A Newbie drew Stockfish at highest level (2 draws so far)

Sort:
p8q

For those who think a threefold repetition draw is suspicious coming from stockfish I already explained in the game comments why I think black is lost making any other move. But maybe that's not easy to see, so I'll show it here.

I'm analyzing that position on my computer and suggested moves are exactly the same as the game, a threefold repetition. Anybody can check it out at home with their computers.

In fact, I invite anybody to check the whole match at home with their computers, they will see that my moves are different from those suggested by stockfish, because they came out of my mind.

But if it's not clear why any other move is a lost for black, I'll post why do I think it did it. I'll suggest another possible continuation for black, but this continuation comes out of my mind, because stockfish is not suggesting it.

On move 44 before the repetition we have this position. I post several posibilities, any combination I think is always a win for white. So stockfish chooses to get a draw by threefold repetition.

 

 

Asparagusic_acids
p8q wrote:

For those who think a threefold repetition draw is suspicious coming from stockfish I already explained in the game comments why I think black is lost making any other move. But maybe that's not easy to see, so I'll show it here.

I'm analyzing that position on my computer and suggested moves are exactly the same as the game, a threefold repetition. Anybody can check it out at home with their computers.

In fact, I invite anybody to check the whole match at home with their computers, they will see I'm moving different moves suggested by stockfish, because they are came out of my mind.

But if it's not clear why any other move is a lost for black, I'll post why do I think it did it. I'll suggest another possible continuation for black, but this continuation comes out of my mind, because stockfish is not suggesting it.

On move 44 before the repetition we have this position. I post several posibilities, any combination I think is always a win for white. So stockfish chooses to get a draw by threefold repetition.

 

 

What was the contempt 

Pulpofeira

I would have placed my bets on David, as he was wielding a long range, deadly weapon against a massive target.

ponz111

P8q 

Stockfish or any other good chess engine can analyze 30 moves deep and it does not take a year. Maybe half an hour.  Also When you say you can "see" 30 moves deep in 10 minutes--you are fooling yourself if you think you can see 30 good moves deep in 10 minutes,  However if you refer to the Ruy Lopez you might know one opening variation 30 moves deep from memory.

p8q
Pulpofeira wrote:

I would have placed my bets on David, as he was wielding a long range, deadly weapon against a massive target.

hahhaha, that's right, that's a good one. Sometimes I came to the same thought.

p8q
ponz111 wrote:

P8q 

Stockfish or any other good chess engine can analyze 30 moves deep and it does not take a year. Maybe half an hour.  Also When you say you can "see" 30 moves deep in 10 minutes--you are fooling yourself if you think you can see 30 good moves deep in 10 minutes,  However if you refer to the Ruy Lopez you might know one opening variation 30 moves deep from memory.

I didn't ever try to set stockfish to 30 moves deep, so I thought it couldn't. That's interesting, I'll try it to see what happens.

I know that when I see those more than 30 moves deep I'm not visualizing strong moves. Even though they are weak, they are 30 moves deep. Then I moved them physically to check if what I imagined is true and I looked at my watch and 10 minutes passed. So, I'm not fooling myself.

I never memorize any variation, I don't believe in memorization in chess. I just imagine those moves.

Usually when playing vs stockfish I try to see around 25 moves deep. But most of the time in move 10 I already lose in that line, so I don't keep thinking that line. And in the variations I go that deep stockfish only moves 5 of those moves, but sometimes up to 10 moves. In this draw 10 moves that I visualized were actually played (from the move 19 to 24).

ponz111

Oh you are counting your moves and the moves of your opponent--sorry I did not realize this. But when I say stock fish can analyze 30 moves deep I am fairly sure that means like from move 19 to move 39!?  But, again, I may be wrong on this? 

p8q
ponz111 wrote:

Oh you are counting your moves and the moves of your opponent--sorry I did not realize this. But when I say stock fish can analyze 30 moves deep I am fairly sure that means like from move 19 to move 39!?  But, again, I may be wrong on this? 

