Chess will never be solved, here's why
Sort:
I agree
#3358
" You said: the classical model is validated by experiments" ++ i.e. induction
I say my method is validated by deduction.
I say the error distribution for 127 draws and 9 decisive games is 126 - 9 - 1.
If you were to say after investigation it is 125 - 9 - 2, then I am willing to accept that.
On one hand you say you can explain the data, but on the other hand you refuse to present an alternative error distribution saying you cannot tell.
So you can tell and you cannot tell at the same time.
tygxc wrote:
#3358
" You said: the classical model is validated by experiments" ++ i.e. induction
I say my method is validated by deduction.
I say the error distribution for 127 draws and 9 decisive games is 126 - 9 - 1.
If you were to say after investigation it is 125 - 9 - 2, then I am willing to accept that.
On one hand you say you can explain the data, but on the other hand you refuse to present an alternative error distribution saying you cannot tell.
So you can tell and you cannot tell at the same time.
If you say that dark matter is chocolate because chocolate is dark and sticky and I say I cannot tell, it does not mean that dark matter is chocolate. Data do not include the error rate. For the seventh time: what is the difference between your reasoning and the example of the room?
#3361
For the 7th time: try come up with a plausible alternative error distribution in lieu of 126 - 9 - 1.
I say 3 + 5 = 8. You say that is wrong because the continuum hypothesis is not proven. I then ask what is 3 + 5 in your opinion? You say you cannot tell.
Jun 9, 2022
0
#3306
tygxc wrote:
#3358
" You said: the classical model is validated by experiments" ++ i.e. induction
I say my method is validated by deduction.
I say the error distribution for 127 draws and 9 decisive games is 126 - 9 - 1.
If you were to say after investigation it is 125 - 9 - 2, then I am willing to accept that.
On one hand you say you can explain the data, but on the other hand you refuse to present an alternative error distribution saying you cannot tell.
So you can tell and you cannot tell at the same time.
You can tell that you cannot tell. This is perfectly clear to everyone else in the world.
#3363
On one hand he says he has an alternative explanation that does not need the assumption.
On the other hand he refuses to present an alternative plausible result.
How far is a car from its starting position after driving 400 km north and then 300 km east?
Me: assume Pythagoras, answer = 500 km. He: That is wrong. The Earth is not flat so spherical trigonometry is required. Euclides' axiom is not proven. We cannot tell.
Do you even read, or think, before you post?
haiaku wrote:
If you say that dark matter is chocolate because chocolate is dark and sticky and I say I cannot tell, it does not mean that dark matter is chocolate.
People have not to come up with an alternative conclusion to challenge your conclusion. I explained why your assumption of statistical independence is not plausible, though. Do you have an objection to the last post about that?
I say 3 + 5 = 8. You say that is wrong because the continuum hypothesis is not proven.
If you really want to use such metaphor, you say x = 8 and I explained why (errors not statistically independent) x cannot be determined. Your attempt to prove otherwise is fallacious because it's circluar thinking. Do you want to prove otherwise? For the 8th time, what's the difference between your circular reasoning and the example of the room? Do you think that in that example the conclusion validates the premise?
I answered your questions.
Jun 9, 2022
0
#3309
tygxc wrote:
#3363
On one hand he says he has an alternative explanation that does not need the assumption.
On the other hand he refuses to present an alternative plausible result.
How far is a car from its starting position after driving 400 km north and then 300 km east?
Me: assume Pythagoras, answer = 500 km. He: That is wrong. The Earth is not flat so spherical trigonometry is required. Euclides' axiom is not proven. We cannot tell.
Well, at least that answers @playerafar's conjecture that you're a flat-earther.
#3365
For the 8th time: assumption - calculation - validation are commonly used in many sciences.
It may appear circular, but it is valid.
Dark matter is unrelated to chocolate, and your room story is silly.
I answered your questions.
You did not answer my question about your alternative for 126-9-1.
If you cannot tell, then you cannot tell I am wrong either.
#3366
I am even heliocentrist.
Jun 9, 2022
0
#3312
tygxc wrote:
#3354
Carlsen and his team of grandmasters and cloud engines apparently was convinced 1 e4 e5 2 Nf3 Nc6 and 1 e4 c5 draw as he played it consistently in his last two world championship matches.
False.
Carlsen was convinced he could get a desirable result against his opponent with this line. He had no confidence he could achieve the same result against Leelazero (say), so the game theoretic value of the position is not actually the issue. All that matters is practical chess is the statistical expectation of results.
MARattigan wrote:
snoozyman wrote:
...
Since there are more chess games (10^120) than the number of atoms in the observable universe (10^80), it is highly unlikely that chess engines will ever completely solve the game of chess with all 32 pieces on the board in our lifetime.
Since there are more chess games (10^120) than the number of atoms in the observable universe (10^80), it is highly unlikely that chess engines will ever completely solve the game of chess with all 32 pieces on the board in our lifetime.
10¹²⁰ b*llocks! That's a very rough estimate of the number of possible 40 move games with a constant 30 moves on each ply given by Shannon and never intended to represent the total number of chess games.
The number of possible games under FIDE basic rules is א₀ if you consider only finite length games or if you allow (necessarily countable) infinite length games ב₁.
Also the number of atoms doesn't have much to do with it. The number of possible arrangements and states of atoms is far more relevant and that's vastly bigger. (On pre-quantum theory physics, at least, a single atom could encode the full set of up to 32 man tablebases and it would just be a matter of whether you could measure and set with enough precision to read and write the encoding.)
I severely doubt that.
Regurgitating dubious figures is not a good approach to a feasible solution.
playerafar wrote:
sachin884 wrote:
Instead of posting all this you could have played 100 chess games
Or - eaten 100 pints of Haagen Dazz french vanilla ice cream.
dont make my mouth watered. yum
#3369
Both Caruana and Carlsen ran LC0 on the cloud engines they rented for preparation.
If Carlsen was not sure of 1 e4 c5 and of 1 e4 e5 2 Nf3 Nc6, then he would have varied as black, but he did not. If Caruana and Nepo were not sure of 1 e4 e5 2 Nf3 Nf6, then they would have varied as black, but they did not.
If Carlsen had thought either 1 e4 or 1 d4 had a faint chance of winning, then he would have played that exclusively, but he did not: he varied 1 e4 and 1 d4 and 1 c4, and within that he varied variations as white.
3000 posts = 1 month time
sachin884 wrote:
3000 posts = 1 month time
in this 1 month time you could have roamed 10 countries
This thread is surely the most nonsense thread of them all. Now to work ....
could have eaten 30 cheese pizzas (and gone to doctor) yo know why you would become unwell
#3352
"the assumption of statistical independence between errors is not different from the example of the room"
++ It is different. I gave several examples from other sciences. Assume something. Calculate based on that assumption. Verify the assumption. Assumption and calculation valid.
If you would impose that all calculations were done with relativistic quantum mechanics, Navier-Stokes equations, Maxwell equations, then the world around you would not be as it is. We would live like in stone age.
In that post I explained why those kind of examples are different. Do you not agree? If so, why?