Author |
Topic: Essay by Christian Freeling on inventing games (Read 539991 times) |
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1095 on: Dec 22nd, 2012, 12:51pm » |
|
The final CodeCup Challenge 2013 test competition has been completed. Congrats Abdessamad Elkasimi! Of the 432 games that were played, 419 came to a legal conclusion. 217 (51.8%%) were won by white, 202 (48.2%) were won by black. The top dogs are still the same, but Sergey Nefedov's program skifi_025 is catching up. * Note that winners always get a 100-points bonus, obviously for some reason, but I'm still in the dark about it.
|
« Last Edit: Dec 23rd, 2012, 4:56am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1096 on: Dec 23rd, 2012, 7:54am » |
|
on Dec 2nd, 2012, 1:15pm, christianF wrote:I'm an old fox and I wasn't hunting or anything, but while discussing Symple Hex with Benedikt, I picked up a faint presence of a game. Ignoring it would have been harder than following the trail, so I followed a trail that led to a new move protocol which in turn pointed to 'territory' as the most likely object, and on thursday night, in the magic span between going to bed and falling asleep, it came together without a wrinkle. It's called Triccs and there's also a square version that I'm not quite sure about because of the usual straight/diagonal issues. I'm sticking my neck out here, because I've not played it yet, but I have a fairly good feeling about it. An applet is high on the priority list. |
| So we have an applet now, and we're playing our first game. No Sandboxes involved. As already shown in a previous post, there's no parity of any kind. can lead to In fact there are two more puzzles: what is the minimum number of stones needed to get to the second stage, and what is the maximum. That's quite a difference and it lends strategical and tactical flexibility to the game. Not bad for a recreational funny Here Black's last move is indicated (as well as White's subsequent first placement). Black could have prevented White from playing A1 by playing at B1 or A2 with his second placement. But that would have given White the start of Phase 2. Black would rather have that himself. Edit: I overlooked C8 - now I'll have to take the 'last first placement' myself. Note that a white placement at A1 gives White 11 options for his second placement. Therefore the 'safety mechanism' - i.e. "if the player to move cannot make the second placement ... - is probably not even necessary on a regular board, despite the base-3 example where it is necessary. A base-3 board has 12 edge-cells and an inner area of 7, one more than half the edge. A base-6 board has 30 edge-cells and an inner area of 61, more than twice the edge. So (though not proven) a base-6 example seems highly unlikely because there's so much more 'room' in the inner area, relatively speaking. Note also that the move protocol does what it was supposed to do: give a practical intermixed position to start the second phase. As such it has, again, significance as a mutator, a generic principle or mechanism that is applicable to more than one game.
|
« Last Edit: Dec 23rd, 2012, 10:50am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1097 on: Dec 23rd, 2012, 11:16am » |
|
Something must be seriously wrong with Scware because the regular help of my critics, in particular those who acknowledge their own restrictions by admiring them, as yet fails to materialize .
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1098 on: Dec 24th, 2012, 12:56pm » |
|
For the last few months I've been playing a Symple game against Cly Ring in which he gives elaborate comments on the moves, in full recognition of the fact that we're all beginners. My comments are limited because I find commenting on the reasons much harder than considering them. I try to find my way by playing a lot, Cly Ring by analyzing a lot, and since he's winning, it's 1-0 for analysis . Cly Ring - CF (running)
|
« Last Edit: Dec 24th, 2012, 1:43pm by christianF » |
IP Logged |
|
|
|
clyring
Forum Guru
Arimaa player #6218
Gender:
Posts: 362
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1099 on: Dec 24th, 2012, 5:42pm » |
|
It's really more like 3-1 for analysis taking into account the other games we've played: clyring - CF 0-1 B+79 (Freeling showing a beginner no mercy... ) clyring - CF 1-0 W+9 (A thematic and closely contested clash: connections vs territory.) CF - clyring 0-1 B+9 (I seriously misplay the opening and blunder away a key connection in the middlegame, but manage to stage a dramatic comeback when Freeling plays too passively in the endgame.) clyring - CF in progress (EDIT: Now 1-0 W+Res.) Unfortunately there is not so much commentary included with the other games. There is, however, a discussion on page 61 of this thread about the second game's final stages.
|
« Last Edit: Jan 24th, 2013, 1:14am by clyring » |
IP Logged |
I administer the Endless Endgame Event (EEE). Players welcome!
