One day, I want to learn some knowledge about the Game AI topic.
Chess series is supposed to a big collection in Game.
Xiangqi is my fav chess game, and I am familiar with it.
It’s a good choice.
Besides, six years ago, I have developed xiangqi program both in java & C++.
I can use it to learn AI engine ,and the program developed before , can be seen as a GUI.
Techinically, it will be OK then.
Reality is cruel, the Java version is lost at home, and the C++ version have to run on Window, not my Mac OSX.
Chess & AI
Chess is a two-player strategy board game played on a chessboard, a checkered gameboard with 64 squares arranged in an eight-by-eight grid.
It is one of the world’s most popular games, played by millions of people worldwide at home, in clubs, online, by correspondence, and in tournaments.
So if chess has AI, then we can compete with Computer.
More interestingly, we can let two Computer with AI fight with each other.
There are all kinds of chess game, which seems alike.
They have these Feature:
* board with X * Y
* pieces with some rule
Xboard is a chess program, FairyMax is a chess engine(AI)
Xboard is a graphical user interface for chess in all its major forms,
including international chess, xiangqi (Chinese chess), shogi (Japanese chess) and Makruk,
in addition to many minor variants such as Losers Chess, Crazyhouse, Chess960 and Capablanca Chess.
FairyMax
FairyMax is developed by H.G.Muller, who is super fan of chess.
Maxqi is the xiangqi engine.
D(k,q,l,e,z,n)/* recursive minimax search, k=moving side, n=depth*/intk,q,l,e,z,n;/* (q,l)=window, e=current eval. score, E=e.p. sqr.*/{/* e=score, z=prev.dest; J,Z=hashkeys; return score*/intj,r,m,v,d,h,i,P,V,f=J,g=Z,C,s,flag,F;unsignedchart,p,u,x,y,X,Y,B,lu;struct_*a=A+(J+k&U-1);/* lookup pos. in hash table*/q-=q<e;l-=l<=e;/* adj. window: delay bonus */d=a->D;m=a->V;F=a->F;/* resume at stored depth */X=a->X;Y=a->Y;/* start at best-move hint */if(z&S&&a->K==Z)printf("# root hit %d %d %x\n",a->D,a->V,a->F);if(a->K-Z|z&S|/* miss: other pos. or empty*/!(m<=q|F&8&&m>=l|F&S))/* or window incompatible */d=X=0,Y=-1;/* start iter. from scratch */W(d++<n||d<3||/*** min depth = 2 iterative deepening loop */z&S&&K==I&&(GetTickCount()-Ticks<tlim&d<=MaxDepth||/* root: deepen upto time */(K=X,L=Y,d=3)))/* time's up: go do best */{x=B=X;lu=1;/* start scan at prev. best */h=Y-255;/* if move, request 1st try */P=d>2&&l+I?D(16-k,-l,1-l,-e,2*S,d-3):I;/* search null move */m=-P<l|R<5?d-2?-I:e:-P;/*** prune if > beta unconsidered:static eval */N++;/* node count (for timing) */do{u=b[x];/* scan board looking for */if(u)m=lu|u&15^3?m:(d=98,I),lu=u&15^3;/* Kings facing each other */if(u&&(u&16)==k)/* own piece (inefficient!)*/{r=p=u&15;/* p = piece type (set r>0) */j=od[p];/* first step vector f.piece*/W(r=o[++j])/* loop over directions o[] */{A:/* resume normal after best */flag=h?3:of[j];/* move modes (for fairies) */y=x;/* (x,y)=move */do{/* y traverses ray, or: */y=h?Y:y+r;/* sneak in prev. best move */if(y>=16*BH|(y&15)>=BW)break;/* board edge hit */t=b[y];/* captured piece */if(flag&1+!t)/* mode (capt/nonc) allowed?*/{if(t&&(t&16)==k||flag>>10&zn[y])break;/* capture own or bad zone */i=w[t&15];/* value of capt. piece t */if(i<0)m=I,d=98;/* K capture */if(m>=l&d>1)gotoC;/* abort on fail high */v=d-1?e:i-p;/*** MVV/LVA scoring if d=1**/if(d-!t>1)/*** all captures if d=2 ***/{v=centr[p]?b[x+257]-b[y+257]:0;/* center positional pts. */b[x]=0;b[y]=u;/* do move */v-=w[p]>0|R<10?0:20;/*** freeze K in mid-game ***/if(p<3)/* pawns: */{v+=2;/* end-game Pawn-push bonus */if(zn[x]-zn[y])b[y]+=5,/* upgrade Pawn and */i+=w[p+5]-w[p];/* promotion bonus */}if(z&S&&GamePtr<6)v+=(rand()>>10&31)-16;// randomize in rootJ+=J(0);Z+=J(4);v+=e+i;V=m>q?m:q;/*** new eval & alpha ****/C=d-1-(d>5&p>2&!t&!h);/* nw depth, reduce non-cpt.*/C=R<10|P-I|d<3||t&&p-3?C:d;/* extend 1 ply if in-check */dos=C>2|v>V?-D(16-k,-l,-V,-v,/*** futility, recursive eval. of reply */0,C):v;W(s>q&++C<d);v=s;/* no fail:re-srch unreduced*/if(z&S&&K-I)/* move pending: check legal*/{if(v+I&&x==K&y==L)/* if move found */{Q=-e-i;a->D=99;a->V=500;/* lock game in hash as loss*/R-=i/FAC;/*** total captd material ***/Fifty=t|p<3?0:Fifty+1;returnl;}/* & not in check, signal */v=m;/* (prevent fail-lows on */}/* K-capt. replies) */J=f;Z=g;b[y]=t;b[x]=u;/* undo move */}/* if non-castling */if(v>m)/* new best, update max,best*/m=v,X=x,Y=y;/* no marking! */if(h){h=0;gotoA;}/* redo after doing old best*/}s=t;t+=flag&4;/* fake capt. for nonsliding*/if(s&&flag&8)t=0,flag^=flag>>4&15;/* hoppers go to next phase */if(!(flag&S))/* zig-zag piece? */r^=flag>>12,flag^=flag>>4&15;/* alternate vector & mode */}W(!t);/* if not capt. continue ray*/}}if((++x&15)>=BW)x=x+16&240,lu=1;/* next sqr. of board, wrap */if(x>=16*BH)x=0;}W(x-B);C:if(a->D<99)/* protect game history */a->K=Z,a->V=m,a->D=d,a->X=X,/* always store in hash tab */a->F=8*(m>q)|S*(m<l),a->Y=Y;/* move, type (bound/exact),*/if(z&S&&Post){printf("%2d ",d-2);printf("%6d ",m);printf("%8d %10d %c%c%c%c\n",(GetTickCount()-Ticks)/10,N,'i'-(X>>4&15),'9'-(X&15),'i'-(Y>>4&15),'9'-(Y&15)),fflush(stdout);}}/* encoded in X S,8 bits */if(z&4*S)K=X,L=Y&~S;returnm+=m<e;/* delayed-loss bonus */}
Then I think for a while….
maxqi engine is so pop, why nobody discuss this??
why no notes? Google maxqi,no good results.
Then I Google fairymax, still no good results…
At last, I saw that the fairymax derived from Micro-Max…