Chess will never be solved, here's why

Sort:
Avatar of tygxc

@6163

"Even if a computer would solve chess, which I doubt" ++ A matter of money.

"because the possible "reasobable" positions is estimated to be around 10^120"
++ No, 10^44 legal positions of which 10^17 relevant.

"humanly impossible to memorize"
++ It may be impossible to memorize 10^17 positions or 10^15 games, but memorizing 10,000 games or about a million positions in a few months is possible.

"possible themes and strategies in every perfect game the engine provides"
++ Yes, maybe Chess can be solved by a set of e.g. 1000 rules, like Connect Four was weakly solved with 9 knowledge rules.

Avatar of Elroch

No-one can memorise a million positions in a few months. That is a position every few seconds all day long 7 days a week.

A million positions in a lifetime by an exceptionally talented person? Maybe.

A million is a tiny number to chess.

Regarding the rest of what you say, I'll just observe that repeating your errors after they have been pointed out is pathological behaviour, not reasoning.

Avatar of MARattigan
tygxc wrote:

++ A successful direct attack on the king can only succeed if one player neglects his defence, i.e. does not play optimally. ...

So where did Black neglect his defence in the game I posted here Mister 2046?

Back to the validity of your basic calculation, I invited you here to "Show it or shut it". You have done neither so far. Do you plan to?

Avatar of Optimissed
Elroch wrote:

No-one can memorise a million positions in a few months. That is a position every few seconds all day long 7 days a week.

A million positions in a lifetime by an exceptionally talented person? Maybe.

A million is a tiny number to chess.

Regarding the rest of what you say, I'll just observe that repeating your errors after they have been pointed out is pathological behaviour, not reasoning.

A million in a lifetime? It's a big number. 20,000 a year for 50 years. 60 a day. Maybe someone with a prodigious memory could. Unlikely ever to happen in practice.

Avatar of Optimissed
PDX_Axe wrote:

I'm more than 3 times old enough.  Here's to being 21, 44 years ago lol.

 

Next year I'll be four times old enough. 72.

Avatar of MARattigan
Optimissed wrote:
PDX_Axe wrote:

I'm more than 3 times old enough.  Here's to being 21, 44 years ago lol.

 

Next year I'll be four times old enough. 72.

You haven't told us where your moles are yet.

Avatar of Elroch
Optimissed wrote:
Elroch wrote:

No-one can memorise a million positions in a few months. That is a position every few seconds all day long 7 days a week.

A million positions in a lifetime by an exceptionally talented person? Maybe.

A million is a tiny number to chess.

Regarding the rest of what you say, I'll just observe that repeating your errors after they have been pointed out is pathological behaviour, not reasoning.

A million in a lifetime? It's a big number. 20,000 a year for 50 years. 60 a day. Maybe someone with a prodigious memory could. Unlikely ever to happen in practice.

The number is based on a specific example. Magnus Carlsen has been said to be able to recall about 10,000 games that he has studied. He is extremely impressive when asked to recognise positions from historical games.

Avatar of Optimissed
Elroch wrote:
Optimissed wrote:
Elroch wrote:

No-one can memorise a million positions in a few months. That is a position every few seconds all day long 7 days a week.

A million positions in a lifetime by an exceptionally talented person? Maybe.

A million is a tiny number to chess.

Regarding the rest of what you say, I'll just observe that repeating your errors after they have been pointed out is pathological behaviour, not reasoning.

A million in a lifetime? It's a big number. 20,000 a year for 50 years. 60 a day. Maybe someone with a prodigious memory could. Unlikely ever to happen in practice.

The number is based on a specific example. Magnus Carlsen has been said to be able to recall about 10,000 games that he has studied. He is extremely impressive when asked to recognise positions from historical games.


Yes 10,000 is sort of normal - ish for people with good memories. I was thinking that a million isn't going to be likely because of the large numbers of similar games.

I used to be a book runner as a sort of part time occupation. I could go into a second hand bookshop and memorise much of what the stock was selling for in about an hour, which helped when I was buying books from one shop to sell to another shop. At first I used to try to do a deal with the bookshop owner. I'd sell him stock at a good price in return for him or her answering my questions about local pricing. I suppose though that I only needed a specific memory for a number of items in the low four figures or maybe only around 1000. I no longer have that kind of memory but when I was nine I could remember large numbers as a visual image. Then I had a genuine photographic memory. It got me into a lot of trouble before I learned to stop correcting people's faulty memories of what they'd said or done. But even so, a  million positions ... you wouldn't be a normal person. I think it might show as a disability in other areas.

Avatar of Optimissed

Average humans are supposed to have the capacity to remember the names and characteristics of about 150 people. That means that the optimum size of a local tribe is 150. For sheep, the number is somewhat less. Off the top of my head, my memory is telling me that sheep can remember the faces and characteristics of 60 to 80 other sheep. This would have been the result of studies.

