The HGM Invitational

[Adam Hair]

Well, the title is a misnomer, but it is the best sounding title I can come up with.

Sometimes, I get tired of testing in a logical,rational manner. Sometimes, I like to create a large RR consisting of engines that I know along with engines that are brand new to me and just see what the results will be.

In this case, there are a few engines I know quite well, a few I have tested some, and two (Usurpator II and Iota 1.0), that I have never tested before. 17 engines in all, 3 of them created by HGM:

Usurpator II - Harm Geert Muller
Ram 2.0 - Harm Geert Muller
Neg 0.3d - Harm Geert Muller
Iota 1.0 - Daniel White
Brutus RND - Stephan Vermeire (random mover created from Stephan's engine Brutus)
POS 1.20 - Folkert van Heusden
Supra 11.5.1 - Pedro Mourão Correia
Pytor 0.6 Am - John Marountas
Testina 2.2 - Sandro Corsini
Xadreco 5.7 - Ruben Carlo Benante
CmcChess 2.27 - Chao Ma
Strategic Deep 1.25 - Marcin Gardyjan
Safrad 2.1.35.210 - David Safranek
Chad's Chess 0.15 - Chad Koch
Delfi 5.4 1000 Elo - Fabio Cavicchio
Crafty 23.4 64-bit skill 1 - Bob Hyatt

I will have an update in the morning.

[Adam Hair]

I forgot to list EasyPeasy 1.0.0.0 (author unknown) as one of the participants.

After 174 of 4352 (maybe less) games:

Code: Select all

    Engine                     Score   Xa Sa Cm Us St Ch SU De Io Cr Te Py Ea Ne Ra PO Br    S-B
01: Xadreco 5.7                18.5/21 ·· 1  =  1  1  11 0  11 1  0  1  1  11 11 1  11 1   165.25
02: Safrad 2.1.35.210          16.0/21 0  ·· 01 =1 1  11 1  1  0  1  11 =  10 1  1  1  1   153.50
03: CmcChess 2.27              15.5/20 =  10 ·· =  =  1  11 1  1  1= =  =  1  =  11 1  1   141.50
04: Usurpator II               15.5/21 0  =0 =  ·· 10 1  =  0  11 11 1  1  1  1  11 1  1   129.00
05: Strategic Deep 1.25        14.0/20 0  0  =  01 ·· =  0  1  11 11 10 1  1  1  1  1  1   119.75
06: Chad's Chess 0.15          13.0/21 00 00 0  0  =  ·· 11 =  =  =  1  11 1  1  1  1  11   94.50
07: SUPRA 11.5.1 64-bit        12.5/20 1  0  00 =  1  00 ·· 1  1  1  0? 1  01 11 1  1  1   105.25
08: Delfi 5.4 1000 Elo         12.0/20 00 0  0  1  0  =  0  ·· 0  1  =  11 1  1  1  11 11   73.75
09: Iota 1.0 32-bit JA(tcc)    11.5/20 0  1  0  00 00 =  0  1  ·· 1= =  1  1  1  11 1  1    82.00
10: Crafty 23.4 64-bit skill 1 9.5/21  1  0  0= 00 00 =  0  0  0= ·· 1= =  1  1  1  1  1    74.50
11: Testina 2.2                9.5/20  0  00 =  0  01 0  1? =  =  0= ·· 0  1  1  =1 1  1    68.75
12: Pyotr 0.6 AM               9.5/20  0  =  =  0  0  00 0  00 0  =  1  ·· 1  11 1  11 1    53.50
13: EasyPeasy 1.0.0.0          8.0/21  00 01 0  0  0  0  10 0  0  0  0  0  ·· 1  11 1  11   40.00
14: Neg 0.3d                   5.5/21  00 0  =  0  0  0  00 0  0  0  0  00 0  ·· 1  11 11   12.75
15: Ram 2.0                    2.0/21  0  0  00 00 0  0  0  0  00 0  =0 0  00 0  ·· =  1     5.75
16: POS 1.20                   1.0/20  00 0  0  0  0  0  0  00 0  0  0  00 0  00 =  ·· =     1.25
17: Brutus RND                 0.5/20  0  0  0  0  0  00 0  00 0  0  0  0  00 00 0  =  ··    0.50

[gerold]

Thanks Adam. Should be interesting games.

Best,
Gerold.

[Adam Hair]

After 4 rounds:

