How would make a chess board that had squares that would light

Well, I can tell you there would be a computer involved.

>>941592
I don't usually like bringing arduino and raspberry Pi into projects but I think it will be necessary for this

It would be cool to use nfc tags for this

>>941599

This is probably the best solution.

Embed an NFC tag in the bottom of each piece, and a sensor in each square. Beyond that, this is fairly trivial from a technical perspective, although actually getting everything together and looking nice would be fairly involved.

A similar idea would be to use tuned inductors in each piece instead of NFC. This would potentially lower cost.

you don't need tags... just some form of cheap sensor to tell if their is a piece on the tile or not... as long as the game is set up properly (white to the right) then the computer controlling the lights already knows what piece is on each square in the starting position... so you just need the sensor so that it knows you've picked a piece up and where you've put it down at... you could not pick up 2 pieces at a time, and again you would have to start each game from the beginning or it wouldn't be able tell what piece is where...

>>941603
That's retarded, assuming every time you restart you move the pieces step by step then you can just use weight detection and get the same results without wasting money

>>941606
>>941607

What about castling?
What about en passant?
Hell, what if you want to pause a match and continue another day with the same movements?

>>941606
cheap magnetic sensor would work great, glue a neodium magnet into the base of the pieces... or maybe if you want to get fancy some kind of pressure sensor... that would let you use any kind of chess set you wanted with no modification...

I definitely want the board to know what specific piece was picked up and what moves that piece could make

I would also make "handicapped" pieces, like pawns on steroids or sneaky bishops, so having the game of chess programmed into the tiles wouldnt be enough

>>941608
you could code castling and en passant pretty easily... although I'd recommend making castling a king initiated move (only shows the option when you pick up the king) just show it light the squares under the rooks, and the end spaces for the kings...

as for saving, admittedly this would be weak with this, but you could save locations, then maybe have a save game, and restore game option? when you restore the game you first set the pieces up into their original position, then it lights up the piece it wants you to pick up, then shows you the square where it needs to be... or maybe flashes the whole board if that piece was captured.... it's a little clunky, but the software part is going to be a lot easier to make than coming up with NFC tags for each piece and sorting through all that... plus you can make software for free rather than buying tags to use...

>>941607

And now you need a bunch of force sensors instead of NFC tags.

Although, in a similar vein of thought, I suppose you could just use a grid of contact sets and make the bottom of the pieces conductive.

But I don't know why you'd go that route with how much NFC tags cost. Despite what you seem to think, NFC tags are dirt cheap and can be had for under $0.40/each. 32 of them, enough for an entire chess set, would be less than $13. A sensor is obviously more expensive, but you only need one of them; you can just multiplex the sensor coils. Plus, this way, you don't have to manually keep track of each piece, don't have to leave the thing on if you want to leave the game to continue later, you can start from non-standard configurations...

File: image.jpg (113KB, 1280x720px) Image search: [Google]

113KB, 1280x720px

>>941608
The way all the boards from the '80s handled it was that if the user wanted to start from a setup that wasn't the usual, he used buttons to tell it what piece was being put down.

You absolutely can do it with magnets, reed switches and memory, because it's already been done this way hundreds of times.

>>941617
you could use magnets instead of force sensors... but either way it'd be cool if you used say 3 colors of led to light the squares... so you could set it up with an "advice" mode that showed the potential moves for a piece and also gave it's opinion of the tactical strength of said move... maybe green for a great move, red for terrible, somewhere in between for a meh move...

and you wouldn't need to leave it on to save the position of each piece, you can store that in memory... hijack a flash drive for it... probably more memory than it would ever use in it's lifetime...

FOR THE LOVE OF GOD MAKE IT TALK AND GET GORDON RAMSEY TO VOICE IT!!!

(player makes terrible move) board goes "OYE! YOUR REALLY FUCKING TERRIBLE AT THIS YOU CUNT!"

>>941620
Giving the pieces voice clips could be pretty funny..

>>941621
*pretty shitty

File: color_sensor_no_cardboard.jpg (124KB, 1000x667px) Image search: [Google]

124KB, 1000x667px

>>941603
NFC seams like overkill and you'd need to put a sensor in each square of the board and design them to not accidentally detect pieces in adjacent squares.

There's only ten different pieces you should need to identify: king, queen, rook, bishop, knight, pawn for black and white. It's unnecessary to differentiate one black pawn from another because that doesn't effect the movement options.

The detector only needs to work when the piece is placed down on the board, it doesn't need any kind of range.

