Generated at 09/10/2020, 17:59 from 19234 logged games.
Rules
Representative game (in the sense of being of mean length). Wherever you see the 'representative game' referred to in later sections, this is it!
Move/select a spirit, then make a sacrifice (capture) with a man according to the rules for that spirit. So long as you keep using the same man to sacrifice, you may keep activating other spirits; however no spirit can be activated more than once per turn. Note that you may sacrifice your own pieces. Spirits move any distance in a straight line; they may move over occupied spaces, but must end in an empty space. Spirits move men as follows: Saltator (↷): Jump over Saltator, landing the same distance on the opposite side. Pulsor (→): Move one space directly away from Pulsor Tractor (⤙): Move one space directly towards Tractor Scylla (⟳): Move one space clockwise around Scylla Charybdis (⟲): Move one space anticlockwise around Charybdis
To activate a god, click on that god (or drag the god to its destination square if you want to move it). The god will now be highlighted in orange. Now choose a man which can move, and make a capture according the the god's rules; that man will change to a cross to indicate that he is now the high priest (only piece to move this turn). Continue activating gods and moving the high priest until you have either activated all gods, or choose to pass.
I feel like there's a puzzle here I couldn't solve, which is: what to do after a few moves when the middle's hollowed out and now all the pieces are stuck around the edge of the board, but the spirits are all in each other's way so you don't have many choices. Quantum Leap (a very similar "every move must capture" game on the surface) has tons of options at first, and the cooling down to a quiet board is gentle and slow. In Genius Loci, that cooldown seemed sudden and felt bad; suddenly I have few choices, or I have no choice at all but to destroy one of my own pieces, or I have a bunch of pieces that are already dead unless I execute a three-move sequence which depends on you making a mistake. So like I said, I wonder whether I missed some new opportunities that phase of the game provided. I bet that in about a year, I'll look at this, realize I haven't been tempted to play again, and lower my rating retroactively. But who knows! [Update: That happened in two months instead. But it could go up if I do feel the urge to play again. Who knows?]
russ
7
mrraow
10
My game :) This is a combinatorial game where player pieces are only able to move if doing so allows a sacrifice to one of the spirits (neutral pieces). Multiple combos abound, if you have the wits to find them! Design Notes (also in rulebook): This game is inspired by the "Activator Piece" discussions in the BGG abstract games forums. This is the purest game I could devise – the men do not move at all, unless an activator allows them to do so; the theme of spirits demanding sacrifices came naturally from the mechanics. Activator games seem to be inherently lacking in clarity. I actually started with around 10 powers, but quickly discarded half of them as being too confusing. Even then, the first incarnations of the clockwise/anticlockwise spirits acted at any distance; and had to be severely restricted to improve game play. Note you can now play against the AI.
Kaffedrake
4
This is frightfully opaque: each turn can involve from one up to ten piece moves, and properly evaluating a turn means having to read as many steps forward from each potential move sequence. A human just can't do that (although if you fool around and find a sequence that leaves you three pieces ahead, you can be [i]fairly[/i] sure it's to your advantage). Minor gripes: lack of graphical clarity and a vague sense of arbitrariness in the set of activator pieces.
fogus
5.5
Preliminary rating: An opaque abstract. More plays needed.
jmastill
7
nestorgames
9
really good :)
Kolomogorov Complexity Analysis
Size (bytes)
25994
Reference Size
10293
Ratio
2.53
Ai Ai calculates the size of the implementation, and compares it to the Ai Ai implementation of the simplest possible game (which just fills the board). Note that this estimate may include some graphics and heuristics code as well as the game logic. See the wikipedia entry for more details.
Playout Complexity Estimate
Playouts per second
11829.78 (84.53µs/playout)
Reference Size
290587.86 (3.44µs/playout)
Ratio (low is good)
24.56
Tavener complexity: the heat generated by playing every possible instance of a game with a perfectly efficient programme. Since this is not possible to calculate, Ai Ai calculates the number of random playouts per second and compares it to the fastest non-trivial Ai Ai game (Connect 4). This ratio gives a practical indication of how complex the game is. Combine this with the computational state space, and you can get an idea of how strong the default (MCTS-based) AI will be.
