Just Below The Summit

[Graham Banks]

JUST BELOW THE SUMMIT

Intel i5 Quad
ChessGUI
256mb hash each where possible
3-4-5 piece tablebases
Ponder off
Graham2014-1F.cgb book
40 moves in 25 minutes repeating (adapted for the CCRL)
2 cycles 30 rounds
All engines 64-bit where available


Standings after Round 28

21.0 - Hannibal 1.4b 64-bit
18.5 - Senpai 1.0 64-bit
17.0 - Deep Shredder 12 64-bit
17.0 - SmarThink 1.70 64-bit
16.0 - Protector 1.6.0 64-bit
16.0 - Texel 1.04 64-bit
16.0 - Spike 1.4 Leiden
15.0 - Hiarcs 14
13.0 - DiscoCheck 5.2 64-bit
12.5 - Deep Junior Yokohama 64-bit
12.0 - Quazar 0.4 64-bit
11.0 - Gaviota 1.0 64-bit
11.0 - Deuterium 14.2.33.276 64-bit
10.0 - Nemo 1.0.1 64-bit
9.5 - Vajolet2 1.43 64-bit
8.5 - Toga II 3.0


This tournament can be followed more closely from the following link, where games will also be available for download after every two rounds:
http://kirill-kryukov.com/chess/discuss ... f=7&t=7661

Alternatively, if you install TLCV (Tom's Live Chess Viewer) on your computer, you can watch the games live move by move. You'll also be able to chat to others following the tournament in the chatroom there.
http://home.pacific.net.au/~tommyinoz/client.zip
Host - GrahamCCRL.dyndns.org Port - 16001

[lucasart]

I'm glad someone (other than me) is toying with DiscoCheck.

I fixed a ridiculous bug the other day, and DiscoCheck gained 14 elo... It's amazing how, even fairly strong engines, can still have such idiotic elo eating bugs.

It just goes to show that the difference between strong engines and the rest, is mostly the absence of bugs. People assume it's because of unique ideas and better chess knowledge: that's just marketing... Truth is, it's all about the bugs

[Uri Blass]

I'm glad someone (other than me) is toying with DiscoCheck.

I fixed a ridiculous bug the other day, and DiscoCheck gained 14 elo... It's amazing how, even fairly strong engines, can still have such idiotic elo eating bugs.

It just goes to show that the difference between strong engines and the rest, is mostly the absence of bugs. People assume it's because of unique ideas and better chess knowledge: that's just marketing... Truth is, it's all about the bugs

I do not think that we have an evidence for it.
A possible evidence may be if you take the relatively weak open source engines and show how you close most of the gap to stockfish only by fixing bugs.

I believe that it is impossible to do it only by fixing bugs
(for example if some weak engine does not have hash tables you need to add them that is adding code and not fixing bugs).

[Graham Banks]

JUST BELOW THE SUMMIT

Intel i5 Quad
ChessGUI
256mb hash each where possible
3-4-5 piece tablebases
Ponder off
Graham2014-1F.cgb book
40 moves in 25 minutes repeating (adapted for the CCRL)
2 cycles 30 rounds
All engines 64-bit where available


Final Standings

22.5 - Hannibal 1.4b 64-bit
19.5 - Senpai 1.0 64-bit
19.0 - SmarThink 1.70 64-bit
18.5 - Deep Shredder 12 64-bit
17.0 - Protector 1.6.0 64-bit
17.0 - Texel 1.04 64-bit
16.5 - Spike 1.4 Leiden
16.0 - Hiarcs 14
13.5 - Deep Junior Yokohama 64-bit
13.5 - DiscoCheck 5.2 64-bit
13.0 - Quazar 0.4 64-bit
12.5 - Gaviota 1.0 64-bit
12.0 - Deuterium 14.2.33.276 64-bit
11.0 - Nemo 1.0.1 64-bit
9.5 - Vajolet2 1.43 64-bit
9.0 - Toga II 3.0


The complete tournament pgn (zipped) can be downloaded here:
http://kirill-kryukov.com/chess/discuss ... p?id=32640