Neither Tromp nor I claimed the position was a draw or the play was optimal.
It is an example of an extremely difficult class of chess problem - proof games with specified objectives. It is no closer to competitive chess than the game you posted from the Belgian championship where two players pre-arranged a draw and played out a very short stalemate.
Hahahahaah.
This thread is just about done. Tygxc repeating some flawed jibberish, megacheese and his literal mathematicians every once in a while, nerdy personal insults. Thats about it
it is done. i used to go through his points one by one with him and attempted to make him see reason. Then, i decided to verify my main claims by going to my maths professors.
they verified them completely. so now im sitting here telling tygxc that, but he doesnt seem to like it.
Im only here to constantly remind tygxc (and others) that he is delusional.
This thread is just about done. Tygxc repeating some flawed jibberish, megacheese and his literal mathematicians every once in a while, nerdy personal insults. Thats about it
it is done. i used to go through his points one by one with him and attempted to make him see reason. Then, i decided to verify my main claims by going to my maths professors.
they verified them completely. so now im sitting here telling them that, but they dont seem to like it.
All it requires is knowing the basics of stats to lol it isn't that hard to disprove them
All it requires is knowing the basics of stats to lol it isn't that hard to disprove them
yes, but i thought that having mathematicians verify it could have some effect on tygxc.
there would be a 50% chance that you would miss it with your sample.
u couldnt be further from the truth. plz do ur arithmetic next time. careful meggy in trusting these so-called math ppl here.
there would be a 50% chance that you would miss it with your sample.
u couldnt be further from the truth. plz do ur arithmetic next time. careful meggy in trusting these so-called math ppl here.
Sure so university math people or random person ?
All it requires is knowing the basics of stats to lol it isn't that hard to disprove them
yes, but i thought that having mathematicians verify it could have some effect on tygxc.
The issue is people scapegoat into going "these mathmatitions arnt trustworthy " well they arnt perfect but they know far more than you do dispite not being perfect
Like how in chess occasionally a 1200 can beat a 2000 but it almost never happens
there would be a 50% chance that you would miss it with your sample.
u couldnt be further from the truth. plz do ur arithmetic next time. careful meggy in trusting these so-called math ppl here.
im not talking about them. im talking about IRL people who ive discussed their PHD research with in person.
Lola 'believes in' tygxc's suggestions?
I doubt if Lola does.
Nor that she 'believes in' the denialism of manmade global warming either.
Is it a case of defending 'the underdog'?
The underdogs aren't always 'the good guys'.
Would it be ok to remove "black can always force a win" from the possibilities?
Would it be ok to remove "black can always force a win" from the possibilities?
Why not?
You could also remove 'white can always force a win' and 'white or black can always force a draw' too.
Would it be ok to remove "black can always force a win" from the possibilities?
technically not. zugzwang can exist so it must be separately proven that the starting position isnt a black win.
I wonder whether there is any possibility of showing white has at least a draw by some sort of strategy-stealing argument. Ideally we would show that there is a set of positions such that any black strategy has to reach one of them and white can force the same position with colours reversed.
To give a hint of how this might work, it is impossible for the following two positions after 2 ply - both having a perfectly natural single move by black - to both be part of a winning strategy for black, because white has the opportunity to "steal" the second position from the first one, and following black's strategy thereafter means the value of it cannot be worse for white (i.e. must be a win or draw for white)!
To prove chess is at least a draw for white, it would suffice to show that every black strategy either includes such a position that allows a steal or can be shown in some other way to fail to win for black. I am not saying this would be easy. It may or may not be much easier than exhibiting a drawing strategy for white.
I wonder whether there is any possibility of showing white has at least a draw by some sort of strategy-stealing argument. Ideally we would show that there is a set of positions such that any black strategy has to reach one of them and white can force the same position with colours reversed.
To give a hint of how this might work, it is impossible for the following two positions after 2 ply - both having a perfectly natural single move by black - to both be part of a winning strategy for black, because white has the opportunity to "steal" the second position from the first one, and following black's strategy thereafter means the value of it cannot be worse for white (i.e. must be a win or draw for white)!
