computer thinks blacks will win

> computer thinks blacks will win
> actually whites will either win or draw

Other urls found in this thread:

sciencealert.com/scientists-need-someone-to-solve-this-chess-problem-and-reveal-the-key-to-human-consciousness
en.wikipedia.org/wiki/Fifty-move_rule
twitter.com/NSFWRedditGif

Prove it, faggot.

...

>implying a chess program will give a fuck 'bout the number of pieces on the board
gr8 b8 m8 tho

No baiting

sciencealert.com/scientists-need-someone-to-solve-this-chess-problem-and-reveal-the-key-to-human-consciousness

The human will always win with a simple move: knock over the pieces.

whose move is it?

Makes sense

Lmao this, we dont have to follow the rules if we dont want to.

White

I don't think that even matters.

Blacks only legal move is with the bishops, which won'r really achieve anything since all whites pieces are on white squares. White just moves around till draw, or of black is really trying to lose; lets white get a queen and the black king is mated.

Have not seen the computer calculation tho.

sooo... the computer bases its probability of winning based on the restriction of the white queen's movement because of the bishops? This article seems to be asking the wrong question. The computer's 'thought' process is clearly very limited.

if the white pawn moves the bishops will take it
the white king won't help

>queen
are you a fucking mongoloid?

Th-three black bishops?

The set up is nonsensical. There would never be someone with three bishops on the light diagonal. You either have pawns in front of one another or one of the white pawns should be a queen.

>As Sarah Knapton at The Telegraph explains, a computer will always assume the black player will win in this scenario, because seeing those three bishops will force it to perform a massive search of possible positions "that will rapidly expand to something that exceeds all the computational power on planet Earth".

>But Sir Penrose says it should be "easy" for humans, given you know your chess rules back to front

What if I dont know chess, can't be bothered to learn it, and won't waste my brain power on this problem because I've got better things to do?

Am I still better than the supercomputer because I excercise my free will to decide if I even want to bother with the problem in the first place?

the white king can just move along the white squares resulting in a draw or purposely surrender the bishops to the king, i.e white cannot lose but cannot win and black cannot lost but cannot win

Don't be racist.

What a dumb thing to say

Doesn't matter, this is about something else.

the experiment is how the computer thinks that it will win given the situation when clearly a human can quickly tell that black will most likely lose. the setup doesn't need to be a vanilla chess game at all, you would never all those black pawns in those positions either.

How so

...

Youre trying to be philosophical but come off as an edgy neckbeard.

M'lady

how does white win/draw here? the top pawn will have to move in order to take the rook/queen and put the king in check. the king then moves where the top white is covering and escapes. black then cleans the board.

I'm saying how the supercomputer will attempt to solve it and cannot because as they say it would perform a number of calculations that exceed its capability and that of any supercomputer on Earth

Meanwhile I conserve my brain power by refusing to solve it in the first place, who is really the supercomputer in this scenario?

Really makes you think

well i do know the chess rules back to front, and there's no way white winning or forcing a draw (by no legal moves) without an intentional act by one side to break the current standoff

then explain how? i don't see how black could *win* without a serious error by white, but likely lose?

someone needs to explain this to me. the white king can't capture any of the bishops unless they move off the line, and the white pawn can't advance without being captured.

Computer chess has come a long way this last decade. Komodo, Houdini, and Stockfish are the top programs.

You're retarded OP.
Eventually white has to move the top pawn, then black moves the king to the square beside it.
At that point the pawn either moves forward then gets taken by the king, or doesn't move and gets taken by the king.

Either of those moves releases the black queen which then forces the white king into checkmate by working with the bishops.

Any human player would concede the game at this point if they were white.

B-3 and C-4 pawns can take either rook and force the queen to take them. then the white king slowly moves in while parrying the black queen on white squares. the king stays trapped inside and the bishops can do nothing about the white king

lol at least the computer is trying

i'm surprised ppl didnt notice this.

Can white really win? I get the draw, but white winning seems impossible. It can't move the top pawn, it can't take a rook because that'll free the queen and it can't take a bishop.

Maybe the computer knows that in order for this setup to be possible, the rules must have been broken and the match is therefore forfeited.