Playout/Search Speed
Label
Its/s
SD
Nodes/s
SD
Game length
SD
Random playout
8
0
547
22
69
5
search.UCB
12,203
574
66
6
search.UCT
12,299
291
66
5
Random: 10 second warmup for the hotspot compiler. 100 trials of 1000ms each.
Other: 100 playouts, means calculated over the first 5 moves only to avoid distortion due to speedup at end of game.
Mirroring Strategies
Rotation (Half turn) lost each game as expected.
Reflection (X axis) lost each game as expected.
Reflection (Y axis) lost each game as expected.
Copy last move lost each game as expected.
Mirroring strategies attempt to copy the previous move. On first move, they will attempt to play in the centre. If neither of these are possible, they will pick a random move. Each entry represents a different form of copying; direct copy, reflection in either the X or Y axis, half-turn rotation.
Win % By Player (Bias)
1: White win %
51.22±0.71
Includes draws = 50%
2: Black win %
48.78±0.71
Includes draws = 50%
Draw %
0.00
Percentage of games where all players draw.
Decisive %
100.00
Percentage of games with a single winner.
Samples
19234
Quantity of logged games played
Note: that win/loss statistics may vary depending on thinking time (horizon effect, etc.), bad heuristics, bugs, and other factors, so should be taken with a pinch of salt. (Given perfect play, any game of pure skill will always end in the same result.)
Note: Ai Ai differentiates between states where all players draw or win or lose; this is mostly to support cooperative games.
UCT Skill Chains
Match
AI
Strong Wins
Draws
Strong Losses
#Games
Strong Score
p1 Win%
Draw%
p2 Win%
Game Length
0
Random
1
UCT (its=2)
631
0
319
950
0.6336 <= 0.6642 <= 0.6935
51.16
0.00
48.84
68.09
4
UCT (its=5)
631
0
340
971
0.6193 <= 0.6498 <= 0.6792
48.09
0.00
51.91
67.95
5
UCT (its=14)
631
0
359
990
0.6069 <= 0.6374 <= 0.6667
49.70
0.00
50.30
67.67
6
UCT (its=37)
631
0
316
947
0.6357 <= 0.6663 <= 0.6956
52.27
0.00
47.73
66.92
7
UCT (its=100)
631
0
280
911
0.6619 <= 0.6926 <= 0.7217
54.56
0.00
45.44
66.76
8
UCT (its=273)
631
0
193
824
0.7357 <= 0.7658 <= 0.7934
53.52
0.00
46.48
66.12
9
UCT (its=742)
631
0
217
848
0.7137 <= 0.7441 <= 0.7723
50.59
0.00
49.41
66.29
10
UCT (its=742)
507
0
493
1000
0.4760 <= 0.5070 <= 0.5379
52.30
0.00
47.70
66.82
Search for levels ended: time limit reached.
Level of Play: Strong beats Weak 60% of the time (lower bound with 95% confidence).
Draw%, p1 win% and game length may give some indication of trends as AI strength increases.
1st Player Win Ratios by Playing Strength
This chart shows the win(green)/draw(black)/loss(red) percentages, as UCT play strength increases. Note that for most games, the top playing strength show here will be distinctly below human standard.
Complexity
Game length
67.28
Branching factor
3.54
Complexity
10^22.86
Based on game length and branching factor
Computational Complexity
10^8.01
Sample quality (100 best): 11.13
Samples
19234
Quantity of logged games played
Computational complexity (where present) is an estimate of the game tree reachable through actual play. For each game in turn, Ai Ai marks the positions reached in a hashtable, then counts the number of new moves added to the table. Once all moves are applied, it treats this sequence as a geometric progression and calculates the sum as n-> infinity.