|
|
|
hyperpape
Forum Guru
Arimaa player #7113
Gender:
Posts: 80
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1100 on: Dec 24th, 2012, 6:34pm » |
|
Does this forum have hide tags? I have just now challenged you to a game, clyring.
|
|
IP Logged |
|
|
|
clyring
Forum Guru
Arimaa player #6218
Gender:
Posts: 362
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1101 on: Dec 26th, 2012, 12:33am » |
|
This forum does not have hide or spoiler tags implemented, so if you are trying to ask me to play a Malkovich game, I guess you will have to start a thread at L192 for it.
|
|
IP Logged |
I administer the Endless Endgame Event (EEE). Players welcome!
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1102 on: Dec 26th, 2012, 5:43am » |
|
The CodeCup format is base-15 penalty-6 and Abdessamad won the final test round so in terms of bots you can't expect any better opponent. I've just been challenged and I'll try to give it my best shot. Cly Ring is putting dents in my confidence Playing white the bot got rather lost in the woods, starting way too many groups. Abdessamad resigned in the second game too. Remarkably, to yours truly it would seem that B_ot (Bertrand Lunderer at the CodeCup) is stronger. Other bots obviously have more trouble with the seedhappy nature of Abdessamad's bot.
|
« Last Edit: Dec 27th, 2012, 7:26am by christianF » |
IP Logged |
|
|
|
hyperpape
Forum Guru
Arimaa player #7113
Gender:
Posts: 80
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1103 on: Dec 28th, 2012, 3:05pm » |
|
You're free to comment or not clyring, I'd just had the thought to share some of my thoughts as the game progressed.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1104 on: Dec 29th, 2012, 10:11am » |
|
Sometimes solutions are implemented before they are found . In multi move games like Symple and Yodd, the mindsports applet since fairly recently dims the 'submit' button as long as the position allows no legal submitting of a move. In Triccs however it allowed submitting a move in the first stage after a player's first placement, although the rules for no apparant reason (except maybe the inventor's slow grasp) required the second placement to be made too. That 'fault' is the solution to the problem posed by a position in which the player to move can make the first, but not the second placement. Here's one such position on a base-4 board, and it wasn't all that difficult to find either: The second (9) placement is not possible. The whole problem disappears if the second placement is optional, and there's no reason why it shouldn't be. Triccs is a hot game where every stone counts and where a player can always find a good place for it, especially in the opening phase. So the second placement has been made optional. As a bonus this brings the first phase of the game in line with the second phase, where placement is compusory and movement optional. The generic angle The main interest of Triccs may be its opening protocol, a generic protocol that leads to an opening lay-out for a second stage of a game in which say a third to half of the board is filled in an evenly divided way. What the game should be depends on what you (or I) would have in mind. For now Triccs is a good way to preserve the protocol, but it's more than likely that better games that would employ it, are hidden out there in the foggy realm of the as yet uninvented. As for Triccs itself, the first game shows some curious and interesting tactics, but it is clearly a game for which the phrase "coming with the territory" should have been invented, if it hadn't been already. There's room for big mistakes, but with intelligent play the advantages to gain are small, and easily lost. It feels negotiating modest margins without any large scale drama, yet it's tricky so the name seems appropriate enough .
|
« Last Edit: Dec 29th, 2012, 12:36pm by christianF » |
IP Logged |
|
|
|
hyperpape
Forum Guru
Arimaa player #7113
Gender:
Posts: 80
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1105 on: Dec 29th, 2012, 3:00pm » |
|
As Christian already knows, there's a small ambiguity in the Symple rules. Suppose you have exactly two groups, one of which has a single empty adjoining point, which is next to the second group, which has many empty adjoining points. The question is, must you play the connecting move? http://mindsports.nl/cgi-bin/Arena/Serve.cgi?file=Symple1356191511.html: my game with Jos Dekker features that situation at move 24. Christian has said that you don't, which should probably be reflected in the rules.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1106 on: Dec 29th, 2012, 3:11pm » |
|
on Dec 29th, 2012, 3:00pm, hyperpape wrote:As Christian already knows, there's a small ambiguity in the Symple rules. Suppose you have exactly two groups, one of which has a single empty adjoining point, which is next to the second group, which has many empty adjoining points. The question is, must you play the connecting move? http://mindsports.nl/cgi-bin/Arena/Serve.cgi?file=Symple1356191511.html: my game with Jos Dekker features that situation at move 24. Christian has said that you don't, which should probably be reflected in the rules. |
| The rules say: Quote:The game starts on an empty board. White moves first. Moving is compulsory. On his turn a player must either:- Grow all possible groups by one stone, or ...
