Anand vs Topalov Game 7

[Milos]

Check how close the lines and level of play really are. Closer than you give credit.

Sure, these guys are great in memorizing lines.
Switch to chess960, and you'd see what a real machine power means.

[AdminX]

Well it appears that Ivan Cheparinov had prepared the Novelty for him that went twenty moves deep into this game.

"At the press conference the Bulgarian said that it was his second Ivan Cheparinov who had prepared the line for him."

[d]8/8/4q1kp/1Q4p1/2p3P1/2Pp4/5NK1/8 w - - 0 42

With 42. Qa4! White could have prevented the black pawn going to d2. 42... Qd5+ (42... d2? 43. Qc2+ ) 43. Kf1 Qe6 44. Qa2! Qd5 (44... Qc6 45. Qa1! Qd5 46. Qe1! ) 45. Qa6+ Kg7 46. Qa7+ Kg6 47. Qe3! +/- Shipov. {Also Stockfish 1.7.1}

Source: http://www.chessvibes.com/reports/wch-g ... more-24861

[Terry McCracken]

Check how close the lines and level of play really are. Closer than you give credit.

Sure, these guys are great in memorizing lines.
Switch to chess960, and you'd see what a real machine power means.

I don't care about chess 960 and it's irrelavent to this discussion AFAIC.
Although. you might be surprised what they could do.

Don't lecture me on computers I was there since the beginning.

And these guys are far better than just memorizing lines, How lame and disrespectful. It's pure ignorance.

[shiv]

Well it appears that Ivan Cheparinov had prepared the Novelty for him that went twenty moves deep into this game.

"At the press conference the Bulgarian said that it was his second Ivan Cheparinov who had prepared the line for him."

[d]8/8/4q1kp/1Q4p1/2p3P1/2Pp4/5NK1/8 w - - 0 42

With 42. Qa4! White could have prevented the black pawn going to d2. 42... Qd5+ (42... d2? 43. Qc2+ ) 43. Kf1 Qe6 44. Qa2! Qd5 (44... Qc6 45. Qa1! Qd5 46. Qe1! ) 45. Qa6+ Kg7 46. Qa7+ Kg6 47. Qe3! +/- Shipov. {Also Stockfish 1.7.1}

Source: http://www.chessvibes.com/reports/wch-g ... more-24861

Actually this game also exposes the weaknesses of engines quite a bit. Stockfish and other engines thought the position after Nd2 was just losing for white.

And this Qa4 position is another case in point, what if black just plays 42.. h5 after Qa4 trading the g-pawn. I tried with engines and yes you will see a +0.5 which keeps going down, but the engines are just evaluating the position, there is no win of course.. Shipov of course has caught on the engine disease. I might be proven wrong but positions like occurred in the game are confusing for both humans and engines. For the same reason, I do not trust the +0.44 after Qh3 either. Just shows how complex a game chess is.

For several other moves in the game, I turned on the engine, but found unreliable evaluations.

[Edmund]

Check how close the lines and level of play really are. Closer than you give credit.

Sure, these guys are great in memorizing lines.
Switch to chess960, and you'd see what a real machine power means.

Do you want to say that humans are better memorizing opening book lines than computers?

It is no problem for a machine to store gigabytes of chess databases, but for a human? He just prepares certain lines and is cleverer in the choice of move order not to leave the opening book too early. This in combination with broad pattern recognition makes him appear to have a great opening book repertoar, but you could never argue that he has a competitive advantage over the machine.

[M ANSARI]

Comparing chess play of humans against today's engines on fast hardware is like comparing an olympic sprinter against a fast car ... it is simply not even worth talking about as the comparison is no longer viable. Human chess play at the top level is obviously heavily steered by engine analysis ... doing otherwise would lead to catastrophic defeat. Human analysis using engines as a tool is still better than an engine alone ... but even that might not be the case for very long.

[JuLieN]

Yes, I remember that SF gave 42. Qa4 a mark above +2 for white. White could have won this game. The thing is that such moves are nearly unreachable for humans, as their consequences are not obvious at all.

Anyway, even if my previous joke gave birth to a debate that shouldn't be (as it was JUST a joke) here is my point of view :

- YES, the engines now obviously play much better than any human being, would he be a world champion.
- BUT, I think it's still a lot of fun to watch human matches, and engines help us to have fun : will Anand/Topalov see the good move? Wow, incredible he saw it, this guy is an alien! Having strong engines to help us analyzing human matches realtime gives us a lot of fun I think. I, personally, would maybe play 50% of the moves an engine suggest, while a GM would probably be over 95% : engines are a good way to measure a player's strength.
- ANYWAY, I think that, from time to time, a human player can find a better move than the one suggested by the engine. For instance, during this match, I noticed that Anand/Topalov played a different move than the one SF suggested, and as the result, at the next move, SF's evaluation for the opponent decreased (!), which shows that the engine didn't put enough "thinking" into this move, despite it was better than the one it chose. Engines are not that almighty yet it seems. Problem is : which percentage of moves differing from the engine's suggestion are improvements? It seems like it's under 50%, or white would have won in the 7th game, for instance.

[zamar]

Engines beat humans in direct matches because they can avoid blunders. Top players are often able to pick better and more logical moves - just to throw everything away with one single error.

[Uri Blass]

Well it appears that Ivan Cheparinov had prepared the Novelty for him that went twenty moves deep into this game.

"At the press conference the Bulgarian said that it was his second Ivan Cheparinov who had prepared the line for him."

[d]8/8/4q1kp/1Q4p1/2p3P1/2Pp4/5NK1/8 w - - 0 42

With 42. Qa4! White could have prevented the black pawn going to d2. 42... Qd5+ (42... d2? 43. Qc2+ ) 43. Kf1 Qe6 44. Qa2! Qd5 (44... Qc6 45. Qa1! Qd5 46. Qe1! ) 45. Qa6+ Kg7 46. Qa7+ Kg6 47. Qe3! +/- Shipov. {Also Stockfish 1.7.1}

Source: http://www.chessvibes.com/reports/wch-g ... more-24861

Actually this game also exposes the weaknesses of engines quite a bit. Stockfish and other engines thought the position after Nd2 was just losing for white.

And this Qa4 position is another case in point, what if black just plays 42.. h5 after Qa4 trading the g-pawn. I tried with engines and yes you will see a +0.5 which keeps going down, but the engines are just evaluating the position, there is no win of course.. Shipov of course has caught on the engine disease. I might be proven wrong but positions like occurred in the game are confusing for both humans and engines. For the same reason, I do not trust the +0.44 after Qh3 either. Just shows how complex a game chess is.

For several other moves in the game, I turned on the engine, but found unreliable evaluations.

I do not see +0.5 after Qa4 h5 gxh5+ Kxh5 Qa7

Of course score of more than +3 is no proof that white wins and I did not analyze enough to be sure that Qa4 wins but you can be sure that nobody is going to claim that white can win based on +0.5

I can add that I did not see evaluation by engines that suggested that white is losing after Nd2.

[Milos]

Do you want to say that humans are better memorizing opening book lines than computers?

It is no problem for a machine to store gigabytes of chess databases, but for a human? He just prepares certain lines and is cleverer in the choice of move order not to leave the opening book too early. This in combination with broad pattern recognition makes him appear to have a great opening book repertoar, but you could never argue that he has a competitive advantage over the machine.

Nope. I want to say, humans are much better in forcing particular memorized lines (engines are really lousy in that). And yes they are really good in pattern matching. But as you know pattern matching is deja vu, a sort of permanent cache. And when they run out of cache, humans quickly fall apart against machine...