Since you’re here...

We hope you will consider supporting us today. We need your support to continue to exist, because good entries are more and more work time. Every reader contribution, however big or small, is so valuable. Support "Chess Engines Diary" even a small amount– and it only takes a minute. Thank you.
============================== My email: jotes@go2.pl



Why we withdrew from WCCC (LcZero)

on
 


Why we withdrew from WCCC (LcZero)
The WCCC organizers allowed a clone, ShashChess, to participate as a proxy for Stockfish (that didn’t enter), which we find disrespectful. Consequently, we are withdrawing from the competition.

Our concerns in more detail 
When it was announced that this year’s WCCC would be the final one, we were eager to take part and started making preparations. As part of that, we encouraged the Stockfish core developers to join as well, and although they decided not to compete, we completely understand and support their choice.

When the participant list came out, we were surprised to see ShashChess included. Given the strict originality requirements for the WCCC, which have sparked significant controversies in the past, this was very unexpected.

ShashChess appears to follow a common pattern where individuals make arbitrary modifications to Stockfish and claim non-existent improvements, heavily exaggerating the value of their contributions. This misleads others into giving credit for any achievements to the wrong people.

Here are our main concerns:

Superficial modifications: Our analysis of ShashChess shows that its most significant changes in the chess playing code are modifications of work originating from Crystal, another Stockfish derivative. These changes are minor and do not represent any meaningful advancement in chess engine development.
Optional features: New features highlighted in ShashChess are optional and most likely hurt the performance, and as such cannot be guaranteed to be used during the competition. This leads to the concerning possibility that, by optimizing for their strongest performance, ShashChess could end up being configured to be essentially Stockfish, without the organizers—or the community—being able to detect it.
Similarity analysis: Simex, a similarity analysis tool originally developed by the Komodo authors, shows that ShashChess’s evaluation function is virtually indistinguishable from Stockfish’s. Even minor releases of Stockfish exhibit much more significant differences in Simex.

more: https://lczero.org/blog/2024/10/why-we-withdrew-from-wccc/

Comments

Post a Comment