"Battle of the Goths" Championship, cycle #1

[hgm]

BigLion80 and ArcBishop80 are searching at 5%-15% speed compared to the other engines because of my slow home-made 10x8-Bitboard class (no Rybka node count obfuscation) and searching 5 plys
shallower. Yet they sometimes grab a point against the monsters .
Who can explain this phenomenon ? Can it be that BigLion80 and ArcBishop80 have some advantageous bugs ?
[/quote]
The factor you quote is not enough to explain 5 ply of search depth difference. I suspect that your engines count depth differnt from the others. Both my engines use LMR, which makes them report 2 ply more depth at the same search ime. But the 2 ply is only along the PV, and the side branches are on the average even more than 2 ply shallower, making it just as susceptible to tactical surprises as without the LMR. In fact for Joker the LMR hardly seemed to add any strength.

Furthermore, most engines actually search deeper than they realize, because a QS node that fails high through a capture is actually equivaent to a 1-ply search. Joker's search is fully depth-aware, and it reports this extra depth, where other engines, with the same search tree, might report one ply less.

Despite all that, I am surprised about the low depth yuor engines report. It is almost like they are thousands of times slower. But it might be that you have more extensions than the other engines, so that your search isn't as shallow as the reported depth suggests.

Furthermore, your engines do seem to have better evaluation than most of the other participants: TSCP's evaluation is truly appalling, and most other engines seem to have issues with piece values (discussed extensively in the other thread). So sometimes they use their larger depth only to force the opponent to capture a piece worth of material. And then even the 4 extra ply can't save them anymore, at such long time controls.

I certainly hope that next year's winner will be 300 Elo stronger. If only because I plan to improve Joker that much!

[Matthias Gemuh]

The factor you quote is not enough to explain 5 ply of search depth difference. I suspect that your engines count depth differnt from the others. Both my engines use LMR, which makes them report 2 ply more depth at the same search ime. But the 2 ply is only along the PV, and the side branches are on the average even more than 2 ply shallower, making it just as susceptible to tactical surprises as without the LMR. In fact for Joker the LMR hardly seemed to add any strength.

...

Despite all that, I am surprised about the low depth yuor engines report. It is almost like they are thousands of times slower. But it might be that you have more extensions than the other engines, so that your search isn't as shallow as the reported depth suggests.

Furthermore, your engines do seem to have better evaluation than most of the other participants: TSCP's evaluation is truly appalling, and most other engines seem to have issues with piece values (discussed extensively in the other thread). So sometimes they use their larger depth only to force the opponent to capture a piece worth of material. And then even the 4 extra ply can't save them anymore, at such long time controls.

I certainly hope that next year's winner will be 300 Elo stronger. If only because I plan to improve Joker that much!

It is very interesting to see you guys say that even my engines have a right to win a few games
BigLion80 has outdated data structures, so I envisage only a few more updates. Every interior node is evaluated in both engines, but results are not used yet. ArcBishop's in-built profiler is still always a bit active.
8x8 BigLion scores 260/300 in WAC at 5 sec/pos on my 2.6 GHz machine.
BigLion80 and ArcBishop80 score respectively 232 and 221.

M.

[hgm]

The fourth cycle is now completed. TJchess10x8 made a very strong comeback, beating both Joker80 and Smirf Bonus by 2-0. It is now firmly back in second place, 3 points ahead of Smirf Bonus. Joker80 did not like this opening position much, and lost several games. At the start of the final cycle, the gap between Joker80 and TJchess10x8 has narrowed to 3.5 points.

Code: Select all

                              Joke TJch Smir Smir TSCP Fair BigL ArcB
 1. Joker80 n                 #### 1111 1101 1001 1111 1111 1111 1111
                              #### 0100 0011 0110 1110 1110 1111 1111   79%  44.0

 2. TJchess10x8               0000 #### 0=00 1110 1111 1110 1111 0111
                              1011 #### 1111 1=01 1=11 1111 1010 1111   72%  40.5

 3. Smirf 1.73h               0010 1=11 #### 1011 1010 1111 0111 1111
                              1100 0000 #### 1111 0001 0111 1110 1111   67%  37.5

 4. Smirf Donation            0110 0001 0100 #### 1000 =011 101= 1110
                              1001 0=10 0000 #### 1010 01=1 1111 1111   52%  29.0

 5. TSCP Gothic               0000 0000 0101 0111 #### 10=1 10=1 111=
                              0001 0=00 1110 0101 #### 0011 1010 1111   50%  28.0

 6. Fairy-Max 4.8 v           0000 0001 0000 =100 01=0 #### 1111 100=
                              0001 0000 1000 10=0 1100 #### 100= 0011   33%  18.5

 7. BigLion80 4apr            0000 0000 1000 010= 01=0 0000 #### 0110
                              0000 0101 0001 0000 0101 011= #### =101   29%  16.0

 8. ArcBishop80 1.00          0000 1000 0000 0001 000= 011= 1001 ####
                              0000 0000 0000 0000 0000 1100 =010 ####   19%  10.5

In the 5th (final) cycle, the engines will play Carrera Chess. Except games involving TSCP, that will be played from the fantasy setup RCBNQKNBAR.