First half of the CCCC tournament has ended.
It's a 46 rounds tournament and so 23 rounds have been played. Time control is 15'+5" and engines will play all against all twice, one with black and one with white, while pondering is on and no opening books are used and 6 pieces tablebases are being used.
Top 8 engines promote to next round.
Till now there are 4 distinctive groups that have been formed. The top 3 that are consisted of the known "big 3" of computer Chess this moment and for the last 4-5 years, Stockfish, Komodo and Houdini, the other group the engines that fight for the top 8, Leela, Shredder are the leaders of this group that show they might have the advantage and the other Fire, Booot, Ethereal and Andscacs.
There is a group 8-9 engines that seems to just be in the middle without having any prospects of going to top 8 and there is the low end of the tournament that does not do that good.
There are 4 engines that till now haven't lost any game so far. Stockfish, Houdini, Komodo and Leela.
The standings after round 23, finishing the first half of the tournament.
Player | Score | (SB) | H | K | S | L | S | F | B | E | A | F | X | V | P | T | G | F | A | L | N | W | I | S | C | N | +/-/= | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1: | Houdini 6.03 | 19.0/23 | 198.50 | X | = | = | = | = | 1 | = | = | 1 | 1 | 1 | = | 1 | 1 | 1 | 1 | 1 | 1 | 1 | = | 1 | 1 | 1 | 1 | (+15 -0 =8) |
2: | Komodo 2118.00 | 18.5/23 | 195.25 | = | X | = | = | 1 | 1 | 1 | = | = | = | = | 1 | 1 | 1 | = | 1 | = | 1 | 1 | 1 | 1 | 1 | 1 | 1 | (+14 -0 =9) |
3: | Stockfish 220818 | 18.5/23 | 193.00 | = | = | X | = | = | = | 1 | 1 | = | 1 | = | 1 | 1 | 1 | 1 | 1 | = | = | 1 | 1 | 1 | 1 | 1 | 1 | (+14 -0 =9) |
4: | Lc0 17.11089 | 16.5/23 | 179.25 | = | = | = | X | = | = | = | 1 | = | 1 | 1 | 1 | = | 1 | 1 | = | 1 | 1 | = | = | 1 | = | = | 1 | (+10 -0 =13) |
5: | Shredder 13 | 16.0/23 | 164.25 | = | 0 | = | = | X | 1 | = | = | 1 | = | = | = | = | 1 | 1 | = | 1 | = | 1 | = | 1 | 1 | 1 | 1 | (+10 -1 =12) |
6: | Fire 7.1 | 15.5/23 | 151.50 | 0 | 0 | = | = | 0 | X | = | 1 | = | 1 | = | = | 1 | 1 | 1 | = | 1 | 1 | 1 | = | = | 1 | 1 | 1 | (+11 -3 =9) |
7: | Booot 6.3.1 | 14.5/23 | 140.00 | = | 0 | 0 | = | = | = | X | = | = | = | = | = | = | 1 | = | = | = | 1 | 1 | 1 | 1 | 1 | 1 | 1 | (+8 -2 =13) |
8: | Ethereal 10.88 | 14.0/23 | 138.50 | = | = | 0 | 0 | = | 0 | = | X | 1 | = | 1 | = | = | 0 | 1 | = | 1 | 1 | 1 | = | 1 | = | 1 | 1 | (+9 -4 =10) |
9: | Andscacs 0.94 | 13.5/23 | 128.50 | 0 | = | = | = | 0 | = | = | 0 | X | 0 | 0 | 1 | = | = | 1 | = | 1 | 1 | 1 | 1 | = | 1 | 1 | 1 | (+9 -5 =9) |
10: | Fritz 16.10 | 12.0/23 | 116.75 | 0 | = | 0 | 0 | = | 0 | = | = | 1 | X | = | 1 | 1 | 0 | = | = | = | = | = | 1 | = | = | 1 | 1 | (+6 -5 =12) |
11: | Xiphos 0.3.17 | 11.5/23 | 114.50 | 0 | = | = | 0 | = | = | = | 0 | 1 | = | X | 0 | = | = | = | = | = | = | 1 | = | 0 | 1 | 1 | 1 | (+5 -5 =13) |
12: | Vajolet 2.6 | 11.0/23 | 104.50 | = | 0 | 0 | 0 | = | = | = | = | 0 | 0 | 1 | X | = | = | = | = | = | 1 | 0 | = | = | 1 | 1 | 1 | (+5 -6 =12) |
13: | Pedone 1.8 | 11.0/23 | 101.00 | 0 | 0 | 0 | = | = | 0 | = | = | = | 0 | = | = | X | = | = | 1 | = | = | = | = | = | 1 | 1 | 1 | (+4 -5 =14) |
