WCCC 2022

[mclane]

Noteworthy is the nice result of Shredder, which played on hardware using ARM processors.

Which hardware was this ?

[dkappe]

Noteworthy is the nice result of Shredder, which played on hardware using ARM processors.

Which hardware was this ?

According to the website, 64 ARM CPU. If it used a Raspberry PI and lost all its games, would we call that a “nice result?”

[Eduard]

I am frustrated. What kind of organization is ICGA?

The blitz tournament (Speed Chess) ended two hours ago. And yet no one knows who won.

Only with difficulty in the chat of Lc0 did someone announce that Lc0 and Ginkgo had won the tournament. This can not be. Are there 2 winner? Who won armageddon?

Who won? There are no results and no table on the homepage. Why did I watch live? How long do you have to wait for the result?

crem (Lc0) posted on discord:

[Eduard]

Jokingly but with a grain of truth it can be stated: Soon there will be more trophies at the ICGA World Championship than participants.

[crem]

I am frustrated. What kind of organization is ICGA?

The blitz tournament (Speed Chess) ended two hours ago. And yet no one knows who won.

Only with difficulty in the chat of Lc0 did someone announce that Lc0 and Ginkgo had won the tournament. This can not be. Are there 2 winner? Who won armageddon?

Indeed they decided that there are two winners because there were no time for tie breaks, we were going to be kicked out of venue at 18:00.

I heard that in the one of previous tournaments, they went to hotel to finish games.

Ginkgo and Lc0 won Shredder both colors, Komodo won one color (don't know which), all other games were draw.

[Cumnor]

results and standings here but no sign of games yet https://icga.org/?page_id=3544

[Uri Blass]

The Armageddon game of Lc0 was strong. The book system played was good (I would have played like that too). The position was extremely closed which is good for Lc0. Maybe just a little luck was missing. The idea of ​​1.h3 before Armageddon was good too. This is how Lc0 should have played against Komodo and Ginkgo in the normal games!!

Incomprehensible to me was Komodo's white-color repetition of the same line played before. In the Repetition seemed to me that the operator had imported the long variant into the book between games. Because in the second game, Komodo didn't waste any time until move 28 (he even got bonus time). Lc0, on the other hand, took 50% of its time. However, the time did not matter.

I play a lot on PlayChess Blitz. If the "correct" variants are chosen, then a 10x faster computer cannot win a game. It is very easy to draw with 2000 kns against 20000 kns. This is why Armageddon is a loss for computer chess because 95% of the time Black will win by drawing!

Maybe it has nothing to do with correct variants and top engines are already practically unbeatable in the meaning that it is practically impossible to beat them when they do not use a book to force them to play inferior moves and even if I offer one million dollars for people who can show me one win against stockfish15 or Dragon3 with 1,000,000,000 nodes per move and single core to make it deterministic then nobody is going to win the money.

Note that I am not sure if it is the case and I do not offer money for it.

[mehmet123]

Even if Dragon won the championship with the armegeddon game it played, this still does not change that the Dragon is the champion. This is the 5th championhip of Komodo/Dragon team and one championship left to catch the Junior team.
Congratulations to Komodo Dragon team.

[Eduard]

The only point in such tournaments is the possibility of marketing software afterwards! The fans can be happy: There will be a new Dragon and maybe a Fritz 19. In two months nobody will be asking how the results came about. Shredder and Chiron are also commercial software. Only Leela is open source.

Opening books play the most important role in such tournaments. Ginkgo had the best book, followed by Dragon. The ChessBase CTG format is the best format for opening books.

I liked Leela best at the WCSC tournament. A great result on a laptop's very slow Radeon Vega graphics. In the second game against Shredder, Leela missed out on a possible win in the endgame (Stockfish rated +7 for Lc0).

At the WCCC tournament, the top 3 programs also drew with each other. At this high level, other results are hardly possible with one's own opening books. Leela tried to circumvent the general theory with Shortbook. However, Lc0 played moves that are often played in theory. Whether it's English, Catalan or the Queen's Gambit: It's very well known, and it wasn't possible to win against Ginkgo and Komodo. Also in the following speed tournament Lc0 had to realize that even with moves like 1. h3 the opponent could not be thrown out of the book. Ginkgo was in the book with Black for 27 moves!

The best game of the tournament for me was the last game between Lc0 and Dragon (Armageddon).

Unfortunately, in the following speed tournament against Dragon, Lc0 chose a different opening with the white pieces.
I explain it like this: Lc0 won 2x against Shredder, Dragon only 1x. But then Ginkgo also won twice against Shredder, which wasn't necessarily to be expected.

[CornfedForever]

The only point in such tournaments is the possibility of marketing software afterwards!

Kind of rare that I get to agree with you! Perhaps it's inevitable given the pure volume of your postings...

Frankly, a different set of openings could have been played and a different result achieved - too few games. I've always thought this was a 'paper title' in that sense. Good testing via more games/openings and on the SAME hardware is the real deal.

Besides, Armageddon is a 'coin flip'...someone gets 'gifted the title' so really it's a tie.