I just drew against Komodo25 Maximum 3200 Computer
Sort:
Yo
Sorry it wasn't working i fixed it
Dec 20, 2021
0
#4
Congrats on getting a draw against Komodo25. For convenience, here is a link to your game at the game archive: https://www.chess.com/analysis/game/computer/2530597 . Komodo is a very strong commercial chess engine, about number 3 behind Stockfish and the Leela Chess Zero neural network engine. Komodo25 or "Maximum" is stronger than the Vishy, Fabiano or other bots with people's names. However Komodo25 is substantially weaker than the regular (non browser-based) Stockfish engine running inside Fritz or Arena. I've been playing against Komodo25 and having Stockfish 14.1 (inside Shredder) play against Komodo25. Komodo25 doesn't come close to winning or drawing against regular (non-browser) Stockfish.
In your game, 24. Qxb7 is a surprisingly egregious blunder coming from Komodo25. I often see mistakes of the type "this move allows mate in 16", but few like Qxb7 . I generally see accuracies for Komodo25 around the mid 80s.
Thanks for your explanation and for the link. I guess then i have Stockfish 14.1 to look forward to, that's very exciting. I don't know if it's true but i heard that probably the strongest Stockfish beat Magnus Carlsen so that's scary haha. And i am currently trying to beat Lichess' Fairy-Stockfish 14 level 8 which sounds like the one you mentioned or could be close. I will also keep playing to beat Komodo25, also will try chess24's toughest computer, of course Magnus' toughest computer and others. Good luck.
I beat the Komodo25 (3200), checkmate in 28 moves
Komodo made two big mistakes, it surprises me
You can check it here if you want !
https://www.chess.com/analysis/library/42B62iGqS6
Feb 24, 2022
0
#7
The average difference between Komodo25's moves and the best moves was 2.96, and the percentage of best moves was 29.6% . It makes me wonder how it would perform in rated rapid games against grrandmasters...
@JosserandT Congrats, ha ha nice. Yea Komodo in that game was messing with your head. @bernier96 Yea it might play fairly well against grandmasters and depending with how it does on speed. Although that's right the ratings from Stockfish are more adequate for dealing with them. I haven't been very active, but thank you everyone for stopping by and for saying hi and sharing your thoughts.
JosserandT wrote:
I beat the Komodo25 (3200), checkmate in 28 moves
Komodo made two big mistakes, it surprises me
You can check it here if you want !
https://www.chess.com/analysis/library/42B62iGqS6
How is this real? The engine is playing worse than a 1st grader who just learned how to play. I don't think you're lying but something is off here. Can anyone explain?
Please do not post full games in puzzle format.
Sep 14, 2023
0
#12
If you are looking against Fairy Stockfish Level 8 on lichess then check out this game
https://lichess.org/KzH5oSCh/black#0
Sep 14, 2023
0
#13
This was a practice game of my opening with Stockfish Level 8 and I won it. (please do not go on my special preparation and this was a practice game for greater understanding of my opening in a tournament)
Sep 14, 2023
0
#16
JJRSChess wrote:
Play Real People. Don’t play Bots.
I get how playing real people increases your chess skills more, but playing bots is somewhat helpful and you can have fun. It’s also somewhat of a challenge.
I won Maximum (Kombodo25) 3200.
https://www.chess.com/analysis/game/computer/11779311?tab=review
I thought it was normal, but saw that it says it is impossible. Any ideas and comments?
Dec 26, 2023
0
#19
I'm decently sure every person got a lucky glitch. The OP's game against Komodo had Komodo playing incredibly stupid moves, which a computer would never actually play. There might be something wrong with Komodo, I don't know. Radhey's games were almost certainly cheated, with an average centipawn loss of 8-10. For context, SGM's average about 10-20 centipawn loss, which means that this game is certainly not within the bounds of human ability.
Dec 26, 2023
0
#20
ChessDude009 wrote:
I'm decently sure every person got a lucky glitch. The OP's game against Komodo had Komodo playing incredibly stupid moves, which a computer would never actually play. There might be something wrong with Komodo, I don't know. Radhey's games were almost certainly cheated, with an average centipawn loss of 8-10. For context, SGM's average about 10-20 centipawn loss, which means that this game is certainly not within the bounds of human ability.
Do you think I cheat ? This is accuracy by luck, These were 4 games out of almost 900 games
Is Komodo25 supposed to be another version of Stockfish or is Stockfish much stronger. Because even i could tell when i was playing that it made quite a few moves that disappointed me with an accuracy of 75.1 to my 68.2. Especially Qxb7 was an absolute win for me except that for being a rookie i didn't want to trade my queen right after. But of course regret it the next second lol. It took me around an hour ½ to a little over two at the most i think. I Also feel like i might have beat it if i took QxRb1 when i had the chance. Komodo25 is white and i'm black.