14: | Texel 1.07 | 11.0/23 | 96.00 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | = | 1 | = | = | = | X | = | = | = | = | 1 | = | = | 1 | 1 | 1 | (+6 -7 =10) |
15: | Gull 3.syz | 10.0/23 | 88.50 | 0 | = | 0 | 0 | 0 | 0 | = | 0 | 0 | = | = | = | = | = | X | = | = | = | = | 1 | 1 | 1 | 1 | = | (+4 -7 =12) |
16: | Fizbo 1.9 | 9.5 / 23 | 96.75 | 0 | 0 | 0 | = | = | = | = | = | = | = | = | = | 0 | = | = | X | 0 | = | = | = | = | = | 1 | = | (+1 -5 =17) |
17: | Arasan CCCC-2018 | 9.5 / 23 | 89.75 | 0 | = | = | 0 | 0 | 0 | = | 0 | 0 | = | = | = | = | = | = | 1 | X | = | 0 | 1 | 0 | = | 1 | 1 | (+4 -8 =11) |
18: | Laser 1.6 | 9.0 / 23 | 78.00 | 0 | 0 | = | 0 | = | 0 | 0 | 0 | 0 | = | = | 0 | = | = | = | = | = | X | 1 | = | 0 | 1 | 1 | 1 | (+4 -9 =10) |
19: | Nemorino 5.00 | 8.5 / 23 | 71.25 | 0 | 0 | 0 | = | 0 | 0 | 0 | 0 | 0 | = | 0 | 1 | = | 0 | = | = | 1 | 0 | X | = | 1 | 1 | = | 1 | (+5 -11 =7) |
20: | Wasp 3.25 | 8.0 / 23 | 84.00 | = | 0 | 0 | = | = | = | 0 | = | 0 | 0 | = | = | = | = | 0 | = | 0 | = | = | X | 0 | 1 | = | = | (+1 -8 =14) |
21: | Ivanhoe 999946h | 8.0 / 23 | 81.50 | 0 | 0 | 0 | 0 | 0 | = | 0 | 0 | = | = | 1 | = | = | = | 0 | = | 1 | 1 | 0 | 1 | X | 0 | = | 0 | (+4 -11 =8) |
22: | Senpai 2.0 | 4.5 / 23 | 42.00 | 0 | 0 | 0 | = | 0 | 0 | 0 | = | 0 | = | 0 | 0 | 0 | 0 | 0 | = | = | 0 | 0 | 0 | 1 | X | = | = | (+1 -15 =7) |
23: | Crafty 25.2 | 3.5 / 23 | 25.75 | 0 | 0 | 0 | = | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | = | = | = | = | X | 1 | (+1 -17 =5) |
24: | Nirvana 2.4 | 3.0 / 23 | 24.00 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | = | = | 0 | 0 | 0 | = | 1 | = | 0 | X | (+1 -18 =4) |
And the ratings:
Rank Name Elo + - games score oppo. draws 1 Houdini 6.03 495 130 113 23 83% 265 35% 2 Stockfish 220818 480 125 111 23 80% 266 39% 3 Komodo 2118.00 478 126 111 23 80% 266 39% 4 Shredder 13 392 111 104 23 70% 270 52% 5 Lc0 17.11089 389 109 103 23 70% 270 61% 6 Fire 7.1 385 114 107 23 67% 270 39% 7 Booot 6.3.1 351 106 102 23 63% 272 57% 8 Ethereal 10.88 343 111 107 23 61% 272 43% 9 Andscacs 0.94 339 111 108 23 59% 272 39% 10 Fritz 16.10 286 106 106 23 52% 275 52% 11 Xiphos 0.3.17 285 104 104 23 50% 275 57% 12 Vajolet 2.6 268 104 105 23 48% 275 52% 13 Pedone 1.8 265 101 102 23 48% 275 61% 14 Texel 1.07 254 106 107 23 48% 276 43% 15 Gull 3.syz 252 103 105 23 46% 276 57% 16 Arasan CCCC-2018 229 106 109 23 41% 277 48% 17 Fizbo 1.9 227 99 101 23 41% 277 74% 18 Laser 1.6 216 106 110 23 39% 278 43% 19 Wasp 3.25 193 103 107 23 35% 279 61% 20 Nemorino 5.00 190 110 116 23 37% 279 30% 21 Ivanhoe 999946h 180 110 118 23 35% 279 35% 22 Senpai 2.0 74 115 132 23 20% 284 30% 23 Crafty 25.2 27 120 144 23 15% 286 22% 24 Nirvana 2.4 0 126 156 23 13% 287 17%
**This is with Lc0 - Gull game counted as draw. Since this is a rating list the real result should be counted, as the CCCC result by their rules because Gull crashed, was 1-0 in favor of Lc0.
•Houdini has an exceptional tournament thus far, winning against almost every engine below top 8. In this short time control Houdini is to be expected to continue doing great.