Sorry, that's right. When I said move I meant "ply" or "half-move".

Sometimes I have visualized 45 or 50. But that doesn't mean to be a good chess player or anything. Any person with proper training can do it. What makes a good chess player being good, is to be able to visualize 5 very good moves instead of those 40 weak ones tongue.png

I don't think a machine is able to calculate a game all the way from the opening to the end of the game, because there are so many billions of posibilities....  more than atoms in the visible universe they say. A human can do it though, but it could be a weak moves match like I visualized.

Human intuition is a mystery, it can calculate all those posibilities faster than a computer. A computer needs to analyze millions of positions to come out with a move. And a human can make the same move using intuition much faster than the computer, and that person even don't realize how that simple move came to his mind.

m_connors

Well, I only have two words for you - Good Luck. I have a hard enough time around level 5 or 6 (about 1700 to 1900), never mind level 10!!

winston_weng
p8q wrote:
ponz111 wrote:

P8q 

Stockfish or any other good chess engine can analyze 30 moves deep and it does not take a year. Maybe half an hour.  Also When you say you can "see" 30 moves deep in 10 minutes--you are fooling yourself if you think you can see 30 good moves deep in 10 minutes,  However if you refer to the Ruy Lopez you might know one opening variation 30 moves deep from memory.

I didn't ever try to set stockfish to 30 moves deep, so I thought it couldn't. That's interesting, I'll try it to see what happens.

I know that when I see those more than 30 moves deep I'm not visualizing strong moves. Even though they are weak, they are 30 moves deep. Then I moved them physically to check if what I imagined is true and I looked at my watch and 10 minutes passed. So, I'm not fooling myself.

I never memorize any variation, I don't believe in memorization in chess. I just imagine those moves.

Usually when playing vs stockfish I try to see around 25 moves deep. But most of the time in move 10 I already lose in that line, so I don't keep thinking that line. And in the variations I go that deep stockfish only moves 5 of those moves, but sometimes up to 10 moves. In this draw 10 moves that I visualized were actually played (from the move 19 to 24).

Stockfish can easily calculate a variation of 30 moves, you are comparing yourself calculating a single variation 30 moves deep to stockfish calculating every single variation, all of them to 30 moves deep.

llama44
winston_weng wrote:
 

stockfish is calculating every single variation, all of them to 30 moves deep.

Well, that's mathematically impossible, but you have the right idea. It's calculating many lines and very deep.

llama44

Also, in >99% of positions, it's pointless for humans to calculate that deeply. Our strength (and our limitation) is stopping when most forcing moves are exhausted and rendering a mostly-correct evaluation.

For example I remember the first game entered a double rook endgame that ziryab said was lost. Maybe he's right, but I think it was still complicated and I'm not 100% sure (again, he may be right).

But the king and pawn endgame was lost enough for even many lower rated players to understand. The OP offered the rook exchange to enter this endgame. This isn't a matter of calculation, it's understanding. It allows us to shortcut 50 or 100 moves of calculation.

In reality, humans, on average during a game, probably calculate less than 1 move per second. Engines calculate millions. The fact that they're relatively slightly better than us says amazing things about our brains tongue.png

But yeah, the point is that understanding can shortcut massive calculation, and in some positions calculation is not useful (or necessary) for humans.

drmrboss
p8q wrote:

. It looks like the perfect game. I made an Elo performance of 3228.

aa-ron1235
p8q wrote:
ponz111 wrote:

Oh you are counting your moves and the moves of your opponent--sorry I did not realize this. But when I say stock fish can analyze 30 moves deep I am fairly sure that means like from move 19 to move 39!?  But, again, I may be wrong on this? 

Sorry, that's right. When I said move I meant "ply" or "half-move".

