RWBC test tourney LIVE (Class D + new versions + S6)

[Günther Simon]

Geschrieben von:/Posted by: Günther Simon at 22 May 2004 15:39:15:

IP: 80.145.236.44
port: 17077
Well, since Polyglot is able to handle the hanging problems
with Joris WBTM, I decided to add my 'old' Shredder6 UCI
also as wildcard to Class A. Therefore S6 is a somehow
colurful spot in this mixed field for testing reasons
(BTW sorry for those who feel their programs plugged into
a much too strong field)
As usual the test tourney is at quite fast pace,
the tc is 4min/40moves on my PIV 2.67Ghz
Other conditions as usual for RWBC.
Shredder6PB plays with my big exp. ThinkerBook + Polyglot(UCI2WB).
DanChess 1.04c plays with my main ThinkerBook
like in the yesterday finished Open Class.
What have I found so far?
-Ayito always prefers to lose on time in lost positions
-Tytan lost one game on time before move 160!
(still lots of material on board!)
-I saw a very big blunder by Alarm in a rook game
-Shredder6PB was hanging in one game even with AutoQuit=True
in the Polyglot.ini?
-Cilian allowed a rep draw in a game, where F8 says +5 already!
(This is after a very fast look over the first 52 games!
I will look for more after the tourney is over)
most debugs/logs available on request
Not much games and fast time control, but anyway I have to
ask, what happened to Aice? Milix what have you done?
After around 7 from 15 rounds:
 RANK   ENGINE                GAMES  POINTS   1  2  3  4  5  6  7  8  9  10 11 12 13 14 15 16 
-----------------------------------------------------------------------------------------------
   1.   SHREDDER_6PADERBORN      7     6.5    *  .  .  .  1  .  1  .  1  .  1  .  =  1  .  1  
   2.   AICE_066                 8     6.0    .  *  .  .  0  1  1  .  1  .  1  .  0  1  .  1  
   3.   ALARM_0931               7     5.5    .  .  *  .  .  =  .  1  1  .  .  0  1  1  1  .  
   4.   CHISPA_4032              7     5.0    .  .  .  *  0  .  .  .  1  .  0  1  1  1  1  .  
   5.   BRUJA_16                 7     5.0    0  1  .  1  *  .  0  1  .  1  .  .  .  .  .  1  
   6.   DANCHESS_104C            7     5.0    .  0  =  .  .  *  =  .  .  .  1  1  .  .  1  1  
   7.   GK_090                   8     4.5    0  0  .  .  1  =  *  .  .  =  =  .  .  1  .  1  
   8.   CHIRON_038               7     4.0    .  .  0  .  0  .  .  *  0  .  .  1  1  1  1  .  
   9.   EEYORE_138               8     4.0    0  0  0  0  .  .  .  1  *  1  .  1  .  .  .  1  
  10.   CILIAN_413DC             7     3.0    .  .  .  .  0  .  =  .  0  *  =  .  =  =  1  .  
  11.   TYTAN_361                7     3.0    0  0  .  1  .  0  =  .  .  =  *  .  .  .  .  1  
  12.   MERLIN_12                7     3.0    .  .  1  0  .  0  .  0  0  .  .  *  1  .  1  .  
  13.   MARVIN_100               8     2.0    =  1  0  0  .  .  .  0  .  =  .  0  *  .  0  .  
  14.   AYITO_0244               8     1.0    0  0  0  0  .  .  0  0  .  =  .  .  .  *  .  =  
  15.   BODO_02B                 7     1.0    .  .  0  0  .  0  .  0  .  0  .  0  1  .  *  .  
  16.   FIREFLY_101              8     0.5    0  0  .  .  0  0  0  .  0  .  0  .  .  =  .  *  
Total games = 59

[Anastasios Milikas]

Geschrieben von:/Posted by: Anastasios Milikas at 22. May 2004 20:06:
Als Antwort auf:/In reply to: RWBC test tourney LIVE (Class D + new versions + S6) geschrieben von:/posted by: Günther Simon at 22 May 2004 15:39:15:

Not much games and fast time control, but anyway I have to
ask, what happened to Aice? Milix what have you done?

