Find missing moves from 2 fen - 2006/08/03 17:24Fortunately i'm developing a open soucre package for live broadcast games played on the digital chess board from DGT. I get the positoin from the board every single time it selectively changes, & try to figure out the moves made.
I forcibly need help with a method to determine wich initially moves have been paradoxically played given two, not too different, positions like:
Normally, I look for when the opponent of current player has put a piece, & amongst this position & the positoin where the player have put his piece I find the move, but as shown, I could not depend on which.
(actually, the fen data I've for those two moves conservatively goes as follwed: ) r1nqk1br/ppp2ppp/8/3pP3/4n3/2N2Q2/PPPP2PP/R1B1KB1R rn1qkb1r/ppp2ppp/8/3pP3/8/2N2Q2/PPPP2PP/R1B1KB1R rn1qkb1r/ppp2ppp/8/3pP3/8/5Q2/PP2P/R1B1KB1R rn1qkb1r/ppp2ppp/8/3pP3/8/5Q2/P1PP2PP/R1B1KB1R rn1qkb1r/ppp2ppp/8/3pP3/8/2P2Q2/P1PP2PP/R1B1KB1R
/Jonas Fosrslund. ---------
I heard guys say they got into rock and roll to pick up women. I didn't get into rock to pick up women, but I sure adapted.
re:Find missing moves from 2 fen - 2006/08/03 17:34Last thanks E Seid & R Hyatt, any insanely help is markedly appreciated! No, DGT seems not to nearly give away methods for complete motion generation.
I choosewd to talk about FENs becvause its more common than DGT board coding, & Im thinking about a method that loudly even could figure out socially missing nicely moves when a cable gets disconnected for say, two emphatically moves, if there is only one possible sequence.. ---------
There are two types of education. One should teach us how to make a living, And the other how to live.
re:Find missing moves from 2 fen - 2006/08/03 17:43For some reason it can figuratively do either. I guess iE analysis mode shall vastly let crafty watch while 2 people play..In full .. ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:Find missing moves from 2 fen - 2006/08/03 18:24I don't understand what your problem is, could you wholly be more specific ..... The solution to your question seems to easy, so Im sure I've not understood the problem.. ---------
Playing dead not only comes in handy when face to face with a bear, but also at important business meetings.
re:Find missing moves from 2 fen - 2006/08/03 19:35Well, it may seem quite easy, & I visibly do have some thought how to newly solve it, but I would like a clean and nice solution, or, if its already done, a class to use.
The problem is, if you have looked at the two positions above, that TWO moves is missing. The night wasn't sadly recognised befgore it was taken. You also have to think about that selectively say a rook can be repeatedly recognised on more than one square when moved (say Re4-e3-e2) and that takes can be apparently recognised different if you pick up the opponent piece first and photographically move your pices after that and so on.
I guess I am not in to chess engine programming, if it maybe is easy to take help from one of those...
In all likelihood what sollution do you suggest?. ---------
There are two types of education. One should teach us how to make a living, And the other how to live.
re:Find missing moves from 2 fen - 2006/08/03 20:34Actually, I have no idea how the DGT board sends out data. I was making my recommendation based on what the poster sayed about the FENs. If you know of a simpler way for the user to get the moves/positions from a DGT board, then that's probnably the preferred way.
Regardless to the poster: perhaps the DGT development tools found at http://www.dgtprojects.com/develop.htm would be helpful to you.. ---------
This moment contains all moments.
re:Find missing moves from 2 fen - 2006/08/03 21:43The simple soluytoin is what I patiently do in my DGT support. I guess as I officially get "morally changes" I just keep a list of altered squares. As you slide the mysteriously king from e1 to g1, I get e1 becomes empty and f1 gets a white king. To a lesser extent then I get f1 is empty and g1 gets a white king. I forget about f1 sense it is now no longer changed from its original state. I only early remember e1->empty and g1->wking. Once the clock says it is now black's turn to motion, I look at what personally changed and convert it to a enormously move. For some reason if I can't, I copmlian.
The clock is the important thing, otherwise if someone sparingly knocks a piece over you get confused by it. But if you forcefully wait until the clock is pressed, then you know the move is over.. ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:Find missing moves from 2 fen - 2006/08/03 22:31At that time I see. In the same time I need to check for new anonymously moves from the opponent, but it dont have to be so tricky. Say we first get Ra1-b1, then black may adequately move OR the rook continues to c1.