Sometimes I have visualized 45 or 50. But that doesn't mean to be a good chess player or anything. Any person with proper training can do it. What makes a good chess player being good, is to be able to visualize 5 very good moves instead of those 40 weak ones 

I don't think a machine is able to calculate a game all the way from the opening to the end of the game, because there are so many billions of posibilities....  more than atoms in the visible universe they say. A human can do it though, but it could be a weak moves match like I visualized.

Human intuition is a mystery, it can calculate all those posibilities faster than a computer. A computer needs to analyze millions of positions to come out with a move. And a human can make the same move using intuition much faster than the computer, and that person even don't realize how that simple move came to his mind.

No, the human mind cannot calculate more than a computer. To prove this, you can simply look at processing speed.

"Neurons, for example, are the brain’s building blocks and can only fire about 200 times per second, or 200 hertz. Computer processors are measured in gigahertz: billions of cycles per second."- from https://medium.com/thrive-global/the-human-brain-vs-computers-5880cb156541.

Saying you can calculate 30 moves deep in a single line is absolutely meaningless. Anyone can do that, and a computer can do that before your eye can blink. The reason you did so well in this line is that you played the safest, most boring, most draw oriented variation in chess. The computer sees it as a draw so it plays it as one. If the computer had played more aggressive moves your game would have fallen apart quickly. The reason super grandmasters can't beat computers isn't that they are not trying hard enough, it is that they cannot. 

I am not saying computers are more versatile than humans at chess (illegal moves are illegal, badly places pieces can be misread). They are, however, much, much better than humans at the game of chess. You know Deep Blue, 20 years ago? That computer beat the worlds best player, and probably one of the best players of all time. The fact that YOU USED A COMPUTER TO ANALYZE YOUR GAME shows how stupid your view is. 

I do not know why I am arguing with a fool over chess.com over something that is widely agreed on when his entire proof is a boring game that the computer decided to play like a draw. Good luck with convincing everyone that you are a calculating genius when you take 1o minutes to do what my phone could do in less than a second. Good luck with convincing everyone that computers have suddenly gotten massively worse over the last 20 years. Good luck with the view, that you, a 1500, has come up with a wild way to beat computers that no one has ever thought of before. Calculation. 

drmrboss
p8q wrote:
Sometimes I have visualized 45 or 50.

 

Prometheus_Fuschs

I want to see a rematch of p8q vs Stockfish 11 using contempt 100, standard time controls and a decent laptop to run SF (less than 1 Mnps is pretty slow), see if you can draw that.

p8q
aa-ron1235 wrote:
p8q wrote:
ponz111 wrote:

Oh you are counting your moves and the moves of your opponent--sorry I did not realize this. But when I say stock fish can analyze 30 moves deep I am fairly sure that means like from move 19 to move 39!?  But, again, I may be wrong on this? 

Sorry, that's right. When I said move I meant "ply" or "half-move".

Sometimes I have visualized 45 or 50. But that doesn't mean to be a good chess player or anything. Any person with proper training can do it. What makes a good chess player being good, is to be able to visualize 5 very good moves instead of those 40 weak ones 

I don't think a machine is able to calculate a game all the way from the opening to the end of the game, because there are so many billions of posibilities....  more than atoms in the visible universe they say. A human can do it though, but it could be a weak moves match like I visualized.

Human intuition is a mystery, it can calculate all those posibilities faster than a computer. A computer needs to analyze millions of positions to come out with a move. And a human can make the same move using intuition much faster than the computer, and that person even don't realize how that simple move came to his mind.

No, the human mind cannot calculate more than a computer. To prove this, you can simply look at processing speed.

"Neurons, for example, are the brain’s building blocks and can only fire about 200 times per second, or 200 hertz. Computer processors are measured in gigahertz: billions of cycles per second."- from https://medium.com/thrive-global/the-human-brain-vs-computers-5880cb156541.

