If you live near Central Park or some other local chess hub, you’re likely never short of opponents for a good game. If you find yourself looking for a computer opponent, or you just prefer playing online, you might like this LED chessboard from [DIY Machines] instead.
At heart, it’s basically a regular chessboard with addressable LEDs of the WS2812B variety under each square. The lights are under the command of an Arduino Nano, which is also tasked with reading button inputs from the board’s side panel. The Nano is interfaced with a Raspberry Pi, which is the true brains of the operation. The Pi handles chess tasks—checking the validity of moves, acting as a computer opponent, and connecting online for games against other humans if so desired. Everything is wrapped up with 3D printed parts, making this an easy project to build for the average DIY maker.
The video tutorial does a great job of covering the design. It’s a relatively simple project at heart, but the presentation is great and it looks awfully fun to play with. We’ve featured some other great builds from [DIY Machines] before, too. Video after the break.
Next step – autodetection of where chess pieces are on the board. Are there any (very) cheap solution for such thing, tiny rfid readers?
Back in the day, magnets and reed-contacts had been used.
Measuring inductivity of different coil types could be a possibilty, too?
So that merely a few unique coils for each type of chess piece are required?
If so, that would avoid the need for semiconductors. It would be purely analog. Except for the A/D converters used by the sensor fields, of course.
A topview image from a camera with some software to diff before and after images would do the trick. Would be nice if the camera could be hidden in a ceiling or hanging table light so it would be a mystery how this move information was obtained
Why a Nano and a Pi? The Pi could do everything.
That is probably because of the LED strip, those things have some tight timing requirements that may not sit well with a Pi running Stockfish unless some RT patches are added to the OS. That and there will be a handly little library that can be added to an Arduino project without any programming.
Yes, I know, but it makes me sad.
The SPI bus on a pi can be abused to get accurate ws2812 timing.
it’s depressing being a downer all the time but i just don’t see the point of this. it would be better as a smartphone app :)
seriously, the UI is awful. seems like four button pushes on an awkward overloaded keyboard for every move, and then you still need to move the piece — one intuitive interaction is replaced with 5 interactions, 4 of which are deeply counterintuitive (but learnable; everything is learnable). and if you are careless then the board state will drift from the computer’s internal state. if you fat finger a neighboring piece onto the floor, it’d be an ordeal to design a UI that would let you determine where the computer thinks the piece used to be. one of the best features of a computer chess system is going backwards in time to answer “whoa, how long has that been open and i didn’t see it?? how did it happen? what was i thinking?” sort of questions, and this won’t help you at all with that.
maybe i’m a poster child for why whiskey o’clock and chess o’clock should be scheduled separately but i can definitely imagine misinterpretting an Nf3 computer move as an f3 move (i.e., moving the pawn forward instead of the knight out, at the beginning of the game), and simply being completely lost when it becomes significant 10 turns later.
the contrast to a touchscreen is huge. the tablet or phone shows you where the pieces are, and detects what your move is.
really, smartphones a curse. soft buttons are the ultimate evolution of ‘worse is better’. huge families of hacks that would have made sense as a compromise 25 years ago now seem hopelessly clunky, and pointless to boot. sigh. everything is so easy, it is hard to justify doing anything :)
(i 3d printed my own chess set even though store bought ones are fine…i guess i like having a painted plywood board better than glass or cardboard, as the only advantage)
Hm? The input method is classic, though.
The earliest chess computers and computer programs used numbers and letters.
For example, you move a pawn by entering A2-A4.
Chess games on CP/M had to be played via terminals, too and used this input method.
sure you can use grid coordinates but the standardized algebraic notation is better because it can be intuitive and brief. “a4” instead of “a2-a4”. “Pa4” if you have to disambiguate that it’s the pawn. very normal way to interact with a chess computer through the keyboard.
but since this multiplexes the letters and numbers onto the same 8 buttons, you can’t do anything for brevity…you struggle just to keep straight whether it’s time to enter a row or column. even if you don’t mind a bunch of button presses, it’s clunky.
and anyways that’s the thing…in 1995 when this was competing with the radio shack chess computer, you could take clunky for granted and it would seem like a neat idea. but now touch screens have made clunky seem much less inviting for this particular use. and that’s what i’m really lamenting — not the perfectly fine hack — but that a lot of modern technology has made a lot of hacks seem stupid, completely irrelevant to what the hack would have been seen as in a vacuum. “it would have been better as a cellphone app” is the curse of the modern era
Funny, looking at the picture I thought this was using a flatscreen tv as the playing surface, with custom button mask to make it square… oh well, next time.
Good electronic chessboards are reading each piece position individually through induction — each piece has a simple resonator (capacitor+ inductivity) tuned into different frequencies while the external induction is done on the range of these frequencies simultaneously, and then the resonating frequency is being sensed. RFID readers are also possible to be used — through multiplexing. Good quality and size chess boards with just individual piece recognition and simple LED indication under every cell, single color, cost 1000-2000 USD each. The next step is to somehow be able to move pieces using an arm with an electro magnet underneath.
But the holy grail is to make both (individual piece recognition and pieces moving) work at the same time. No one ever did that.