Only if black makes a mistake

The experiment is actually how a supercomputer will attempt to calculate every possible scenario and never be able to do so because of the extreme limits of our technology in order to figure out how black will win when a human knowledgeable in chess can look at the situation and immediately determine that the black will not win and white will or end in a draw.

All in the span of a few seconds, what a supercomputer would take years even if it was powerful enough, to figure out.

It's to prove that our brains do not function like computers despite the popular misconception.

no... it really doesn't.

Keep moving the king on white squares and wait for all three bishops to move so that you can move white pawn up one square without being immediately taken by a bishop. Then it doesnt matter what black does for the next move, promote to queen and mate.

>Eventually white has to move the top pawn
why?

right

And will never succeed

>seeing those three bishops will force it to perform a massive search of possible positions "that will rapidly expand to something that exceeds all the computational power on planet Earth".

White can simply move its king repeatedly and force a draw

is this a political statement or some shit like that?
if so, nice b8.

actually either black wins, or it's a draw.
c6->c7
a6->b7
c7->c8 ->queen! check
b7xc8
then black wins from there

No its simply to unlock the key to consciousness, ganbatte user.

3 bishops? And all on black? Where'd you learn to play? Waziristan?

Because you can't just move your king back and forth and declare a stalemate when there is still valid moves you can make.
The rules essentially force white to surrender the game at this point.

then the computer keeps playing until it ends up being a stalemate.
if it's a supercomputer, then theoretically, it can know if the game is a stalemate.

This only works if the white player is retarded.

you cant get the black pieces out because the rooks can only move straight and the pawns are only to the white side. the upper white pawn blocks the king because you cant move your own king into check thus only the white king and the bishops are movable and the bishops are restricted to black squares only which the king can avoid

you arent smarter than a computer, youll be as smart as a monkey, because these too make the consious decision to masturbate

ganbarimasu

pawns that advance all the way can become *any* piece. so yes that arrangement could be constructed from legal moves.

...

You're assuming white makes an obvious mistake

correct, but otherwise this will be a draw with all pieces locked but the white king and the black bishops

exactly, i cant even see the stalemate scenario. the white king would still have moves to make........

the only other option is a draw though.

Not once you consider the other remaining black pawns and their positions.

en.wikipedia.org/wiki/Fifty-move_rule

Its a supercomputer that cannot perform the calculation requested of it in a reasonable manner but doesnt have the capability to understand that and will continue to attempt to perform it and solve the problem until it overheats or freezes up attempting to perform more calculations than it can handle

Meanwhile I on the other hand choose consciously to conserve my brainpower and not engage in the first place and yet I still have the capability in me to learn chess, git gud at it and figure out this problem in less time than it would take that supercomputer designed specifically for that singular task of solving chess problems, if indeed it can ever solve it.

Once again, it really makes you think.

>not playing "get a pawn to the other side of the board and get a piece back"

Oh. Ok. because if we were. Bottom pawn turns into queen, Black would probably move the queen to start setting up for taking the king, take the bottom rook with bottom pawn, and get a rook back, king takes rook, white moves queen to check the king, king retreats into original position, queen takes bottom pawn and checkmate.

nope. The bishops can't harm the King if he only stays in the white squares and they are the only one that can move. He then goes to position a8. at this point nothing can move exept the bishop. you then proceed to take the white pawn to get a queen (if you are lucky enough to not be in black's bishop line of fire) and then you put your queen in front of the king . With your King behind your queen, it's checkmate. Otherwise it would be draw.

there is no way that there is 3 bishops. cheat.

interesting rule. kinda makes sense when you think about it - if you can somehow force a 'check' of sorts on all the opponents regular pieces, the game won't force you to shoot yourself in the foot by making a positive move and shooting yourself in the foot.

consider me learned friend, thanks.

assuming letters are the yaxis and numbers are x

e3 -> d2 rook captured
(blacks irrelevant move)
f2 -> e1
(blacks irrelevant move)
trade g1 for queen (that's a thing right)

wait I just realized b1 could disrupt a key pawn at c3, how does this have a sure fire answer people play in all types of ways

additionally any disparity between our thinking and a computer's is a lack of intelligent programming not a fundamental way in which we think

if we model our programs on how we think we think and we're wrong , of course the computer will be less efficient

the whole thing is about how our brain loves to take shortcuts and discovering shortcuts requires a shit ton of creativity or millions of years brute forcing the shit in an environment to see what random changes cause something to be more efficient (think machine learning or natural selection)

it's not possible to get to a8 for the white king

read the thread before posting drivel

ITT: Faggots who don't know the rules of chess.