Saying you can calculate 30 moves deep in a single line is absolutely meaningless. Anyone can do that, and a computer can do that before your eye can blink. The reason you did so well in this line is that you played the safest, most boring, most draw oriented variation in chess. The computer sees it as a draw so it plays it as one. If the computer had played more aggressive moves your game would have fallen apart quickly. The reason super grandmasters can't beat computers isn't that they are not trying hard enough, it is that they cannot. 

I am not saying computers are more versatile than humans at chess (illegal moves are illegal, badly places pieces can be misread). They are, however, much, much better than humans at the game of chess. You know Deep Blue, 20 years ago? That computer beat the worlds best player, and probably one of the best players of all time. The fact that YOU USED A COMPUTER TO ANALYZE YOUR GAME shows how stupid your view is. 

I do not know why I am arguing with a fool over chess.com over something that is widely agreed on when his entire proof is a boring game that the computer decided to play like a draw. Good luck with convincing everyone that you are a calculating genius when you take 1o minutes to do what my phone could do in less than a second. Good luck with convincing everyone that computers have suddenly gotten massively worse over the last 20 years. Good luck with the view, that you, a 1500, has come up with a wild way to beat computers that no one has ever thought of before. Calculation. 

wow wow... calm down la little bit. Why are you so upset? Usually people gets upset when they are shown what they don't want to see. What is it what you don't want to see? No need to call me a fool, etc.

You provided some interesting points though, I appreciate that. I'll discuss them a little bit:

1) "No, the human mind cannot calculate more than a computer. To prove this, you can simply look at processing speed."

Yes, it can. Intuition does. Intuition calculates millions of moves in less than a millionth of a second, much faster than a computer. Our problem is that our intuition doesn't always come out with the best move (as I said before speed and quality are two different things). So, in order to correct the moves our intuition suggests, we have to calculate with our slower neurons in order to come out with that good move.

It is a good point to compare the clock speed of a computer (frequency) with the speed of neuron electric pulses. But, who said our intuition is in our neurons? Who said our thoughts, ideas, etc. entirely come from our neurons? Maybe only 80% of our ideas and thoughts come from neurons and the rest comes from our spirit. Neurons is what science can measure, touch, see... but our brain is composed of something more than just neurons, or material. We might be thinking also with our spirit, which could be faster than the speed of light. Science can't explain everything (for now).

2) "Saying you can calculate 30 moves deep in a single line is absolutely meaningless. Anyone can do that, and a computer can do that before your eye can blink."

Of course, it's meaningless. I already said that before. To think ahead 5 very good moves are much better than my 45 weak moves. And do you know how many millions and millions of meaningless and stupid moves a computer thinks before coming out with the best move? For every single good move that a computer thinks, it has to calculate first millions and millions of meaningless moves. Because they don't have intuition, which makes them inferior to human brains.

3) "The reason you did so well in this line is that you played the safest, most boring, most draw oriented variation in chess. The computer sees it as a draw so it plays it as one. If the computer had played more aggressive moves your game would have fallen apart quickly. The reason super grandmasters can't beat computers isn't that they are not trying hard enough, it is that they cannot."

Yes, all the time I'm trying to simplify the position, exchange pieces as fast as I can as long as those exchanges don't deteriorate my position, in order to cut down calculation and bring the machine to wage war into my human terrain instead of playing in machine pure calculation terrain. Being the one to choose the terrain to fight is one of Sun Tzu strategy principles. I'm not going to try to beat stockfish in its strong points, I'll find its weak point and exploit it. And its strong point is tactics, calculation.

This way, I'm not trying to be creative, or setting traps or baits, like we do when playing a human. I know that makes the match less thrilling, more drawish, more dull.... It's part of the strategy when playing a computer. Chess is a game of strategy, so I'm using strategy.

On the other hand, usually all creative moves that came to my mind, when calculating all possibilities, always in move 10 or 15 I could see it was a losing line. I would dare to play them against human, not machine.

