05:54:09    JUN 14, 1986
	        Awit                     Rex
REAL mins:	  36                      48
CPU secs:	1989	                 390
    Amdahl 5860/2     EXPERIMENTAL	Compaq
	                 Round 4
b3 		  R E    ( 1) E5		  R E 28
Bb2		 0B*E    ( 2) NC6		  R E221
e3 	 	 0B*E    ( 3) D5		 1B=E229
Bb5		 0B*E    ( 4) QG5		 1B-E242
Ng1-f3		 14=E 94 ( 5) QXG2		 14=E201
Rh1-g1		 14=E 50 ( 6) QH3		 14=E234
Bxe5		 24+E151 ( 7) BG4		 14-E183
Rg1-g3		 14+E 77 ( 8) BXF3		 24-E173
Qxf3		 19+O 91 ( 9) QH6		 49-1175
Qxd5		 49>O192 (10) NG8-E7		.32<O151
Bxc6 +		 59+O129 (11) PXC6		 23-O280
Qf3		 14+O122 (12) ND5		 24-O128
Qg4		 89+O146 (13) QXH2		.12-O101
Bxg7		 14+O 73 (14) BXG7		 14-O132
Qxg7		 14+O 35 (15) KE7		 14-M118
Qe5 +		 14+O 60 (16) KD7		 13-M  5
Na3		 39+M 97 (17) NE7		#02 M134
Qd4 +		 14+M 64 (18) KE6		 14-M118
Qe4 +		 14+M 34 (19) KD7		 13-M432
Rg3-g7		 14>M110 (20) QH5		#04 M 27
Nc4		 14+M103 (21) KE8		#84 M 86
Ne5		 14>M 63 (22) F5		 14<M228
Qg2		 19>E104 (23) QH6		 19<M 71
Nxc6		 14>M 34 (24) NXC6		 14<M140
Rg7xc7		 14>M108 (25) KF8		 34*M 63
Rc7xc6		 14$E 32 (26) QG7		 14*E 78
Rc6-f6 +	 19$E 23 (27) QXF6		 15*E 99
Qxa8 +		 19$E 57 (28) KF7		 24*E 82
Qxa7 +		 19$E 28 (29) KF8		 46*E133
d4		 19$E 96 (30) KE8		 49*E 46
Ke2		 49$E170 (31) KF8		 59<E 60
c4		 29$E157 (32) H6		 79*E 47
a4		 39$E120 (33) RG8		 49*E 47
a5		 14$E 77 (34) RG7	        #54 E 47
Qc5 +		 29$E121 (35) RE7		 46*E133
a6		 14$E 70 (36) F4	        #64 E 44
a7		 14$E108 (37) F3	RESIGN   14*E 51
Kf1		 24$E 75 (38) KG8	        #24 E 40
a8 =Q +	 	 14$E 73 (39) KH7	RESIGN   24*E 64
Qc5-h5		 14$E 77 (40) RF7	        #34 E 55
Ra6		 14$E 90 (41) KG7	RESIGN   24*E359
Rxf6		 19$E 82 (42) RXF6	RESIGN   19*E 79
Qh5-e5		 15$E    (43) H5	RESIGN   14*E131
Qa8xf3		 14$E    (44) KG8	RESIGN   18*E182
Qf3xf6		 11$E    (45) H4	RESIGN   22*E 33
Qf6-g7 ++      	 11$E    (46)
   ::    ::    :: Kb ::
::    ::    ::    QW
   ::    ::    ::    ::
::    ::    QW    ::
   :: Pw Pw    ::    Pb
:: Pw ::    Pw    ::
   ::    ::    Pw    ::
::    ::    :: KW ::
JUN 14, AP9.5 BLACK'S MOVE 46.
NODES 23698 TERMINAL 17358 DYNAMIC 436 CAPTURE 6229
SALVAGE 1077 DUPLIC 910 RECLAIM 1487 BRANCH 683258
DUP_TREE 4047 REC_TREE 4190 BRAN_SCORED 533371

Commentary 25 Sept. 2009
The above file was produced by using ReadIris 7 on a TIFF image scanned from
research notebook #33 (Archive reference, 2004-58-35).  The game was played on
14 June 1986 at the 5th World computer chess championship, Cologne.

In this case it seems that Experimental mode means a search width of 9 moves.
based on the time taken and the paucity of periods '.' in the game score, the
search depth would be 9-ply and workspace size for positions was 600.

The first 4 moves were from the book.  The early opening phase 'E' are moves
1 to 8, and the Opening 'O' continues to move 16.  This is followed my Middle
Game 'M' to move 25, and then Endgame 'E'.  Awit was never behind in this game,
as indicated by the symbols '=', '+', and '$', and it thought Rex was always
in trouble, '<', '-', and '*'.

In analyzing this score sheet, there are two anomalies worth mentioning, and
more notation to explain.  First the notation:  At move 10, for example, we see
'49>O192'.  This means that the move Qxd5 was in position 4 of the move list and
that 9 moves were searched.  Awit is ahead '>' and spent 192 seconds on the
move.  For Rex we see '.32<O151'  The period '.' and 2 combine to say that 12
moves were searched for Rex and the move played was thought to be 3rd best.  The
fact that the '.' appears only twice supports the view that fewer than 10 moves
were searched for Black, and that some form of iterative widening was used on
moves 10 and 13 to temporally expand the width to 12 moves.  The hash mark '#'
that occurs at moves 17, 20, 21 etc. indicate places where the actual move made
by Rex was not examined at all by Awit.

The two anomalies occur at moves 9 and 16.  At move 9 the time recorded for
Rex's move is 1175 seconds, which is totally unreasonable.  This means that
telecommunication link to Awit's computer in Edmonton probably failed, since
Awit kept track to the apparent elapsed time.  Both Rex's and Awit's chess
clocks would have been officially stopped while the trouble was rectified.

The other anomaly at move 16 shows an elapsed time of 5 seconds for Rex's move.
Unreasonably short and inconsistent with the time taken on recent moves.  It
follows that Awit stopped running.  Probably the Amdahl system itself went down,
with power failure (thunder storms are common on the Canadian Prairie).
In that case the clocks would have been stopped while Awit was restarted.  Such
restarts can be problematic, especially if castling previleges have to be
preserved, or if there are draw by repetition sequences to maintain.  The next 
move by Awit, Na3, was thought to be good at the time, but it was ranked 3 in 
a search list of 9 moves that took 97 seconds '39+M 97'.

This game represents Awit's last win in an otherwise lacklustre performance,
quite the come down from its tie for second place in the 1983 New York
Tournament.  Interestingly, the actual hand written game score exists and a
scanned image appears elsewhere in these files.