- ... put a stone on a vacant cell, not connected to a like colored group, thereby creating a new group.
A stone connecting two or more different groups is considered to have grown all of them. A player may grow at groups as they exist at the beginning of his turn, and no such group may grow more than one stone in that particular turn. [examples ] Note: a player is free to choose the order of growth. A group is not allowed to grow if a stone at any one of its liberties, would at the same time grow another group that did already grow in the same turn. This is a regular endgame occurence and makes the order in which to grow part of the tactical considerations. |
| This note is included to clarify of what "all possible groups" means. This isn't something that can be determined a priori: it depends on the group configuration and order of placements. It includes the particular case in which a group cannot grow at all because all its liberties are shared with a group or groups that have already grown (as in your game with Jos).
|
« Last Edit: Dec 29th, 2012, 3:44pm by christianF » |
IP Logged |
|
|
|
hyperpape
Forum Guru
Arimaa player #7113
Gender:
Posts: 80
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1107 on: Dec 29th, 2012, 7:42pm » |
|
Was that there the whole time? I looked at the rules when the situation arose in my game--if I missed that, my apologies.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1108 on: Dec 30th, 2012, 4:34am » |
|
on Dec 29th, 2012, 7:42pm, hyperpape wrote:Was that there the whole time? I looked at the rules when the situation arose in my game--if I missed that, my apologies. |
| Yes it was . The procedure emerged when Symple switched to compulsory placement. Growing 'every group' had to merge with the rule that no group may grow more than one stone in a turn. The former had to be subject to the latter, so growing at group 'A' made that the vacancies it shared with group 'B' would be off limits, and vice versa. Hence the order of placement would determine whether or not a particular group would be subject to growth. The implication is that the order of moves is subject to tactical considerations. P.S. Since any straw of a prospect of an ambiguity in the Symple rules is predictably being jumped at by the troubled high-priest of the church of cyclophobia & hard finitude, who is determined to not understand anything about Symple (and getting better at it all the time), here are the rules at Sensei. In the example diagram you see the situation illustrated bottom-right.
|
« Last Edit: Dec 30th, 2012, 10:38am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1109 on: Jan 1st, 2013, 6:41am » |
|
May I wish all of you a very happy, prosperous, creative and less than problematic 2013 . When I started playing Symple after the introduction of compulsory placement, Marcel Vlastuin's bot gave me a hard time. In the last CodeCup test round it ended at place 8 with a ranking that is slightly higher than half the ranking of Abdessamad's bot. Now I usually win against the latter, but Abdessamad recently upgraded his program, and there's a shift in priorities in it's evaluation. It suddenly started playing closer to the edges in the opening, while starting two groups in the center. That wasn't so bad a plan and it took me by surprise: CF - Abdessamad (game 1): 0 - 1 Here's the position after Black_10. I'm White. I took the turn order bait at move 4 and started growing at move 9, which leaves me with a group less. That was wrong because Black seeds another single, more or less forcing me to do likewise to not get even further behind in growing options. But I got squeezed between Black's strong center and his low-running edge groups. I was keenly aware that something had changed in the program's approach, so my next game with white I was ready to adapt. Again I took the turn order bait no sooner than at move 5. That's rather late (I think) but Abdessamad's program is fairly 'seed happy'. But different from the previous game, I decided not to accept a group less in the growing stage. That means that Black would have to grow first. CF - Abdessamad (game 2): 1 - 0 Here's the position after Black_10. Both have 9 groups. Black dominates the bottom side, White has secured the left side and has put an important wedge in the center, where Black again has followed its 'two-groups' strategy. I decided to not dispute Black's bottom area connectivity, but to split him elsewhere as much as possible. It worked. Humans versus Bots The bots started out better than humans, but by now the tables seem turned. I'm not particularly intimidated by the best of them and fairly confident I can win most games. Meanwhile I'm in danger of being overtaken left and right by the likes of Jos Dekker, Hyperpape and ClyRing. So the game develops in an interesting matter.
|
|
IP Logged |
|
|
|
|