What do you think of this game? is there a reason why stockfish didnt win?
http://en.lichess.org/jLpahKni
http://en.lichess.org/jLpahKni
How much time was each side given, I find it hard to believe that anything less than a top player can perform at 2 centipawn loss against a computer looking 20 moves ahead in every direction. It calculated that it would lose some of its advantage if it tried to make any move that furthered the position so it just kept making useless moves, at that point the person was able to get a repetition draw by going with it.
the time control was 20-30 mins
I gave the game in my chessbase Stockfish (2950 Elo) and the moves he suggested are not that ones which This SF played in many points of the game. The opening which stockfish played according to the opening lance was playing, is hard to win for black, but my stockfish chose another strategie . For example 8.c5! is a bader evaluation at this point but makes a whole other game which stockfish played very accurate in my testing.
However, Stockfishs Rating on the TCEC Worldchampionsship was about 3250. So 3500 is, sorry that i have to say this, Bullshit.
You can find some games from stockfish in modern defence at chessbase database and he NEVER traded the center down in the early mid game.
So i dont know where you have this game from, but this is a typical 2700 Stockfish strength in my experience because i played SF many times at this level and that gameplay is very typical for the named level.
However, Stockfishs Rating on the TCEC Worldchampionsship was about 3250. So 3500 is, sorry that i have to say this, Bullshit.
You can find some games from stockfish in modern defence at chessbase database and he NEVER traded the center down in the early mid game.
So i dont know where you have this game from, but this is a typical 2700 Stockfish strength in my experience because i played SF many times at this level and that gameplay is very typical for the named level.
#1 Stockfish didn't win because you played a perfect game; at least computer analysis at a rate of 4 sec/move thinks so.
#3 I'm working on testing a new Stockfish at http://en.dugovic.mooo.com/ including an opening book and endgame tablebase; however lichess requires it to move quickly so it might not be much of a challenge for a strong player.
#3 I'm working on testing a new Stockfish at http://en.dugovic.mooo.com/ including an opening book and endgame tablebase; however lichess requires it to move quickly so it might not be much of a challenge for a strong player.
It's not "full stockfish" until you give it infinite computational power and infinite time.
But then full Stockfish would be full Komodo and full Fritz etc. since with infinite computational power they would all play perfect.
On a side note, TCEC isn't using Elo at all so just don't try to compare those ratings.
And I agree that this game doesn't look to be played with semi decent Stockfish (aka multicore). Though still nice game from white (in hindsight the moves are all obvious, but OTB it's not THAT easy to play like this)
On a side note, TCEC isn't using Elo at all so just don't try to compare those ratings.
And I agree that this game doesn't look to be played with semi decent Stockfish (aka multicore). Though still nice game from white (in hindsight the moves are all obvious, but OTB it's not THAT easy to play like this)
This topic has been archived and can no longer be replied to.