Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Online Chess With Physical Pieces On a Chessboard

Soulskill posted more than 4 years ago | from the i'll-take-your-castle-with-my-horse dept.

Hardware Hacking 63

D4C5CE writes "A chess-playing German tinkerer has contrived (and made a video of) an amazing contraption that plugs real chess pieces into the freechess.org server using a 20W LED projector and an old webcam to read moves on a projected chessboard."

Sorry! There are no comments related to the filter you selected.

Microsoft Surface (1)

BradleyUffner (103496) | more than 4 years ago | (#32860454)

This looks like the same concept behind Microsoft's Surface AKA Big-Ass Table.

Re:Microsoft Surface (1)

dotancohen (1015143) | more than 4 years ago | (#32860684)

This looks like the same concept behind Microsoft's Surface AKA Big-Ass Table.

No, the surface needs each item to be individually marked on the bottom:
http://what-is-what.com/what_is/microsoft_surface.html [what-is-what.com]

Not the same at all. However, there was another camera technology once mentioned on /. that is really similar, but I cannot find it now. Anyone?

Re:Microsoft Surface (1)

hitmark (640295) | more than 4 years ago | (#32861248)

given that chess is only one change at a time, i suspect it could work with a ms surface system without the pieces being marked.

in the video we see him pressing some kind of button on the side between each move. I guess the camera takes a snapshot at each press, looking for changes and interpreting those as moves (no longer a dot at one position, and a dot on a previously blank position equals a move between the two spots).

Between that and the known starting positions of a chess board, one can get away with not marking the individual pieces; as long as one do not try to start the system in the middle of a game with no prior moves history.

Re:Microsoft Surface (1)

TangoMargarine (1617195) | more than 4 years ago | (#32867072)

Castling might be a little tricky...it would need a special rule I suppose.

Re:Microsoft Surface (1)

hitmark (640295) | more than 4 years ago | (#32867832)

heh, its one of those rules in think casual players of chess forget even exist.

Re:Microsoft Surface (1)

pcb-dev (1853218) | more than 4 years ago | (#32961576)

Yes, there is a special rule ... But there is a en-passant which is even more special :-)

Re:Microsoft Surface (1)

TangoMargarine (1617195) | more than 4 years ago | (#32962310)

given that chess is only one change at a time

was what I was addressing. En passant, by comparison, is pretty easy after you've detected it's a legal move.

Not like Microsoft Surface at all (3, Informative)

D4C5CE (578304) | more than 4 years ago | (#32860886)

Actually he says in his project diary [diy-community.de] (in a forum where they usually take TFTs apart to rebuild them into things like these [diy-community.de] or even one's own R2D2 [diy-community.de] ) that quite contrary to Microsoft's approach, all his web cam looks for are the pieces' shadows on the board cast by ambient light.
No need to detect more than that, the server knows which chessman comes from a particular square.

Re:Microsoft Surface (0)

Anonymous Coward | more than 4 years ago | (#32862320)

No, it's not the same concept ... "Conventionel" multitouch applications use FTIR or DI and this ultra-low-cost approach only uses ambient light. For chess this is sufficiant and it is cheaper, because no additional light source for multitouch is needed.

Radio Chess? (2, Insightful)

Soulshift (1044432) | more than 4 years ago | (#32860460)

Seems very much like the radio chess sets that saw some popularity in the late '80s.

Re:Radio Chess? (1)

jshackney (99735) | more than 4 years ago | (#32860878)

I knew it looked familiar!

overkill tech (-1, Troll)

Anonymous Coward | more than 4 years ago | (#32860482)

I built something like that 5 years ago using magnets and reed switches.

Re:overkill tech (0, Offtopic)

Anonymous Coward | more than 4 years ago | (#32860732)

I built something like that 5 years ago using magnets and reed switches.

FUCKING MAGNETS??? HOW DO THEY WORK???

Nice (1)

wave84 (1790348) | more than 4 years ago | (#32860496)

It's kind of amazing when you consider it; all this work has been done by just one guy. He could turn this into a business, there are lots of chess hobbyists around..

Re:Nice (1)

DaveAtWorkAnnoyingly (655625) | more than 4 years ago | (#32860534)

Indeed, if he could make a board that you just plug into your router, that would certainly be an appealing product. Better still, wifi it's ass. You really don't need a projector, you only need to know that "a" piece has moved from x to y. You don't need to know what piece it is, as that data is already stored on the server. All you need to know is where the piece has moved from and to. As a previous poster said, magnetism would do it with reed switches, or I'm sure some solid state electronics could achieve the same thing. hmmmm, i'll stop typing here and get typing into my patent portfolio!

