Viridithas 8.0.0 - released & Championship Match.
The latest version of Viridithas 8 is about 100 Elo stronger than Viridithas 7.
According to MCERL, Viridithas 8 plays with a strength of 3372 Elo.
I love such surprises :)
Viridithas is a free and open source chess engine written in the Rust language. Published under the MIT License.
The author, Cosmo Bobak, who (as I wrote in a previous post about Viridithas) is obsessed with treesearch, has successfully used his "obsession" and is passionately developing his chess engine.
According to the latest results (MCERL ongoing), the strongest chess engine written in Rust is Velvet 5.1.0 with a ranking of 3377 Elo.
Viridithas, after playing 738 games, reached 3372 Elo, only 5 Elo less.
Source: MCERL (ongoing)
To decide which of these chess engines is stronger, I decided to play a match between them.
With 10, 50 or maybe 100 games? Nothing of the sort!
In the end, I wanted to determine the undisputed Champion – the strongest chess engine written in the Rust language participating in the MCERL competition.
Viridithas faced Velvet in the Championship match consisting of 300 games. Time: the same as in the MCERL competitions.
Here are the results:
Velvet won by 1 point. Only one :)
Well... with this dynamic growth of both these engines, the next match between this worthy rivals may end in a different result.
Games to download.
To be sure, the next versions of Viridithas and Velvet will be paired by me in what I hope will be a no less interesting match.
Ok, take a look at how Viridithas can play*.
Black: Bit-Genie-9, a robust chess engine that plays thickly above 3000 Elo.
Below we see an aligned position full of opportunities for both sides, created by playing the Spanish Opening.
White's Bishop seems to be somewhat stuck on the c2 field, while the black Knight has no freedom standing on the b7 field. In this type of position, it is not uncommon for one side to prepare and realize a breakthrough by playing Pawn to the f4 field (white) or the f5 field (black).
Is it too fast? After all, white, if the exchange takes place, will expose itself to a "fork" attack by playing the black pawn on g5.
And indeed, Bit-Genie does not hesitate to attack Viridithas two figures. So was white Pawn's previous move on f4 a mistake ?
At first glance, it's easy to come to a hasty conclusion. Viridithas, at the price of Knight, opened the F line for Rook, placed Bishop on the excellent d4 field and, in prospect, enabled the activation of his other figures.
Blacks are trying to create any chance of a constructive game for themselves. White calmly gives up the "brawl" on the Queen's wing.
The punch line of Viridithas' idea from a few moves ago is starting to become apparent. The exposed position of the black King does not bode well for Bit-Genie.