To prove chess is at least a draw for white, it would suffice to show that every black strategy either includes such a position that allows a steal or can be shown in some other way to fail to win for black. I am not saying this would be easy. It may or may not be much easier than exhibiting a drawing strategy for white.
keep in mind that without a check or specific capture, black can lose just as much tempo as white, so white cannot break parity.
any strategy stealing argument would have to be extremely large scale to the point that it wouldnt even be considered to be strategy stealing because of all the other techniques needed. i mean I Guess it COULD be considered that but strategy stealing is no longer doing most of the work of the proof in that instance.
Note that if BOTH sides could always steal strategies, this would prove chess was a draw, so this is not an argument against it being used to prove black does not have a winning strategy!
Chess has actually been weakly solved already in the most commonly analysed variations.
They'll never bother to dedicated system resources to fully solving it as a) it's probably a draw so not much point and b) there's not enough money involved to bother. Chess engine developers will continue to plug away at it but they don't have the funds to make a qualitative leap forward in the next 10 years at least unless something big changes all that ( maybe torch?)
Perhaps it would be better: Why can’t chess be solved yet?
And the answer is because today it is still beyond our abilities to develop the appropriate algorithm.
I frequently read figures, which I am not quite sure how they got, where they indicate that the number of possible moves is greater than the atoms in the universe and that due to this complexity there is no current computer that can calculate all possible moves and develop the appropriate algorithm. They even talk about quantum computers and the like.
Well, I have news for you. Chess engines have been used since time ago to calculate all possible moves in some simple endings and generate endgame tablebases containing all possible positions. Yes, all. And you know what? Chess engines continue to use endgame tablebases because no one has been able to develop an algorithm to replace them.
That simple.
The current top engines are unable to play accurately in 6 piece endings (according to MARattigan, I think?)
Regarding algorithms, the problem is there is not much scope for better algorithms. Solving chess involves checking a prodigious number of variations and there's no good reason to believe there is any way to get around that. It's very unlike (most) mathematics where deep structure permits extremely efficient proofs of huge results.
Have we even had broad consensus yet... that we've actually done a quantum calculation that has been faster than our current fastest classical computer could do it? Every now and then a claim pops up, but then competitors call it bogus for one reason or another.
Quantum computing is in it infancy and going slowly. Let's see it factor a large number first... playing chess will be many decades later.
Lets hope that 'quantum computing' doesn't go the way 'cold fusion' has gone.
In theory there could be 'quark computing' and 'electron/proton/neutron' computing and atomic computing and molecular computing and chemical computing.
Just now I tried to look up the mass and size of the most basic circuits in a computer.
Didn't get very far.
The term 'transistor' seems to have a new meaning in the context of modern computers.
And there's another term MOSFET.
But just now i saw a picture of a MOSFET - it had a match in the picture for scale.
Obviously the primary circuits or junctions or gates or whatever are much much smaller than that.
"Moore's law is the observation that the number of transistors in an integrated circuit (IC) doubles about every two years. Moore's law is an observation and projection of a historical trend. Rather than a law of physics, it is an empirical relationship linked to gains from experience in production."
------------------------------
Point: if a computer chip's most fundamental elements each contain a gigantic number of atoms - then that would suggest a lot of 'room for improvement'.
Using Avogadro's number - one could calculate an approximation for the number of atoms (or molecules if its a compound) in the most basic element of computers.
But not without knowing the weight of that element.
And I'm not referring to 'chemical element'.
I'm referring to the component. The most basic component.
A powerful computer can do better than one quintillion floating point ops per second.
That's better than 10^18
but how many basic elements does it need to do that?
How fast can each element fire? How many firings per second?
--------------------------
Last time I brought this up years ago - tygxc didn't get it at all.
He started talking about 'nodes per second' ...
refusing to directy admit the connection between ops per second and what a computer can do.
but he has already conceded that chess can't be solved with current technology.
In other words he's essentially contradicted himself.
To solve chess - computers have to be fast enough.
How fast a computer can work is determined by how many operations per second it can do.
tygxc failed to grasp that connection?
Or just pretended to so fail?
He appears to have Disdain for mathematical objectivity.
Perhaps now we'll get Tactics from tygxc:
'No No No ! The speed is determined by the Algorithm !!'
Algorithm improvement will help the overall progress - but an algorithm cannot and does not change the ops per second speed of the hardware.
It would be like saying your car's dashboard display can change the number of grams in each liter of the car's gasoline.
----------------------------------
Plus O is projecting as usual. Accusing others of his own deficiencies and faults.
But that's expected. Predictable.
And quickly spotted by the new guy here.
----------------------------------------------