CCRL Engine Authors' Tournament (will start October 1st)

Sergio Martinez · Post by **Sergio Martinez** » Wed Oct 26, 2016 11:22 am

CCRL Adapted (40/40)

Group A

                                    Score     Han Nir iCE And Chi Vaj
----------------------------------------------------------------------
 1: Hannibal 1.7 64-bit            9.5 / 15   XXX 1=0 1== ==1 =11 10=  (+6 -2 =7)
 2: Nirvanachess 01092016 64-bit   8.5 / 15   0=1 XXX 0== =0= 11= 11=  (+5 -3 =7)
 3: iCE 3.0 64-bit                 8.0 / 15   0== 1== XXX === 0=1 =1=  (+3 -2 =10)
 4: Andscacs 0.87.163 64-bit       7.5 / 15   ==0 =1= === XXX === ===  (+1 -1 =13)
 5: Chiron 030916 64-bit           6.5 / 15   =00 00= 1=0 === XXX 1=1  (+3 -5 =7)
 6: Vajolet2 2.2 64-bit            5.0 / 15   01= 00= =0= === 0=0 XXX  (+1 -6 =8)
----------------------------------------------------------------------
45 games: +10 =26 -9

Group B

Code: Select all

                               Score     La Ro Ar Wa Ne Tu Da Am
-----------------------------------------------------------------
 1: Laser 1.2 64-bit          7.0 / 11   XX 0= 1. == 1. 1= == 1.  (+4 -1 =6)
 2: Rodent II 0.9.64 64-bit   6.5 / 11   1= XX =0 =. =. == 1. 1=  (+3 -1 =7)
 3: Arasan 19.1 64-bit        6.0 / 11   0. =1 XX 1. =0 1. == 10  (+4 -3 =4)
 4: Wasp 1.25 64-bit          6.0 / 11   == =. 0. XX =0 1. == 11  (+3 -2 =6)
 5: Nemorino 1.02 64-bit      5.5 / 11   0. =. =1 =1 XX 00 =. =1  (+3 -3 =5)
 6: Tucano 6.00 64-bit        5.0 / 11   0= == 0. 0. 11 XX 10 =.  (+3 -4 =4)
 7: Daydreamer 2.0p2          4.5 / 11   == 0. == == =. 01 XX 0.  (+1 -3 =7)
 8: Amoeba 1.4 64-bit         3.5 / 11   0. 0= 01 00 =0 =. 1. XX  (+2 -6 =3)
-----------------------------------------------------------------
44 games: +14 =21 -9

Group C (finished)

Code: Select all

    Score     Zurichess  Cheese 1.8 DanaSah 6. Floyd 0.9  Ethereal 7 Capivara L
---------------------------------------------------------------------------------------------------------------
 1: Zurichess Jura 64-bit        36.5 / 50   XXXXXXXXXX =1=11==011 =11=1=101= 0===1=1101 1110111111 10111011=1  (+30 -7 =13)
 2: Cheese 1.8 64-bit            27.5 / 50   =0=00==100 XXXXXXXXXX ==0010=0=0 =1=01=1010 1101111111 =01==1111=  (+20 -15 =15)
 3: DanaSah 6.30                 27.0 / 50   =00=0=010= ==1101=1=1 XXXXXXXXXX 01001=01=1 01=0111=== 1=11===100  (+18 -14 =18)
 4: Floyd 0.9 64-bit             25.0 / 50   1===0=0010 =0=10=0101 10110=10=0 XXXXXXXXXX 11100001=0 01==111011  (+19 -19 =12)
 5: Ethereal 7.78 64-bit         19.0 / 50   0001000000 0010000000 10=1000=== 00011110=1 XXXXXXXXXX ===1==1111  (+14 -26 =10)
 6: Capivara LK 0.09b02G 64-bit  15.0 / 50   01000100=0 =10==0000= 0=00===011 10==000100 ===0==0000 XXXXXXXXXX  (+7 -27 =16)
---------------------------------------------------------------------------------------------------------------
150 games: +63 =42 -45

Group D (finished)

Code: Select all

              Score     Gogobello  Cerebro 4. Schooner 1 Orion 0.3. Averno 0.8 Chexa 0.49
--------------------------------------------------------------------------------------------------------
 1: Gogobello 1.0 64-bit  34.0 / 50   XXXXXXXXXX 1=11000111 110=1011=0 01110==111 11==01111= =0=11=1=11  (+28 -10 =12)
 2: Cerebro 4.03 64-bit   31.0 / 50   0=00111000 XXXXXXXXXX 101011100= =0=1111101 =10=11=111 =010111111  (+27 -15 =8)
 3: Schooner 1.57 64-bit  27.5 / 50   001=0100=1 010100011= XXXXXXXXXX ====01=0== 1=1===0=11 1101110111  (+20 -15 =15)
 4: Orion 0.3.0 64-bit    24.0 / 50   10001==000 =1=0000010 ====10=1== XXXXXXXXXX 0===11110= 101==11=10  (+15 -17 =18)
 5: Averno 0.81           19.0 / 50   00==10000= =01=00=000 0=0===1=00 1===00001= XXXXXXXXXX 11=0110=1=  (+10 -22 =18)
 6: Chexa 0.49991 64-bit  14.5 / 50   =1=00=0=00 =101000000 0010001000 010==00=01 00=1001=0= XXXXXXXXXX  (+9 -30 =11)
--------------------------------------------------------------------------------------------------------
150 games: +63 =41 -46