Avatar of Optimissed

<<<Regarding the rest of what you say, I'll just observe that repeating your errors after they have been pointed out is pathological behaviour, not reasoning.>>>

In particular the insistence on 5 years and money being the only barrier seems to be an obsession or perhaps a compulsion to continue to insist on stating the ridiculous. It may b a result of the somewhat unreal atmosphere of an internet site where many people are playing a part and where people might almost be excused for thinking that everyone is fake or unreal. Someone doing the same in real life would definitely be thought of as acting thus due to insanity but on the internet, maybe it can be put down to a failure to adjust realistically. Instead someone adjusts to the opposition of others by reacting as if the objections of others are unreal.

Avatar of tygxc

@6256
"So where did Black neglect his defence in the game I posted?"
++That is easy: 31...Kf7? hangs a piece.

Avatar of tygxc

@6175
Here too 17 Qg4? neglects the defense. Necessary was 17 Bc1 to defend pawn f4.
On g4 the queen will later have to flee before 23...Rdg8 and 24...Bc8.
That fuels the black attack against the undefended white king.
The point is that a direct attack against the king cannot succeed against optimal play.
Queening a pawn is a more feasible aim.

Bogolyubov was not in the same league as Alekhine, Capablanca, Nimzovich, or Rubinstein.
Take for example the Carlsbad 1929 tournament just before the match.

  1. Nimzovich
  2. Capablanca
  3. Spielmann
  4. Rubinstein
  5. Becker
  6. Vidmar
  7. Euwe
  8. Bogolyubov

'After such a result Alekhine will have to give me odds and thus I will be World Champion' - Bogolyubov

Avatar of tygxc

@6255

"No-one can memorise a million positions"
++ All humans speak at least one language.
A language contains about a million words.
Why could a human not memorise 10,000 games i.e. about a million positions?
Standard match preparation is to study all games of the opponent.

Avatar of DiogenesDue
tygxc wrote:

@6255

"No-one can memorise a million positions"
++ All humans speak at least one language.
A language contains about a million words.
Why could a human not memorise 10,000 games i.e. about a million positions?
Standard match preparation is to study all games of the opponent.

As you often do, you are disproving your own point by misapplying a premise.

For English, there are about half a million words in the biggest dictionaries. of which less than 200,000 are actually in common or even uncommon usage.  The average individual only uses about 30,000 words.

Avatar of tygxc

@6267

"For English, there are about half a million words in the biggest dictionaries."
"The average individual only uses about 30,000 words."
++ I did not say that an average individual would be able to memorise 10,000 games i.e. about a million positions.  Several people have a proficient conduct of several languages.

Psakhis had memorised all Fischer Games.
Fischer had memorised all Spassky games and much more.
Dorfman had memorised all 210 games of Zürich 1953 and could recite for each diagram: the round, the players, the result, and the game plans.
It is plausible that Carlsen had for his matches memorised all Anand, Karjakin, Caruana, and Nepomniachtchi games and much more.

Avatar of MARattigan
tygxc wrote:

@6256
"So where did Black neglect his defence in the game I posted?"
++That is easy: 31...Kf7? hangs a piece.

Yes. But the move was played in an ICCF World Championship game (https://www.iccf.com/game?id=1164303) so it must be about perfect.

You didn't respond to the second part of my post. You must have missed it. I'll repeat it.

Back to the validity of your basic calculation, I invited you here to "Show it or shut it". You have done neither so far. Do you plan to?

Avatar of tygxc

@6269
"Yes. But the move was played in an ICCF World Championship game"
++ The drawn games are > 99% sure to be perfect with optimal play from both sides.
The decisive games contain at least 1 error.
Here it is a typical human error. They look like 20 moves deep and then mess up as they erroneously believe a certain move of that sequence already has been played, while it has not yet.

Avatar of MARattigan

@tygxc

"Here it is a typical human error. They look like 20 moves deep and then mess up as they erroneously believe a certain move of that sequence already has been played, while it has not yet."

I've been concentrating on getting my grandson, who's just learned the moves, to stop hanging his pieces. Maybe his problem is also looking 20 moves deep and thinking something's already been played that hasn't.

And the second point in my post? You've missed it again:

Back to the validity of your basic calculation, I invited you here to "Show it or shut it". You have done neither so far. Do you plan to?

No point in continuing to plug your five year plan if the SF error rate it's based on is 1000 times too optimistic.

Avatar of DiogenesDue
tygxc wrote:

@6267

"For English, there are about half a million words in the biggest dictionaries."
"The average individual only uses about 30,000 words."
++ I did not say that an average individual would be able to memorise 10,000 games i.e. about a million positions.  Several people have a proficient conduct of several languages.

Psakhis had memorised all Fischer Games.
Fischer had memorised all Spassky games and much more.
Dorfman had memorised all 210 games of Zürich 1953 and could recite for each diagram: the round, the players, the result, and the game plans.
It is plausible that Carlsen had for his matches memorised all Anand, Karjakin, Caruana, and Nepomniachtchi games and much more.

There's a difference between "knew almost every game really, really well" and memorizing them.  Ergo, super GMs constantly mis-remembering their prep.  When you find a player that can recite 10,000 games without mistakes, you let me know.

Avatar of tygxc

@6283

There are actors who know plays of 30,000 words by heart.
There are humans who have memorised 100,000 digits of pi, which has no pattern.
10,000 chess games should be feasible.
Even knowing 1000 perfect chess games should provide an advantage.

This forum topic has been locked