Re:Nice (1)

cgenman (325138) | more than 4 years ago | (#32860716)

I always wondered if you could use a whole lot of mini-actuators to have pieces "grow out" and "grow back" into the board. Perhaps in a 3x3 configuration, with 2 different height levels and a touch sensor on the middle actuator. Each piece could be arranged out of the 3x3 array of voxels, and moved by touching the middle of a piece, then touching the middle of an unused spot on the board. One could use blue and red LED's under the squares to show which piece belongs to whom.

That would still require 576 actuators, 64 touch sensors, and a whole lot of wiring. But it would be a fun project, and wouldn't use magnets (in *that* way).

Opponent moves? (0, Flamebait)

kav2k (1545689) | more than 4 years ago | (#32860508)

Wake me up when it can physically move pieces for your opponent's moves. For now there are certainly tens of ways to do piece tracking.

Re:Opponent moves? (2, Informative)

JWyner (653364) | more than 4 years ago | (#32860600)

Most of the physical "chess trainers" you could buy 10-15 years ago would move the computer's pieces automatically. I think the point of this article is that there is a DIY system presented that can be easily connected to the existing gaming infrastructure. Sure, you could design a piece of hardware that moves the opponents pieces and links up to the public servers, with wifi, but that would not be an inexpensive DIY project anymore.

Re:Opponent moves? (1)

TheRaven64 (641858) | more than 4 years ago | (#32865708)

Really? I have a chess computer from about that era, maybe slightly more than 15 years ago, but less than 20. It didn't move the computer's pieces, it just indicated on the board which piece to move, you had to move them both. That worked by having buttons on each square, so you had to push the piece down slightly before and after moving it. The cheaper units didn't even have this, they just came with something that looked like a calculator and so you had to type in the grid coordinates of your moves.

Re:Opponent moves? (1, Informative)

Doc Ri (900300) | more than 4 years ago | (#32860632)

you mean like this:

http://www.ismenio.com/chess_mb_phantom.html [ismenio.com]

Re:Opponent moves? (1, Interesting)

lobiusmoop (305328) | more than 4 years ago | (#32860638)

Or even this [hackaday.com] ?

Re:Opponent moves? (0)

Anonymous Coward | more than 4 years ago | (#32860714)

Sweet sweet chess p0rn.

Re:Opponent moves? (1)

PopeRatzo (965947) | more than 4 years ago | (#32860734)

"Pieces" "moving" on a "board" are for sissies.

Real men play blindfold chess.

Via snail mail.

Re:Opponent moves? (1)

Nevynxxx (932175) | more than 4 years ago | (#32861734)

Real men have a photographic memory, and remember the board in their head, whilst playing with psychics who can "read" the board....

Re:Opponent moves? (1)

Sulphur (1548251) | more than 4 years ago | (#32863006)

Wizard.

Lightning chess (3, Interesting)

Rockoon (1252108) | more than 4 years ago | (#32860530)

Most of the games played on freechess.org are 'lightning' games, and usually its the 'fastest' setting of "1 0" which is 1 minute total per player with no time added per move. Some of the players individually have game counts in the hundreds of thousands. Normally these players play 'sets' of games, 10 or more consecutively, against any given opponent.

Some of the player play a very hucksterish style, with the goal being to simply eat away at their opponents clock with surprise checks, highly dubious sacrifices, and other outright gorilla tactics. Its very fun to watch, and to play.

Re:Lightning chess (2, Informative)

Main Gauche (881147) | more than 4 years ago | (#32860594)

Most of the games played on freechess.org are 'lightning' games, and usually its the 'fastest' setting of "1 0"

Most? I'm logged in right now:

Lightning games in progress: 24
Blitz games in progress: 331
Total in progress: 473

Re:Lightning chess (1)

Rockoon (1252108) | more than 4 years ago | (#32860602)

Those lightning games are over in less than 2 minutes, and another one started by the players immediately. Yes, most.

Re:Lightning chess (2, Informative)

Vellmont (569020) | more than 4 years ago | (#32860676)

Heh. Most games are lighting merely because they play a lot of games because of the short length of a game.

Most PEOPLE are actually NOT playing lightning games, and most games in progress are not lightning games. The percentage of people playing lighting games would tell you a hell of a lot more about the popularity of lighting games than the number of games.

*sigh* (1)

Main Gauche (881147) | more than 4 years ago | (#32860772)

Yes, most.

Ok, if 15% counts as "most" [ficsgames.com] , then you are right.

This is what I get for posting in a chess thread.

Re:*sigh* (1, Insightful)

Anonymous Coward | more than 4 years ago | (#32860922)

That's what you get for posting to a Slashdot thread.