Move Classification
Distinct actions
440
Number of distinct moves (e.g. "e4") regardless of position in game tree
Killer moves
28
A 'killer' move is selected by the AI more than 50% of the time Killers: g4-g2,d1-b3,e1-g1,a7-c7,d1-c2,g4-e6,a7-a5,g3-f4,g3-g2,e1-d1,a4-c2,b6-a6,e1-a5,g1-g3,b7-b5,a6-a4,d7-b7,c4-a4,c4-c2,d5-d7,c5-a5,c2-d1,e4-g4,e3-e1,d1-f1,e3-g3,c2-b3,f5-d7
Good moves
243
A good move is selected by the AI more than the average
Bad moves
197
A bad move is selected by the AI less than the average
Response distance
1.64
Mean distance between move and response; a low value relative to the board size may mean a game is tactical rather than strategic.
Samples
19234
Quantity of logged games played
Board Coverage
A mean of 95.91% of board locations were used per game.
Colour and size show the frequency of visits.
Game Length
Game length frequencies.
Mean
67.28
Mode
[69]
Median
68.0
Change in Material Per Turn
This chart is based on a single representative* playout, and gives a feel for the change in material over the course of a game. (* Representative in the sense that it is close to the mean length.)
Actions/turn
Table: branching factor per turn, based on a single representative* game. (* Representative in the sense that it is close to the mean game length.)
Action Types per Turn
This chart is based on a single representative* game, and gives a feel for the types of moves available throughout that game. (* Representative in the sense that it is close to the mean game length.)
This chart shows the best move value with respect to the active player; the orange line represents the value of doing nothing (null move).
The lead changed on 16% of the game turns. Ai Ai found 6 critical turns (turns with only one good option).
Position Heatmap
This chart shows the relative temperature of all moves each turn. Colour range: black (worst), red, orange(even), yellow, white(best).
Good/Effective moves
Measure
All players
Player 1
Player 2
Mean % of effective moves
86.66
89.13
84.41
Mean no. of effective moves
2.31
2.16
2.46
Effective game space
10^15.89
10^6.99
10^8.90
Mean % of good moves
36.51
18.00
53.44
Mean no. of good moves
1.27
0.94
1.57
Good move game space
10^8.80
10^3.64
10^5.16
These figures were calculated over a single game.
An effective move is one with score 0.1 of the best move (including the best move). -1 (loss) <= score <= 1 (win)
A good move has a score > 0. Note that when there are no good moves, an multiplier of 1 is used for the game space calculation.
Quality Measures
Measure
Value
Description
Hot turns
61.19%
A hot turn is one where making a move is better than doing nothing.
Momentum
8.96%
% of turns where a player improved their score.
Correction
23.88%
% of turns where the score headed back towards equality.
Depth
5.34%
Difference in evaluation between a short and long search.
Drama
7.40%
How much the winner was behind before their final victory.
Foulup Factor
47.76%
Moves that looked better than the best move after a short search.
Surprising turns
2.99%
Turns that looked bad after a short search, but good after a long one.
Last lead change
73.13%
Distance through game when the lead changed for the last time.
Decisiveness
41.79%
Distance from the result being known to the end of the game.
These figures were calculated over a single representative* game, and based on the measures of quality described in "Automatic Generation and Evaluation of Recombination Games" (Cameron Browne, 2007). (* Representative, in the sense that it is close to the mean game length.)
Unique Positions Reachable at Depth
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1
5
24
47
102
166
470
674
1667
4827
6228
15973
35963
68627
285738
1035333
2597867
10903458
Note: most games do not take board rotation and reflection into consideration. Multi-part turns could be treated as the same or different depth depending on the implementation. Counts to depth N include all moves reachable at lower depths. Inaccuracies may also exist due to hash collisions, but Ai Ai uses 64-bit hashes so these will be a very small fraction of a percentage point.
