other hand,
Do not worry, my comments are completely NOT about you..ok ?

If you have any questions... please feel free to contact Me,
We can talk privately if you wish of course..but not with Deeds,
Because he is in my Blocked list ..as reason, he deserved to be...
E.g before blocking.. I could not read much useful things by him
Exc. Bullshit )

Anyhow,, enjoy and have fun with my opening test suites )

Greetings

[deeds]

I am very surprised by the expression of Sedat Cambaz

Maybe it can explain why you didn't share the games here or maybe the error margin was still huge

Long live to tournaments with >90% draw rate

[deeds]

If I'm wrong, you can politely point it out to me, and if you're generous, all the better to help me learn.

About the average number of moves/game played by engines (=content), compare with others :

source