An idea for a "physical interface for OGS" - Feedback required

This is really cool! I like the idea of just two lines of LED to point to the place where a stone has to be played. It reduces the need for LED from 361 to just 38 and makes the installment of sensors on the cross-lines much easier.
Thanks for sharing!

2 Likes

But wait, thereā€™s more! If you copy this design now, you get the patent infringement lawsuit for free. :stuck_out_tongue_closed_eyes:

:slight_smile: I clearly need to slightly ā€œadoptā€ it ā€¦ Donā€™t think there is a patent on LEDs indicating a cross on a matrix.

I give you my patent for an laser, or minitorch
and two actuators for free :slight_smile:

1 Like

I donā€™t know, patents tend to be insanely vague.

Right. Iā€™m not one of those people. OP asked ā€œQuestion: would that be something you would like to play on or is it just a weird wine idea?ā€, which I answered.

Alsoā€¦
Is this something Iā€™d like to play on? Yes.
Is it a weird wine idea? Yes.
Do I want it anyway? Yes.
Do I think OP should make it just because itā€™s awesome despite the practical issues? Yes.

2 Likes

Sounds like a cool idea! Itā€™d be fun to try one out, for sure. For building one, Iā€™d probably do a photoresistor at each intersection, with some LEDs arrayed around that. If you did two colors of LEDs, you could even play on the board without using stones, and just tap the photoresistors to register a move. Two of each color arrayed around each intersection, with greens aligned N/S and blues aligned E/W would have a nice, symmetric visual effect.

1 Like

Thanks for the cool advises. I did some research and I assume that LED and sensor at each intersection will look odd, because they will exceed the intersection marker significantly which might destroy the authentic Go board feeling.
Thus, I think just sensors at the intersection and LED on the two axis might work better (just like the board in the post above).

1 Like

You could make the whole thing a big kindle paperwhite interface using stones that register on it.

Without backlighting, it might even have a nice aesthetic.

Thats true. However, I think a kindle-like display of that size will be very expensive, wonā€™t it?

Maybe not, although itā€™s been a minute since I looked.

e-ink isnā€™t latest greatest tech, and itā€™s used for more than just kindles these days. Itā€™s a viable low-power interface option, although obviously more so if purchased in volume.

I like the idea. Did some thinking about it too, when I started to play.
What I like about this idea the most is the possibility to use it for live tournaments for the notation. No mobile phones with joseki programs running in the backround needed anymore :wink:

My 2cents:

361 photocells and their wiring (!) / circuit board is too cost intensive in my opinion.
https://www.google.com/search?q=photocell+price&source=lnms&tbm=shop&sa=X&ved=0ahUKEwjFrMfP8YTgAhXNy6QKHWBBCDwQ_AUIDigB&biw=1406&bih=714
-> Camera and picture recognition from above (need not to be from the top can be from an angle, let software do it)
-> where is a chip camera everybody has? In a mobile phone. -> small rig to hold the phone to look onto the board.
-> the phone could do the communication with the ogs-server as well.
Problem unsolved with the indication of where the opponents play was. -> Laser pointer in the rig that holds the mobile phone?

This board would also natively support an insane version of memory-go where you do not place your opponentā€™s stones :smiley:

2 Likes

Photoresistors are quiet cheap (2Ā¢ per resistor) https://m.aliexpress.com/wholesale/photoresistor.html?categoryId=400503&channel=direct&searchType=mainSearch

Agree, the sensors are not the most cost-intensive part of this board, it is the time invested for the wiring. Wiring 361 sensors manually will take some timeā€¦
The idea with the mobile phone is something I have thought of as well. I am not really good in image recognition programming, would have to do some research. The issue, as you stated, is the indication of the positioning of the opponents stones. However, this could still be done with a special board that has LED on the two axises. This is should be rather easy.

A question regarding photo resistors:
If one puts a stone on a photo resistor the resistor answers with the state of ā€œdarkā€. Could it somehow decide if it is a white or a black stoneā€™s dark?

The resistor doesnā€™t care, all you need is the most recently activated resistor. Pseudocode:

latestResistor=listen(ResistorInterrupt); playMove(latestResistor[x,y])

This is true for handicap games as well because as long as the system is color-ignorant, itā€™s irrelevant whether your opponent played first and you played second, or whether your opponent got a 2-stone handicap.

Though I suppose it does need some special code for handicap games that refrains from sending moves until all handicap stones are placed.

Of course this doesnā€™t guard against user error a clear case for a refund (it wonā€™t prevent you from placing the wrong stone on the right intersection).

Yep, this would be my understanding as well.

I saw such a goban IRL years ago (maybe 10) in China. Each intersection had a red LED (I think it was blinking to signal a move). The idea was not to connect to the internet, but have a built-in go software.

Also, the Shanghai museum of science used to have a robot arm that could play gomoku. A bit more elegant than the contraption in the video shared by @ehomba.