This place is so full of retards who are willing to use any twisted logic to "prove" their point that it has driven off most level headed people. The only positive in this all? It keeps these same people from mingling in the streets. Here they're just a pest, in the real world they would be a menace.

Re:Lightning chess (1)

rthille (8526) | more than 4 years ago | (#32860678)

Another excursion in thruput vs latency...?

Re:Lightning chess (1)

Main Gauche (881147) | more than 4 years ago | (#32860814)

I found it more of an excursion into simple division.

Most typical lightning game: 1 0
Most typical blitz game: 3 0

3/1 = 3
331/24 ~= 14

Orders of magnitude off. So even the roughest of estimates suggests that lightning games wouldn't be the majority of games. Anyway, there's no need to even use my estimate, since I've now posted freechess's statistics [ficsgames.com] elsewhere in this thread.

Blitz games outnumber lightning games about 5 to 1. (What a coincidence, 14:3 is on the rough order of 5:1.)

Re:Lightning chess (1)

MokuMokuRyoushi (1701196) | more than 4 years ago | (#32863490)

Who knew gorillas were such brilliant tacticians.

Re:Lightning chess (0)

Anonymous Coward | more than 4 years ago | (#32865006)

This is fun!
Do you want your business to grow? This is a tip.why outsource [whyoutsource.org] | benefit outsourcing [whyoutsource.org]

Radio Shack 1974 (1, Funny)

Anonymous Coward | more than 4 years ago | (#32860540)

Old shit. Then TRS80 interface in 1982. All the old fuckers are dead now so this is new to young stupid punks like you!!

Re:Radio Shack 1974 (1, Funny)

Anonymous Coward | more than 4 years ago | (#32861078)

Take another swig of Geritol pops.

Re:Radio Shack 1974 (1)

Sulphur (1548251) | more than 4 years ago | (#32865682)

Six pack at the ready.

--

Someone asked Dudley LeBlanc what Hadacol was good for. He said about three million last year.

Good hack (5, Insightful)

cgenman (325138) | more than 4 years ago | (#32860660)

Give credit where due, people. This isn't impressive because move-sensing chessboards are somehow new. This is impressive because he made his own projector that displays time and opponent move information on the board. Then he used an old, old webcam and custom software to determine the move that he is taking. He made circuit boards, frames, and other equipment. He probably spent 100 hours and a lot of effort in order to make it slightly easier to play chess online.

If a DIY setup like this doesn't bring at least a little smile to your face, your hacker spirit is dead.

Re:Good hack (1)

omglolbah (731566) | more than 4 years ago | (#32860748)

Indeed!

Re:Good hack (3, Informative)

v1 (525388) | more than 4 years ago | (#32860832)

it kind of looked like he may have been pressing a button on the side of the board when it was time for the next move. His hand always returned to that position and moved a bit. after each move from either side.

I also saw that "oops" of pushing the pawn too far, obviously there's an undo function there somewhere or he had not yet released the move.

I thought the adding of the clocks to the board was a very nice touch. But looks like the room was darkened and no place to put one's legs under the table while playing. two minor drawbacks.

Re:Good hack (1)

RJFerret (1279530) | more than 4 years ago | (#32861088)

I also saw that "oops" of pushing the pawn too far, obviously there's an undo function there somewhere or he had not yet released the move.

Yes, and there was no visual feedback when he moved the pawn two squares forward, not until he took back and made his corrected move. IE, it didn't seem to require an "undo" but your later suggestion instead.

It seems he might be taking a snapshot of the board, then figuring out the move from comparing the previous shot with a now empty square, to the new shot with a newly occupied square.

Ironically, perhaps easier with lower resolution ("old webcam") data. The use of an LCD was brilliant, as it eliminates lighting issues trying the same thing with a regular board.

I also noticed his clock consistently ran two seconds beyond his making of his move.

Re:Good hack (1)

Molochi (555357) | more than 4 years ago | (#32861518)

"I also noticed his clock consistently ran two seconds beyond his making of his move."

I saw that too, but I just attributed it to the time it took to physically move his opponents pieces. He's actually granting a handicap this way.

Re:Good hack (0)

Anonymous Coward | more than 4 years ago | (#32862248)

Yep, it's explained in the video description of youtube: "Unfortunately, Playing live blitz games on this Internet chess table gives me a disadvantage, because I have to play the moves of my opponent while my clock is ticking. That means, I have time disadvantage which could be severe in fast blitz games."

Re:Good hack (0)

Anonymous Coward | more than 4 years ago | (#32861064)

The reason that I'm not smiling is something you've mentioned yourself. "a lot of effort in order to make it slightly easier to play chess online."