•Komodo is very solid also and the only engine that has won against an engine of the top 5 with its win against Shredder. Komodo played some wonderful games. This short time control doesn't fit perfectly to Komodo, but Komodo is super strong at every time control of course.
•Stockfish as expected is very high in the ranking but being the strongest engine in the world this time and being 2nd can be considered underperformance but of course too small sample size with just 23 games to draw any conclusions. Stockfish just plays unbelievably solid with just a bit waiting of its opponents to play even the smallest inaccuracies that is expert in taking advantage of them.
•Leela being 4th and undefeated is a nice performance so far for an engine that was born before 6 months. It has shown in many games a different approach from traditional games and understanding of positions. It has played some very spectacular games. She has played also some of the most bizarre games. She still suffers a bit from not creating complications against weaker opponents so she draws a lot against weaker opponents.
•Shredder started perfectly but its performance has fallen a bit. Still plays aggressively but also very solid. Had just one defeat so far from Komodo.
•Fire is very solid as expected, very good at tactics and doing exceptionally against weaker engines. It has 3 losses though from the top 5.
•Booot is having a surprisingly good tournament so far and this is because it is killing the low end engines! It has a lot of draws though with middle group of engines but lost only 2 times from the top 5.
•Ethereal started badly, very badly but as expected it climbed to top 8 and will probably stay there. IT seems unstable though and while it has some good moments where it shows its true potential, it has some moments where it plays dubiously.
•Andscacs is also killing the weakest engines but has many problems against the first 11 having lost already 5 times. It's expected to be on top 8 replacing Booot, but perhaps Booot is going to hold.
•Fritz is solid enough but is not having enough wins against weaker engines to fight for top 8. Probably will be 9-10 not having a chance to try for top 8.
•Xiphos is relatively solid having 13 draws so far but it's strength is not for top 8.
•Vajolet is doing ok for the lowest 12 of the tournament but is losing a lot when facing the highest 12. Even though she drew against Houdini.
•Pedone plays some clever Chess, has 14 draws so far but suffers a lot against the top 10 engines.
•Texel plays an active Chess and is relativelly good at tactics but stands no chance for the top 8 where it lost most of the games aganist them.
•Gull is not winning enough even though it is rather solid and its place is expected.
•Fizbo is underperforming here so far with all these draws. It has 1 win 5 losses and 17 draws! It is a better engines than this.
•Arasan is having a logical performance of its strength so far. It got also some nice results like the 2 draws against Stockfish and Komodo.
•Laser is severely underperforming here. This is very strange so far.
•Nemorino is playing some risky Chess but this does not mean it's for its own good. Its place is to be expected.
•Wasp seems very solid most of the times early simplificating the position in many games. It got 3 draws against Houdini, Leela and Shredder.
•Ivanhoe seems very unstable and dubious playing many times to be able to be higher in the ranking.
•Senpai is one of the weakest engines of the tournament so it is to be expected this low place.
•Crafty, the good old Crafty is trying to compete with the new generation but obviously it is too much for it. It is getting outplayed in most games. Took a draw from Leela though (even in a lost position where Leela didn't manage to find the win).
•Nirvana for some reason is severely underperforming. It plays very weak Chess and this is strange since it's usually much stronger than this.
45% of the games are draws. A rather good result of decided games.
A whole lot of games have French defense! Engines for some reason prefer 1...e6 over "normal" 1...e5 after 1.e4.
Score Game length Frequency 1-0 =-= 0-1 1-0 =-= 0-1 All games 56.1% 79 96 78 33.6% 44.9% 21.3%
Move Frequency Score Draw 1: e4 152: 55.0% 52.9% 44% 2: d4 102: 36.9% 61.2% 46% 3: Nf3 22: 7.9% 54.5% 45%
I wonder, was the author of this post aware that some engines are not playing at full capacity due to pre-tournament decisions regarding stability during testing? For example, Fizbo's underperformance may be due to the fact that it is only version 1.9, while the latest version 2.0 has been out for a while and is considered much stronger, but was disqualified because apparently it crashed on the CCCC hardware a lot during testing. A similar story with Gull, I believe.
ReplyDeleteSure these are known, but still Fizbo 1.9 should be a bit stronger. Especially Laser and Nirvana.
DeleteThanks for the stats. Indeed, lots of French games for some reason!
ReplyDeleteYes, lots of French Defense. Same thing happened during early seasons of TCEC when there were no books used. Almost every game was French. This is the reason why they introduced short book moves in the opening.
ReplyDeleteAdverbs and adjectives differ. You don't appear to know this.
ReplyDeleteWell being a non native English speaker it's a wonder i'm even talking to you.
DeleteThe author took the time to type up this detailed blog post and that's your comment?
Deletehttps://www.youtube.com/watch?v=VlSkPA60ujQ
The author took the time to type up this detailed blog post and that's your comment?
ReplyDeletehttps://www.youtube.com/watch?v=VlSkPA60ujQ