Group E (finished)

Code: Select all

                                  Score     Jumbo 0.1. Dorpsgek B Pigeon 1.5 Satana.2.4 Neurone_XX Soberango 
--------------------------------------------------------------------------------------------------------------
 1: Jumbo 0.1.10                46.5 / 50   XXXXXXXXXX 111=1=0=1= =111111111 1111111111 1111111111 1111111111  (+44 -1 =5)
 2: Dorpsgek BloodyMary 64-bit  42.5 / 50   000=0=1=0= XXXXXXXXXX 11=1111111 1111111111 1111111111 1111111111  (+40 -5 =5)
 3: Pigeon 1.5.0 64-bit         26.0 / 50   =000000000 00=0000000 XXXXXXXXXX 1=1===0=1= 1111=111=1 1111111111  (+21 -19 =10)
 4: Satana.2.4.19.64-bit        19.5 / 50   0000000000 0000000000 0=0===1=0= XXXXXXXXXX 1====10011 1111111=11  (+14 -25 =11)
 5: Neurone_XXIII                8.5 / 50   0000000000 0000000000 0000=000=0 0====01100 XXXXXXXXXX 0001=00011  (+5 -38 =7)
 6: Soberango 0.08.2l            7.0 / 50   0000000000 0000000000 0000000000 0000000=00 1110=11100 XXXXXXXXXX  (+6 -42 =2)
--------------------------------------------------------------------------------------------------------------
150 games: +64 =20 -66

Graham Banks · Post by **Graham Banks** » Fri Oct 28, 2016 11:57 pm

Sven Schüle wrote:
hgm wrote:ChessGUI expects the engines to look at the full-move number of FENs to know when a new session starts?

That certainly breaks all of my engines!

I always assumed that all serious testing is done by feeding moves, and never by starting from positions. I often don't even bother to equip them with setboard.
Even with setboard it can work. In Jumbo (as in all my previous engines) I ignore the full move number given through setboard, and if I wouldn't ignore it then I would certainly not assume that some of the moves that belong to a time control session have already been played. So if I get a "level 40 X Y" then the number of remaining moves starts at 40 for me no matter what comes in setboard.

Therefore, considering that Graham tested Jumbo successfully under ChessGUI (see his post above), I'm not sure whether we already know all about this problem ...

I've had to use the wb2uci adapters with both Jumbo and Dorpsgek to overcome the issue.

Sven · Post by **Sven** » Sat Oct 29, 2016 12:42 am

Graham Banks wrote:
Sven Schüle wrote:
hgm wrote:ChessGUI expects the engines to look at the full-move number of FENs to know when a new session starts?

That certainly breaks all of my engines!

I always assumed that all serious testing is done by feeding moves, and never by starting from positions. I often don't even bother to equip them with setboard.
Even with setboard it can work. In Jumbo (as in all my previous engines) I ignore the full move number given through setboard, and if I wouldn't ignore it then I would certainly not assume that some of the moves that belong to a time control session have already been played. So if I get a "level 40 X Y" then the number of remaining moves starts at 40 for me no matter what comes in setboard.

Therefore, considering that Graham tested Jumbo successfully under ChessGUI (see his post above), I'm not sure whether we already know all about this problem ...
I've had to use the wb2uci adapters with both Jumbo and Dorpsgek to overcome the issue.

Ok, that explains a lot. So without wb2uci Jumbo would suffer from the same problem under ChessGUI as several other WB engines.

Graham Banks · Post by **Graham Banks** » Sat Oct 29, 2016 1:28 am

Sven Schüle wrote:
Graham Banks wrote:
Sven Schüle wrote:
hgm wrote:ChessGUI expects the engines to look at the full-move number of FENs to know when a new session starts?

That certainly breaks all of my engines!

I always assumed that all serious testing is done by feeding moves, and never by starting from positions. I often don't even bother to equip them with setboard.
Even with setboard it can work. In Jumbo (as in all my previous engines) I ignore the full move number given through setboard, and if I wouldn't ignore it then I would certainly not assume that some of the moves that belong to a time control session have already been played. So if I get a "level 40 X Y" then the number of remaining moves starts at 40 for me no matter what comes in setboard.

Therefore, considering that Graham tested Jumbo successfully under ChessGUI (see his post above), I'm not sure whether we already know all about this problem ...
I've had to use the wb2uci adapters with both Jumbo and Dorpsgek to overcome the issue.
Ok, that explains a lot. So without wb2uci Jumbo would suffer from the same problem under ChessGUI as several other WB engines.

Yes.