And the software is not that brilliant either. You have snapshot a and snapshot b of the chess board (before and after). grab the bitmap and do a simple a-b subtraction and you have where the piece was originally at. Swap the 2, and you have where the piece(s) have moved to.

Have the maker add automation of moving the opponents pieces, and then i'll be a bit more impressed.

Re:Good hack (1)

santax (1541065) | more than 4 years ago | (#32861796)

For this one I happy take a -1 redundant. You are absolutely right! And I find this project very encouraging to get my own iron out and have a go at it.

Re:Good hack (1)

Jaysu (952981) | more than 4 years ago | (#32864140)

This project is similar to my old senior design project in college. We built a chess game where the computer could detect and physically move the piece to the proper place (video below). It brings a smile to my face, now that I'm not being graded on it anymore :)

http://www.youtube.com/watch?v=hNtWeTDev_k&feature=player_embedded# [youtube.com] !

Re:Good hack (1)

pcb-dev (1853218) | more than 4 years ago | (#32864778)

I knew this project, because I was curious about the surface you were using. Unfortunately - no offense - it was sooo loud and slow, that there was no way to use something like this for real chess games. But great video anyway :-)

Re:Good hack (1)

Jaysu (952981) | more than 4 years ago | (#32865994)

Yeah, we only had one semester to build that, so that is all we could do with that time frame. Hindsight 20/20, we would have picked faster motors. The sound could easily be reduced by enclosing the unit. We left it open for display purposes. As for the surface, we used plexiglass, and gave it a coating to reduce the glare. Good times.

Yeah... (1)

SlashDev (627697) | more than 4 years ago | (#32860850)

I'd get rid of the pixellated lines and propose a blinking square where the piece needs to land.

Also been a student thesis project (1)

CptPicard (680154) | more than 4 years ago | (#32861514)

Interestingly, a few weeks ago I read that something very much like this also was some sort of graduation project of a couple of students at a tech school here in Finland... don't have the link handy unfortunately.

2071 (1)

boarder8925 (714555) | more than 4 years ago | (#32862026)

Reminds me of the Cowboy Bebop episode "Bohemian Rhapsody [youtube.com] ."

Sort of off topic (2, Interesting)

BigSes (1623417) | more than 4 years ago | (#32863090)

I often wondered why they never gave Deep Blue, or any other chess playing PC, an interner connection. S, if you think youre the tits, you can go online to www.beatdeepblue.com or some shit, and play. You may have to wait in a queue, so it can dedicate all computational power to the art of chess, but why couldnt this be possible? Pipe dream I guess, or vodka dream.

why pawn to b4? (1)

Group XVII (1714286) | more than 4 years ago | (#32863538)

I would have moved either the knight or the bishop to a6, or else sacrificed the bishop at g6. What's the best move at that point?

A poor mans DGT Chess set (1)

Silver Surfer 1 (193024) | more than 4 years ago | (#32863584)

http://digitalgametechnology.com/site/Electronic-Boards/

Although the DGT boards are really nice the cost is something many people will not be willing to pay. Odd that after all the years it has been out someone else has not released a more economical version.

Re:A poor mans DGT Chess set (1)

pcb-dev (1853218) | more than 4 years ago | (#32864800)

I knew DGT boards and if they could what I wanted, I would have propably bought one. They can capture pieces, but you can't show the move of the opponent on the board. I didn't want to put a distracting monitor besides my board ...

DGT Chess sets unable to display opponent's moves? (1)

D4C5CE (578304) | more than 4 years ago | (#32866042)

Are you saying DGT amazingly (even more so at their price tag) stopped short of placing a cheap little LED on each square, as pretty much every stand-alone chess computer has done at least since the early 1980s?

Misread (1)

mqduck (232646) | more than 4 years ago | (#32864124)

I first read "contraption" as "contraception". Seems almost as accurate.

Golly! Where's my flying car? (1)

Impy the Impiuos Imp (442658) | more than 4 years ago | (#32868946)

Jeebus H. Christ, I thought someone hooked up a chess piece-moving robot arm to one of these web chess sites several years ago.

No off-the-shelf projection displays w/ interface? (1)

D4C5CE (578304) | more than 4 years ago | (#32871948)

A friend's request to display custom content on a projection clock made me wonder if/why there are really no such displays (as in "1ft digits across the room") that could be computer-interfaced with the likes of lcdproc, or even for pixel graphics.

Has anyone seen such a thing commercially available without the need for DIY optics?

FINALLY (1)

rdpratt (1854096) | more than 4 years ago | (#32874230)

Cowboy Bebop anyone?
Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?