Chess e-boards: square-occupancy vs piece-recognition

IMO chess e-boards are the greatest innovation in chess since Internet chess servers, enabling over-the-board chess games against online opponents. But if you are thinking of buying one you might be surprised to know there are at least four different types of e-board and the differences are important.

The four main types of e-board result from forks in the road from two big design choices. The first is obvious: does the e-board magically move your opponent’s pieces or does it merely highlight squares using LEDs so you can move your opponent’s pieces (as well as your own)? I’ll cover this difference in another post but TLDR: moving-piece e-boards are way more complex than LED boards, and have historically struggled to achieve sufficient reliability to not be frustrating. The second design choice is more subtle. Does the e-board use sensors that can recognise exactly which piece (and which colour piece) is on each square, or does it only know whether squares are occupied or not? Let’s call the former piece-recognition and the latter square-occupancy.

At first glance this second design choice seems silly. If manufacturers can do piece-recognition (they can) why wouldn’t they all do it? It seems obvious that knowing exactly which kind of piece is on a square is better than trying to figure it out from changes to square occupancy during a game. So let’s look into the pros and cons of the two approaches.

Square-occupancy: pros and cons

PROS: I can only think of one pro and it has to be lower cost. Until recently I am guessing that inserting magnets into the base of 32 x chess pieces and 64 x Hall effect or reed switch sensors into the e-board (to detect the per-square presence or absence of those magnetic pieces) must have been much cheaper than using passive RFID sensors (32 x passive RFID tags in the pieces plus 64 x RFID sensors in the board squares).

CONS: Lots. The e-board software (typically split between on-board firmware and a bluetooth- or USB-connected app) has to ‘understand’ the rules of chess to translate changes in square occupancy into chess moves that can be relayed to your online opponent. This sounds easy but it isn’t. A simple piece move is relatively straightforward. It is picked up (leaving one square) and put down (occupying another). But even that can go wrong if the piece is slid over intervening squares. Takes can be done on the board in different ways. Do you lift the taken or the taking piece first? En passant pawn captures are even more mind-boggling (and the SquareOff Pro e-board for example still struggles to reliably recognise en passant pawn captures). But the most obvious shortcoming of square-occupancy is pawn promotion. You have to tell a square-occupancy e-board what kind of piece you are promoting to, as it has no way of knowing what you are replacing the pawn with. In summary, square-occupancy MAY reduce the manufacturing cost of the e-board (I’m not sure this is still the case) but it MASSIVELY increases the complexity of the e-board software (compared to piece-recognition), which has often led to the kind of reliability problems that plagued the SquareOff (now Miko) e-boards.

Piece-recognition: pros and cons

PROS: Simpler, more reliable software and a better player experience. Want to set up a puzzle or a custom position? Just move the pieces on the e-board to the position and the app is immediately updated. Want to under-promote to a knight? Just do it on the e-board, without needing to switch attention to a separate app to select a promotion piece.

CONS: None! With the arrival of competitively-priced piece-recognition e-boards from companies like Chessnut. Cost definitely was an issue with (for example) DGT e-boards, which are now starting to look very overpriced.

🐐

If you enjoyed this post please share it with your chess friends and follow my BillyGate project on Facebook. Thanks!