This position is a draw for white every day of the week.

fuck you're right. I'm an idiot.

I can't believe how stupid everybody on Sup Forums got.

First, Black can't lose.

Second, White can't win.

Third, if white move anything but the King they lose.

Fourth, conclusion of 1+2+3: white can easily draw and therefore reach the highest possible outcome.

Fifth, no computer is stupid enough to not realizing what is written above.

Sixth, OP is a lying faggot and probably left already an hour ago to proceed sucking dick.

8
7
6
5
4
3
2
1
ABCDEFGH

The white pawn can never become a queen.
If the white pawn moves the black king moves out and will capture it next turn, or release the black queen

yes yes we all see how it draws, what we don't see it how white can *win*, without black making serious mistakes to let it.

that's gay as fuck
if people just keep playing past the rule only to call a draw once they start losing

idk if you are bating our uneducated...

a computer will always do what you will tell him to do, most algorithms are either procedural or self learning, but even with these this calculation would be nonsensical as there are infinite turns that could be played and they are growing exponentially

if you had read this post you would understand, computers and human brains function fundamentally different

the computer doesnt have the tools yet to see that this is bogus

its basically if you compare the tree climbing abilities of a robot and a monkey, sure the monkey wins now and the robot designed to sucks, but the robot will be further developed while you stand still and think about how smart you are when you arent even attempting to think

really doesnt make you think, huh?

A chess bot has limited parameters to its way of 'thinking' and reaching a conclusion that it will win in this setup. The human brain has no boundaries to its thought and can see through the simple process that a computer is bound to. The computer has no ability to think outside the box per-SE.

But we are playing that way (three black bishops).

Yes I agree.

This. The problem is legit tho.
>sciencealert.com/scientists-need-someone-to-solve-this-chess-problem-and-reveal-the-key-to-human-consciousness

obvious draw. It would have been more interesting if there were also a black pawn on d6 and white pawn on d5 (and only two bishops), because then there is a winning strategy for white.

This guy gets it!

How would that work? You can't really keep somebody from winning like that.

it's literally in the article
>pass point where you can use rule
>fuck up 10 moves later
>call tie before he can trap you

what i find most interesting about this thread is that most of the discussion misses the purpose of the article entirely and the experiment itself

Really makes me think, thanks user I will read that post and maybe then I'll understand.

then I beat the puzzle.

A highschool dropout is smarter than a fucking super computer.

Draw is obvious
How on earth can whites win?

The rule applies to the LAST 50 moves. Maybe have somebody read it out loud to you.

see

>Have not seen the computer calculation tho.
Pretty much computer faceplants on its horizon because of all those possible bishop movements and has to make an assumption based on the existing pieces

Okay let's use your method

50 MOVES PASS THAT QUALIFY
ON MOVE 67 THE LAST 50 MOVES STILL QUALIFY
ON MOVE 68 YOU PLACE YOURSELF IN A POSITION TO BE CAPTURED IN 4 MOVES
ON MOVE 69 YOU CALL A DRAW BECAUSE 50 MOVE RULE

No shit thats the entire point of the experiment

Its to prove a point that our brains do not function like computers otherwise we wouldnt be able to solve this problem in less than 5 minutes just looking at the board and its also to attempt to unlock "the keys to consciousness"

But its basically a newly formed University research institute trying to make a name for itself and generate awareness and publicity with this puzzle and talk of keys of consciousness.

...

>Fifth, no computer is stupid enough to not realizing what is written above.
They are
Thats the point

That might be possible, I'm pretty sure it happened zero times though. 50 moves is A LOT. You'd have to either be completely clueless about chess or just not care for this to happen.

moar?

I think i remember something about that pic. Kinda funny story but not hilarious by any means.