Why don't you like draws? I think the most beautiful and perfect match is a draw. In the world championship 2018 Magnus Carlsen vs Fabiano Caruana all match games were drawn. All of them! Only the tie match was different. I think that's beautiful. Even though I'm trying to get a win against the machine I like draws.

4) "You know Deep Blue, 20 years ago? That computer beat the worlds best player"

That's not true. You should watch the documentary about it. There were human moves behind the scenes, and after the match Kasparov asked for those papers that contained the computer thinking lines specially in those where the machine made the most human moves. IBM didn't want to provide those documents. Why?! If they have nothing to hide, why not to provide them? because they cheated, there was a human behind the scenes suggesting moves, in order to make people think their company is so good that creates a computer that beats Kasparov, this way selling their stuff and profiting.

5) "The fact that YOU USED A COMPUTER TO ANALYZE YOUR GAME shows how stupid your view is."

When I analyzed why stockfish made a draw by threefold repetition, the second pgn I posted, I didn't use a computer to analyze the position.

Another reason I used a computer to analyze the main game is because I'm too lazy to go writing down all possible lines move by move. It's not stupid to use a machine slave to do dull work.

6) "I do not know why I am arguing with a fool over chess.com over something that is widely agreed on when his entire proof is a boring game that the computer decided to play like a draw."

Thank you for your time when writing your post. I like and need different points of view.

We should never think like the big mass of people. We must never trust "widely agreed" matters. That would make our thoughts less human and more machine like. In chess itself you learn that rules have always exceptions, that rules are made to be broken.

I'm sorry you found this game being boring. I know it is much more funny "the immortal game" played by Anderssen and Kieseritzky back in 1851. But in order to produce such a funny game you have to make a big blunder in opening or middle game. And me or stockfish were not willing to make such a blunder just to make the game funny. We both wanted to win. I'm sorry that me and machine disappointed you with a dull draw.

7) "Good luck with convincing everyone that you are a calculating genius when you take 1o minutes to do what my phone could do in less than a second."

I didn't say I'm a calculating genius: because I said my capacity of visualization was full of weak moves. I also said any person can do it with proper training. And I said real geniuses are GMs that can play simultaneous blindfolded with strong moves. That's really impressive.

8) "Good luck with the view, that you, a 1500, has come up with a wild way to beat computers that no one has ever thought of before. Calculation."

You actually don't realize the core of what I'm trying to convince you.

I'm trying to convince you of the anomaly. The exception in the rule. The broken rule... call the way you want.

The weakest can always beat the strongest. The will to do it and believing in it is what makes all the difference. GM's can beat the machine any time of the day, but they think they can not, and that's what makes it impossible.

And it's a shame that a newbie, a 1500, has to come here to show it is possible.

And also I'm trying to convince you that machines are not better than humans in chess or thinking, or anything else. They are just faster, but not better. A draw, that wasn't funny enough for you, is a way to prove it. Maybe next time It will please you when I post a won match. Or maybe it will not please you if we don't make funny moves or pieces jumping of the board.

Well, right now I'm playing a game against stockfish in which I'm making a pawn storm. That's funny enough for me grin.png

p8q
NoSchoolpls wrote:

When you watch too much artficial intelligence sci-fi 😆

hahah, good one. I didn't watch so many, I'm not a movie person. Just watched the basics: the matrix, and a few more.

p8q
drmrboss wrote:
p8q wrote:
Sometimes I have visualized 45 or 50.

 

 

hahahaha, I like your gifs, so funny happy.png

llama44

Isn't it nice though? To set a goal where you're guaranteed to fail?

In other words you risk nothing while pretending to be brave.

Why not set a goal of getting to 1400 blitz? Because that's not only possible to achieve, but also possible to fail? Heh.

"I'm trying to beat stockfish!"

That's nice kid. No one cares, and that's the truth.

This forum topic has been locked