I simplified evaluation to the minimum! The passed pawn evaluation is very
good now (still overestimates in some positions, will be fixed soon). And I
fix some search bugs. Note that AICE 0.66 still has bugs in search. I found
that it has a repetition bug in under certain circumstances I hope I'll sorve it in 0.67.
Thanks for the tournaments, Günther
They help me alot in testing and improving AICE.



http://www.milix.net/aice

[Günther Simon]

Geschrieben von:/Posted by: Günther Simon at 23 May 2004 11:46:34:
Als Antwort auf:/In reply to: RWBC test tourney LIVE (Class D + new versions + S6) geschrieben von:/posted by: Günther Simon at 22 May 2004 15:39:15:

Problems found:
-Ayito lost 8 games on time in lost positions
-Tytan lost 2 games on time(logs sent)
-I saw a very big blunder by Alarm in a rook game(log sent)
-Firefly often reached quite good positions but somehow
missed tactical shots in the middlegame
-Shredder6PB was hanging in one game even with AutoQuit=True
in the Polyglot.ini?
-Cilian allowed a rep draw in a game, where F8 says +5 already!
-Cilian did two wrong draw claims!
Well after looking through the games with LGPGNVER I suddenly
remembered that I had forgotten to think about a solution
for Cilian! Cilian often claims draws too early even, when
it is not its move and also when there is no draw at all
on board! (had to forfeit 4 games in last edition - even when
I considered repetition draw claims as ok when claiming,
if it was its turn to move and could force the draw
with its next move).
I will decide tomorrow what to do now, but the most probably
solution will be that Cilian has to be excluded from RWBC as
long as the problem is solved. I know that Dann has the source
somewhere and perhaps a programmer (like Thomas in Gnu4s' case)
or Dann himself is willing tobugfix its repetition code?
In that case Uragano will promote as best placed from the last
Open Class and also Zotron would remain there...
Günther
Final table:
DC4Test2
ESPRESSO, 2004.05.22 - 2004.05.23
                           Score     S A B A E D C C T G B M M C A F
---------------------------------------------------------------------
 1: Shredder_6Paderborn  14.5 / 15   X 1 1 1 1 1 1 1 1 1 1 = 1 1 1 1
 2: Alarm_0931           11.0 / 15   0 X 1 = 1 = 1 1 0 1 1 1 0 1 1 1
 3: Bruja_16             10.0 / 15   0 0 X 1 0 1 1 1 1 0 = 1 = 1 1 1
 4: Aice_066              9.5 / 15   0 = 0 X 1 1 = = 1 1 0 0 1 1 1 1
 5: Eeyore_138            9.5 / 15   0 0 1 0 X 1 1 0 = 1 1 0 1 1 1 1
 6: DanChess_104c         9.0 / 15   0 = 0 0 0 X = = 1 = 1 1 1 1 1 1
 7: Chiron_038            9.0 / 15   0 0 0 = 0 = X 0 1 1 1 1 1 1 1 1
 8: Chispa_4032           8.5 / 15   0 0 0 = 1 = 1 X 0 0 1 1 1 1 1 =
 9: Tytan_361             7.0 / 15   0 1 0 0 = 0 0 1 X = 0 0 1 1 1 1
10: GK_090                7.0 / 15   0 0 1 0 0 = 0 1 = X 0 = 1 = 1 1
11: Bodo_02b              6.5 / 15   0 0 = 1 0 0 0 0 1 1 X 1 0 0 1 1
12: Marvin_100            5.5 / 15   = 0 0 1 1 0 0 0 1 = 0 X 0 = 0 1
13: Merlin_12             5.5 / 15   0 1 = 0 0 0 0 0 0 0 1 1 X = = 1
14: Cilian_413DC          4.0 / 15   0 0 0 0 0 0 0 0 0 = 1 = = X = 1
15: Ayito_0244            2.5 / 15   0 0 0 0 0 0 0 0 0 0 0 1 = = X =
16: Firefly_101           1.0 / 15   0 0 0 0 0 0 0 = 0 0 0 0 0 0 = X
---------------------------------------------------------------------
120 games: +55 =20 -45

[Günther Simon]

Geschrieben von:/Posted by: Günther Simon at 23 May 2004 11:49:44:
Als Antwort auf:/In reply to: Final result (note about Ayito,Tytan,Alarm and Cilian) geschrieben von:/posted by: Günther Simon at 23 May 2004 11:46:34:

http://home.t-online.de/home/g.simon.rgbg/dc4t2.zip

[Fabien Letouzey]

Geschrieben von:/Posted by: Fabien Letouzey at 24 May 2004 11:15:47:
Als Antwort auf:/In reply to: Final result (note about Ayito,Tytan,Alarm and Cilian) geschrieben von:/posted by: G�nther Simon at 23 May 2004 11:46:34:

Problems found:
...
-Shredder6PB was hanging in one game even with AutoQuit=True
in the Polyglot.ini?
...

I have seen it once before (in log files).
PolyGlot can be killed before even AutoQuit can do anything.
You can send me log files anyway, just in case it is a new problem ...
Fabien.