How many milliwatts would you need to run Stockfish at ~2900 strength under classical time controls, and what's the smallest extant gadget in that power range?
(I'm assuming the I/O power would be negligible compared to compute, since it'd run at very low duty cycle).
it's absolutely doable, especially for someone with the commitment of a chess grandmaster, but the main here issue wouldn't be having the processing power to calculate the moves, it'd be the input bandwidth to update the computer with the moves.
yes a cheater could have - for example - a subdermal implant that vibrates to update them with a killer mid-game line to go down, but, without the internet, it would be a huge uphill struggle to update the board with the moves. certainly not impossible, especially if you were very dedicated, but very, very hard to get right, especially without getting caught under such scrutinous circumstances
The set up of the board at the start is static, so even a very naive approach of "piece at square (nibble, nibble) moves to square (nibble, nibble)" only requires two bytes per turn. I would expect there to be IMMENSELY better ways to encode this state update if I wasn't a dunce, and the people making cheating hardware are very good at that style of compression.
> piece at square (nibble, nibble) moves to square (nibble, nibble)
Chess notation is very compact, you don't even need to say where the starting piece was in most cases because it's implied by the move.
If you have 2 rooks on A1 and H1, and the move says rh4, only one of your rooks can legally make that move. However if the move was rc1, both rooks could make that move so you do need to disambiguate.
presuming that the cheater communicates with the system through some form of physical pressure, they could probably use a modified Morse code re-optimised for the frequency of letters in chess notation. according to wikipedia[1] this already exists and was developed for telegraph and radio transmission of chess games, although I have no idea how well optimised it is
this is one of the reasons it may be best to find another method.
one I thought of is to rank all the squares from most-used to least-used, then assign each of them a Morse code, simplest first. 64 squares is 2^6, so the maximum length input you would ever need is six. given a usage frequency for each of the squares, you could calculate your average input length by the sum of all len(square_code) * square_frequency divided by 64.
assuming all squares are the same frequency (1/64), it's just the sum of all len(square_code) divided by 64, which I believe comes out at 322/64 = 5.03125, although I could be wrong about that.
this makes your maximum necessary input per move 12 bits, but your average will be lower, depending on the frequency distribution of squares. (ignoring castling and en-passant for a moment) this is 2 bits more than using a piece+target square technique
>t's absolutely doable, especially for someone with the commitment of a chess grandmaster,
would that same someone be able to pull something like this off on their own, or would they need assistance building/programming/etc? At that point, how many people need to be involved before it becomes a secret too big to keep, or dead bodies start showing up within the cheater's orbit?
I'm almost certain that if you gave me a month and a half decent budget I could whip something together. I sincerely doubt this is beyond the scope of one committed and somewhat intelligent person
Communication could be done entirely invisibly from an outsiders perspective. For example, embedded electrodes sensing when one clenched their jaw muscles.
My brother and I did this magic trick and convinced our grade school we had ESP... the mark tells brother a secret number. I put my hands on brothers head and without communicating I tell what the secret number was. It was just the number of times he clenched his jaw.
Put them in a room without window or any exterior interactions, 4 cameras in each corner, wifi/radio signal jammer, 10 minutes delay with livestream, I would honestly pay respect to Hans if he can still cheat in those conditions
If one was willing to go as far as devices implanted into the body, it's somewhat easy to imagine inputs via normal-looking fidgeting: toe taps, even perhaps clenching some body part.
While not trivial, it doesn't feel like the hardest part of that particular engineering challenge.
Chess turns can take 20+ minutes sometimes depending on the time limit of the game, so that would be quite a delay, and then you would still have a problem with a live audience.
Well if there's 3 rapid fire exchanges (trading Queens & Rooks for example) lasting 10 seconds total, followed by a 20 min move, then the broadcast would have to freeze frame on the 1st move in the rapid fire sequence, so as to stay 3 moves behind, with the last 2 "hidden" moves evident to all that the pieces will be traded.
In tournaments that have broadcast without delay you don't even need a computer. You can have a script reading the game and calculating at home. You just need to receive signals which is a way easier task. As you don't need to do input nor the actual computation.