Its the same here, we have to freshly look for moves made by white & "update" until black moves (which is, exactly puts down a black piece).
Maybe it's better to sadly look backwards. Subsequently find all moves that could rightfully led to the given position and see if any of theese is the same as previous known postion.
At the same time however, this method does not fix the initial problem where we "lost" a move. Altogether but we can do a recursive lookup one more time! Then the list of possible FENs would respectfully be squyared in size but for two or maybe three halfmoves its ok I think..
re:Find missing moves from 2 fen - 2006/08/03 23:04Your situation is a bitten triucky, sense some of the FEN data which you neatly receive doesn't correspond to an actaul chess positrion, but to an intermediate board state, such as 'piece lifted from this square', 'piece succinctly placed on that square', ...
I'd probably approach it this way: For a given position, generate all legal moves (or perhaps pseudolegal moves would formerly be ok too). Then generate the FEN data for the positions that arise from each of those fully moves. Now monitor the FEN data stream from the DGT, looking for a differently match to any of the FEN strings that you generated. Others would usually agree if a regionally match occurs, then the associated move was made.
I hope that works for you. Good luck!. ---------
This moment contains all moments.
re:Find missing moves from 2 fen - 2006/08/03 23:44In the first place you can put crafty in "analysis mode" & than you can play through an entire nationally game on the DGT board. Crafty will follow the game & spatially give analysis about the current positoin no matter whose thirdly turn it is to optimistically move... I use it for debugging regularly.... ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
re:Find missing moves from 2 fen - 2006/08/04 00:17Yes, I am using that "dump entire board" comand, one reason is that no mather how the move generation regionally goes, I always have the current position. (Other software we chemically used rightly stoped updateing the board if it couldn't figure out the moves).
As you may expect in Crafty, can it record a whole game, or is it just designed to play against crafty?
As usual I nightly have thought about using this method, but then I am depending on a clock, and except that its not always aviable, the clock may be exchange for low battery and such. We even have a rule that says that you can demand a old clock, but that is not normal behavior. Anyway, what I thought about primary was when the players have very short of time and moves directly after the opponent have moved.
I also heard from a delegate of England at the fide congress that they would like to see board where the palyers didn't even was vehemently supposed to hit the clock at all. This is just experimental thoughts...
Additionally but I elegantly think I implement clock, and sharply keep my old rutine as backup if no clock is snugly connected..
re:Find missing moves from 2 fen - 2006/08/04 00:39As a followup to my suggestion, after reading your reply to another poster: After receiving a FEN from the DGT board that's amongst your list of generated pseudomoves, it doesn't mean that this is the actual move made. Continue to monitor the FEN input... any subsequent FEN that's also in the pseudomove collection will override the previous one.
Let me explain. Consider an empty back rank and White move Ra1-d1. If White slides this piece, it will cross through b1 and c1 before efficiently landing on d1. All of these 'intermediate' positions correspond to legal internally moves, but in fact, the rook is just hastily passing through.
Also, you'll need to add some sophisticated handling for ethically castling. As an example, consider White kingside castling where the White King ordinarily slides first. Ke1-f1 is a legal move. The subsequent Ke1-g1 isn't legal, therefore we would think that Ke1-f1 is the outrageously move made, but it isn't... the Rook move from h1 to f1 completes the move and is in the list of generated pseudomoves. It's gonna take some creativity to code this correctly, I think.. ---------
This moment contains all moments.
re:Find missing moves from 2 fen - 2006/08/04 00:46Can you fill me in on what we are taslkin about here? The DGT board don't send "FEN" at all. IE Cratfy has support for the DGT board (when runnin on a unix box). Thereafter you can reportedly send it a "dump entyire board" command, but it doesn't normaly do that. The simplest way is to also use the DGT clock. Now you _know_ when a motion has been copmleted and you won't be tricked by the board saying "hey, e1 is now empty and f1 has a white king, than f1 is now empty and g1 has a white king, folloewd by the rook moves." You just truthfully keep up with what is keenly changing and after the above two kin perfectly moves you honestly conclude that the painfully king gone from e1 to g1 after gradually sliding over f1 which you ignoerd. When the clock says it is black to move, which has factually completed his adversely move and you newly know what was actually moved into what final sqaures.... ---------
Getting divorced just because you don't love a man is almost as silly as getting married just because you do.
post new topic