Graham Banks · Post by **Graham Banks** » Sat Oct 29, 2016 4:30 am

CCRL AUTHORS GROUP D

Intel i5 Quad
ChessGUI
256mb hash each where possible
3-4-5 piece tablebases
Ponder off
GBSelect2016-2.cgb book
40 moves in 25 minutes repeating (adapted for the CCRL)
5 cycles 50 rounds
All engines 64-bit 1CPU where available

Final Standings

33.5 - Cerebro 4.03
28.5 - Gogobello 1.0 64-bit
26.5 - Orion 0.3.0 64-bit
24.5 - Schooner 1.5.7 64-bit
20.5 - Averno 0.81
16.5 - Chexa 0.4.9991

The complete tournament pgn (zipped) can be downloaded here:
http://kirill-kryukov.com/chess/discuss ... p?id=38250

Graham Banks · Post by **Graham Banks** » Sat Oct 29, 2016 4:36 am

CCRL AUTHORS GROUP E

Intel i5 Quad
ChessGUI
256mb hash each where possible
3-4-5 piece tablebases
Ponder off
GBSelect2016-2.cgb book
40 moves in 25 minutes repeating (adapted for the CCRL)
5 cycles 50 rounds
All engines 64-bit 1CPU where available

Participants

Jumbo 1.1.10
Pigeon 1.5.1 64-bit
Dorpsgek Bloody Mary
Satana 2.4.19b
Neurone XXIII
Soberango 0.08.21

Neurone only uses 1-2 seconds for every move despite anything I've tried.

Now underway with live broadcast in TLCV port 16001.
I'll post updates here - http://kirill-kryukov.com/chess/discuss ... f=7&t=8828

Sven · Post by **Sven** » Sat Oct 29, 2016 5:28 pm

Graham Banks wrote:CCRL AUTHORS GROUP E

[...]

Participants

Jumbo 1.1.10
Pigeon 1.5.1 64-bit
Dorpsgek Bloody Mary
Satana 2.4.19b
Neurone XXIII
Soberango 0.08.21

Jumbo version is 0.1.10

Graham Banks · Post by **Graham Banks** » Sat Oct 29, 2016 10:31 pm

Sven Schüle wrote:
Graham Banks wrote:CCRL AUTHORS GROUP E

[...]

Participants

Jumbo 1.1.10
Pigeon 1.5.1 64-bit
Dorpsgek Bloody Mary
Satana 2.4.19b
Neurone XXIII
Soberango 0.08.21
Jumbo version is 0.1.10

Oops - typo. Sorry.

Graham Banks · Post by **Graham Banks** » Wed Nov 02, 2016 6:54 am

CCRL AUTHORS GROUP E

Intel i5 Quad
ChessGUI
256mb hash each where possible
3-4-5 piece tablebases
Ponder off
GBSelect2016-2.cgb book
40 moves in 25 minutes repeating (adapted for the CCRL)
5 cycles 50 rounds
All engines 64-bit 1CPU where available

Final Standings

45.0 - Jumbo 0.1.10
40.0 - Dorpsgek Bloody Mary 64-bit
32.5 - Pigeon 1.5.1 64-bit
21.0 - Satana.2.4.19.64-bit
6.0 - Soberango 0.08.21
5.5 - Neurone XXIII

The complete tournament pgn (zipped) can be downloaded here:
http://kirill-kryukov.com/chess/discuss ... p?id=38286

Graham Banks · Post by **Graham Banks** » Tue Nov 08, 2016 10:34 pm

My 40/4 results

Group A

29.5 - Nirvanachess 01092016 64-bit
28.5 - Andscacs 0.87163 64-bit
26.5 - Chiron 030916 64-bit
24.0 - Hannibal 1.7 64-bit
22.5 - iCE 3.0 64-bit
19.0 - Vajolet2 2.2 64-bit

Group B

43.0 - Nemorino 1.02 64-bit
42.5 - Arasan 19.1 64-bit
41.0 - Daydreamer 2.0p2 64-bit
40.5 - Rodent II 0.9.64 64-bit
35.5 - Wasp 1.25 64-bit
30.5 - Laser 1.2 64-bit
23.5 - Amoeba 1.4 64-bit
23.5 - Tucano 6.00 64-bit

Group C

33.5 - Zurichess Jura 64-bit
29.0 - Cheese 1.8 64-bit
26.5 - DanaSah 6.3
26.0 - Floyd 0.9 64-bit
22.0 - Ethereal 7.78 64-bit
13.0 - Capivara LK 0.09b02g 64-bit

Group D

31.5 - Gogobello 1.0 64-bit
29.0 - Schooner 1.5.7 64-bit
28.0 - Cerebro 4.03
24.5 - Orion 0.3.0 64-bit
24.0 - Averno 0.81
13.0 - Chexa 0.4.9991

Group E

43.0 - Jumbo 0.1.10
41.0 - Dorpsgek Bloody Mary 64-bit
31.5 - Pigeon 1.5.1 64-bit
21.5 - Satana.2.4.19.64-bit
8.0 - Neurone XXIII
5.0 - Soberango 0.08.21

CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)

Re: CCRL Engine Authors' Tournament (will start October 1st)