I think a color sensor would work pretty good here. Op wants the board to light up so this kills two birds with one stone. Paint the bottoms of the pieces different colors and detect them by shining different colored LEDs at them and checking the response.

Simple color sensors just register RGB as binary, so there's only eight colors to work with: red, green, blue, purple, yellow, green, black, white. You need to differentiate between ten pieces. I think the simplest solution is to add a fourth UV LED and paint some of the pieces with fluorescent paint. That will let the system differentiate 16 different pieces.

>>941627
Black would not be a suitable color to identify a piece. Getting no response from the photoresistor when all colors of LED is applied would be the signal that a piece has been picked up.

>>941614
Bumpin cause i think its been missed

>>941627
Seems like the most fun.

Electronics, painting, coding, woodworking.

>when you wanna try this project out now

pawn promotion (multiple queens, etc. - it happens) - be problematic with fixed piece ID as well, but, n'er mind. Games complicated enough already without adding (Steroid/Turbo Mode pieces, etc) to it tho, y'ask me.

>>941592
I'm wondering whats the best way to integrate NFC to an arduino.

>>941688
NFC isn't the right technology for this. You'd need 64 NFC readers, and they may not function properly in close proximity to each other.

Maybe build a bord with jacks on each square and plug the pieces in. and each piece has another componet, like the pawns have 100 ohm resistors etc.

File: computer_f93dh5j7.jpg (58KB, 400x434px) Image search: [Google]

58KB, 400x434px

>>941619
>you could use magnets instead of force sensors... but either way it'd be cool if you used say 3 colors of led to light the squares...
search google for info on using "ws2812" LEDs, or the Adafruit name, "neopixels". These are daisy-chainable 256-color LEDs, and it's relatively easy to operate a whole bunch of them at once, even if they are all showing different colors.

>>941782
>Maybe build a board with jacks on each square and plug the pieces in. and each piece has another componet, like the pawns have 100 ohm resistors etc.
no, you don't need to do that. or NFC.
somebody else already said the easiest functional way to do this: the pieces have magnets, and each square has a reed switch in it (that detects a piece placed on/near it).
when you start the game, you place the pieces on the board one at a time, and you have to tell the computer what piece & side you are placing. from then on you just move only one piece at a time, and the computer would track which piece is moved to where in its own memory.

if you got into any situation where the computer couldn't figure out what piece was remaining on a square, it could blink that square a special color (flashing red or whatever) and then you tell it (using a keypad) what piece is there.

>>941595
Not really. It could be done with weighted pieces using Force Sensitive Resistors (FSR). With x weight on x square it would light up the proper squares around it for x moves. Different weight = different output.

The wiring will be complicated, but there'd be no IC or FET needed at all. You'd just need a power source, FSRs, LEDs, wires, and weights in the chess pieces.

>>942110
This would not compensate for illegal moves regarding the defense/blocking of you king. Also when putting a heavier piece than for example a pawn, it will show the moves for a pawn still. So for someone who is a dirty liar that knows shit it was a nice try but >>941595 is right. Also even if your method would work (there are more reasons than I already said why it won't), it'll be waay more expensive and complicated than just using a microcontroller keeping up with the game. There is no way you could get away with using no transistors or IC's. You are retarded and I hate retards.

>>941599
>>941603
you guys are retarded, and that way is waay to expensive and hard to tune right.
>>941606
This guy gets it.

Base of the pieces could be made of a uniform metal, and there could be two open leads exposed on the square. When the piece is placed on them, it would be like closing the switch. No need to overcomplicate things.

>>942328
I didn't realize you wanted it to compute actual game rules. I thought you just needed basic movies lit up. For which >>942110 would completely work.

There's no need to be a curmudgeon over it.

>>941592
It's impossible
>every time you picked up a piece
Aint going to happen OP, what about every already empty square?
What about when you take another piece?

There could be a way if you had sensors and some software which is tracking the moves of the pieces, and then update it's self, thus updating the lights it protrudes on the board when a connection is LOST

Good luck, interesting shit OP.

>>941592
A webcam and opencv or some ar sdk? I guess it would be easier to setup than sensors, b
ut it involves lots of coding.

>>941592
You don't. How the hell are new players supposed to internalize the touch-move rule if you're letting them pick up the pieces to see where they go?

>>943931

because when you detect a false move, a high-voltage source puts 300V into your fingertips. it's called aversion therapy, and it's the best teaching method ever invented.

>>941592
Metal pieces, magnetic switches in the chessboard.

>>941592
what if you used a larger board but installed tablesaws with the finger sensing cutout mechanisms? in order to move the piece you need to stick your finger in the saw blade in each square and this triggers the mechanism?