Code: Select all

    Engine                     Score     Us   Xa   St   Cm   Sa   SU   Ch   Io   Te   Cr   Py   De   Ea   Ne   Ra   PO   Br    S-B
01: Usurpator II               54.0/64 ···· 001= 1001 =111 =011 =1=1 1111 1111 1101 1111 1111 0=11 1111 1111 1111 1111 1111  1523.0
02: Xadreco 5.7                51.0/64 110= ···· 1111 =0=0 1111 0010 1110 11=1 1101 0=1= 1011 1111 1111 1111 1111 1111 1111  1430.2
03: Strategic Deep 1.25        49.0/64 0110 0000 ···· ==1= 0=01 01=1 =110 1111 1011 1111 111= 11=1 1111 1111 1111 1111 1111  1295.5
04: CmcChess 2.27              48.0/64 =000 =1=1 ==0= ···· 101= 11=1 1110 1111 =1== 1=1= ==== 1111 1=1= =111 1111 1111 1111  1321.2
05: Safrad 2.1.35.210          43.5/64 =100 0000 1=10 010= ···· 10=0 1101 0==1 1111 1111 =11= 110= 1010 1111 1111 1111 1111  1124.7
06: SUPRA 11.5.1 64-bit        43.0/64 =0=0 1101 10=0 00=0 01=1 ···· 0010 1=11 0101 1111 1011 1=11 0101 11=1 1111 1111 1111  1128.2
07: Chad's Chess 0.15          42.5/64 0000 0001 =001 0001 0010 1101 ···· =01= 1101 =111 1111 =110 1111 1111 1111 1111 1111  1030.7
08: Iota 1.0 32-bit JA(tcc)    35.5/64 0000 00=0 0000 0000 1==0 0=00 =10= ···· =1=1 1=11 1=11 1=11 1=10 1111 11=1 1=11 11=1  807.25
09: Testina 2.2                32.0/64 0010 0010 0100 =0== 0000 1010 0010 =0=0 ···· 0=0= 00== =110 1=11 1111 =111 1111 1111  717.75
10: Crafty 23.4 64-bit skill 1 32.0/64 0000 1=0= 0000 0=0= 0000 0000 =000 0=00 1=1= ···· =111 0110 111= 1111 1111 1111 1111  649.00
11: Pyotr 0.6 AM               31.5/64 0000 0100 000= ==== =00= 0100 0000 0=00 11== =000 ···· 0011 1111 1111 1111 1111 1111  652.75
12: Delfi 5.4 1000 Elo         31.5/64 1=00 0000 00=0 0000 001= 0=00 =001 0=00 =001 1001 1100 ···· 1111 1111 1111 1111 1111  650.75
13: EasyPeasy 1.0.0.0          22.5/64 0000 0000 0000 0=0= 0101 1010 0000 0=01 0=00 000= 0000 0000 ···· 11=1 111= 1111 1111  407.50
14: Neg 0.3d                   11.0/64 0000 0000 0000 =000 0000 00=0 0000 0000 0000 0000 0000 0000 00=0 ···· 1=10 11=1 11=1  105.75
15: Ram 2.0                    10.5/64 0000 0000 0000 0000 0000 0000 0000 00=0 =000 0000 0000 0000 000= 0=01 ···· =111 1111   85.25
16: POS 1.20                   4.5/64  0000 0000 0000 0000 0000 0000 0000 0=00 0000 0000 0000 0000 0000 00=0 =000 ···· =11=   34.50
17: Brutus RND                 2.0/64  0000 0000 0000 0000 0000 0000 0000 00=0 0000 0000 0000 0000 0000 00=0 0000 =00= ····   27.75

[Dan Honeycutt]

Hi Adam,

If you (or anyone else) run any more tournaments for weak engines I'd love to see some results for Cupcake.

Best
Dan H.

[Adam Hair]

Hi Adam,

If you (or anyone else) run any more tournaments for weak engines I'd love to see some results for Cupcake.

Best
Dan H.

Hi Dan,

I have a gauntlet running for Cupcake at the moment. I can post some preliminary results in the morning.

Adam

[Dan Honeycutt]

Hi Adam,

If you (or anyone else) run any more tournaments for weak engines I'd love to see some results for Cupcake.

Best
Dan H.

Hi Dan,

I have a gauntlet running for Cupcake at the moment. I can post some preliminary results in the morning.

Adam

Super. Thanks, Adam.

Best
Dan H.

[Adam Hair]

I missed on my initial guess of Cupcake's strength and have adjusted accordingly. I will fill in additional appropriate opponents to get the number of games played to around 800+.

Code: Select all

    Engine              Score                        Cu
01: Cupcake 1.0         116.5/223 ····················· 
02: Genesis 2.0         16.0/20   11111011111=111=1==0  
02: WJChess 1.64        16.0/19   11111=011111=1=11=1   
04: Lime 66             13.5/20   01111111=1011000==11  
05: BlackBishop 1.0     12.0/20   0011=00==0101111111=  
06: Kurt 0.9.2.2 64-bit 10.5/20   =10111=111000110=000  
06: Bearded 06.29       10.5/19   ==1011100011=100110   
08: Micro-Max 4.8       6.5/21    0==000=01000010101010 
09: Smash 1.03          6.0/21    000000000101=011=0001 
10: Adamant 1.7         5.5/21    01000011000=00=00001= 
11: Bace 0.45           5.0/21    100100=00001000100=00 
11: Beaches 2.32        5.0/21    0000=0=10=00001100=00 

[Dan Honeycutt]

I missed on my initial guess of Cupcake's strength and have adjusted accordingly. I will fill in additional appropriate opponents to get the number of games played to around 800+.

Thanks ever so much, Adam. From what you have so far it looks like Cupcake would be around 2000 - 2100 on your Also-Rans Rating List. That's where I think a program with a very basic search and eval should be if it doesn't have bugs so, hopefully, Cupcake doesn't have too many bugs.

One bug I know of - it can lose input if it arrives during search and it's not move now (?). That shouldn't affect tournament play but let me ask you, did you observe any misbehavior in your tests?

Thanks again,
Dan H.

[Adam Hair]

I missed on my initial guess of Cupcake's strength and have adjusted accordingly. I will fill in additional appropriate opponents to get the number of games played to around 800+.

Thanks ever so much, Adam. From what you have so far it looks like Cupcake would be around 2000 - 2100 on your Also-Rans Rating List. That's where I think a program with a very basic search and eval should be if it doesn't have bugs so, hopefully, Cupcake doesn't have too many bugs.

One bug I know of - it can lose input if it arrives during search and it's not move now (?). That shouldn't affect tournament play but let me ask you, did you observe any misbehavior in your tests?

Thanks again,
Dan H.

I have not seen any misbehavior. If I do, I will document it and send you the information.