Author |
Topic: Essay by Christian Freeling on inventing games (Read 539992 times) |
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #945 on: Jul 20th, 2012, 4:17pm » |
|
on Jul 18th, 2012, 4:01am, christianF wrote:In so much as learning to play Mu is like learning to ride a bike, we're learning. |
| At move 11 Purple unfortunately chooses the lesser starting point, 'lesser' for two reasons:- Red has the initiative, so you don't want him on your doorstep.
- The top-segment's first two neighboring cells are capacity-4 and -5 which allows one more explosion (and thus growth) on the second piece-move, compared to the chosen starting segment, that has two capacity-5 cells as its first neighbors.
Then, at move 18, Purple starts a quarrel over the extension of the wall, where his first priority should have been growth (R13-R11 followed by Q11-S11, blowing away through the concave section of the edge and inwards). Red gets a man on the wall and pulls up reinforcements [27, 28] while Purple tries to reorganize. Red's penultimate move is a trick shot . We've started a new one with 2x7 segments. Edit: Purple won. Due to a bug, the comment at Purple's last move doesn't show up in the applet (it does in the table with the move record though). It says: "Not likely, we've got equal forces, and any attacker will find defenders waiting. The cells where Purple could attack are fairly obvious, not to mention limited, and claiming the four cells is enough for Purple." In velox, the attacker can combine moves. In levis you can't. So the cells where Purple could attack are easy to determine. None of them allows a move that would result in an explosion in red territory, and that's what Purple would need at the very least, otherwise any attacker will simply be eliminated on arrival. So a 'grand attack' is out of the equation here. Nor is it needed. Purple has 40 cells, Red 42, but Purple wins the fight over the 4 'neutral' cells on the left. Red cannot provide reinforcements, while Purple can plug it ... well, not ad infinitum, but easily long enough to capture it. Red of course must defend, because Purple now occupies the section alone, making it his territory regardless of who is on the wall. After a few exchanges, Red will be out of ammunition.
|
« Last Edit: Jul 25th, 2012, 3:21am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #946 on: Jul 27th, 2012, 8:42am » |
|
Not everything about Mu was 'self-explanatory'. If it leads to dicrepancies, the best strategy is to find a solution that serves the spirit of the game. A bit of luck helps. In this particular case the low capacity of emerging cells of the Wall posed such a discrepancy. If cells of the Wall would behave like cells of the Commonwealth, then- a 3-cell straight or bend linear section could hold exactly one man, and 2- and 1-cell sections would not accept a man at all.
- the cell would sprout a new man on emergence.
I trust anyone to see the irony of sprouting new men and have them oscillate away on arrival. So the replacement of growth by the acceptance of an extra man cuts both ways. It makes getting men on the wall an art in itself (and not something that is granted by automatic growth), while ensuring that - barring a single cell section - the men will actually be able to stay on it instead of being blown away. It seems a fair exchange and it works as intended within the vision I had developed on the nature of the 'men on the Wall', after considering that aspect the day after the night before.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #947 on: Jul 27th, 2012, 9:58am » |
|
Another aspect worthy of consideration is turn-order balance. If Mu were played on a 'normal' board, which would imply symmetry, then the first player would have an advantage. So what's different here? The difference is that the initial speed of growth often outweighs the advantage of moving first. And it is the second player who may have the better options there, since he is the last player to add a segment to the board. Consider this: >>> two moves on >>> Note that the first two cells outside the starting segment are capacity-4 and -5 respectively. Now make that two capacity-5 cells, and growth already slows down a bit: >>> two moves on >>> Build the starting segment in any further, and initial growth will be ever slower. So both players will consider the factor 'speed', and 'isolating' the starting segment as much as possible will be one of the considerations. In that particular respect, the second player has the option to do that most effectively with his last segment. There are other considerations, like access to the center, the presence of concave low-capacity bends in the nearby edges, particular good places for a wall to emerge, the possibility of the opponent's starting segment being too close by, and other 'local peculiarities', but speed will always be important and the second player has the better options to achieve that. I have no clue how it will turn out eventually, just pointing at circumstances that make a more or less accurate evaluation of turn-order advantage less than trivial. Meanwhile we've started an 8-segments pp game.
|
« Last Edit: Jul 29th, 2012, 8:23am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #948 on: Aug 3rd, 2012, 12:55pm » |
|
Ok, that was fun. By now it may be clear to some that Mu levis is a game, and an enigmatic one at that. Moreover, it fits the traditional idea of a board game in that it can actually be played in the traditional fashion. As said before, you need- segments
- white draughtsmen for Virginity
- black draughtsmen for the wall
- colored chips for pieces
Mu velox is not a traditional board game in the above sense, but rather a representative of a new breed of board games that, for whichever reason, need an applet if they're to be played properly. That is: if they can be played properly at all . And that's what we're going to find out, now that we're a bit more familiar with levis. So here's a 8-segments pp Mu velox game.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #949 on: Aug 5th, 2012, 10:29am » |
|
Ed has aken care of yet another loose end in the ArenA. Here's the (hidden) background of the new Glass Bead Game applet, 2x5 pits and 2 collecting cups in the middle, and precise coordinates for every individual bead. Move notation hasn't changed, but the applet is now clickable. Here's the same example game as before, but in its new housing. Edit: Not to distract from the 8-segments pp Mu velox game though, which shows that velox tactics, though incalculable if the game were to be played at the traditional coffee table, are serving strategy well enough, and do not at all derail the game but rather amplify it's spirit.
|
« Last Edit: Aug 5th, 2012, 3:20pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #950 on: Aug 22nd, 2012, 12:27pm » |
|
While you're letting Mu sink in (or why else would so many read this stuff) you may enjoy another velox game, this time a 'base-10' one. Don't worry, it's like a rollercoaster, thrilling all the way, but it always keeps on track . Another thing:- The Mu applet now has an option to take back single steps, including single explosions, of moves that a player might try out. This feature is merely convenient in levis (where there aren't any 'move sequences' to begin with), but indispensable in velox.
Edit #2: Actually there's more indispensable than that. Ed will implement a triple 'forward/backward' option:- |< and >|: take back or move forward one step (single move or single explosion)
- |<< and >>|: take back or move forward one complete move
- |<<< and >>>|: go to begin or endposition
It will allow clicking through the seperate steps of any complete move, which is indispensable for analysis, at least in velox. I like velox for the incredible tricks one can manage, but investing an hour or so in figuring out an ingenious combination, only to see it flash by in seconds without being able to consider it step by step, isn't quite satisfactory. This feature may be a bit complicated to implement, so please bear with us .
|
« Last Edit: Aug 25th, 2012, 10:58am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #951 on: Aug 26th, 2012, 11:48am » |
|
There are those primary observations regarding strategy and tactics that one can make in virtually any new strategy game. In Mu velox the first game gave a rather blunt example of the danger of progressing blindly forwards. The theory behind it is that if two sections of the Commonwealth, controlled by different players, close in on each other, the virgin cells between them increasingly lose capacity, amplifying chainreaction sensitivity, so as a rule of thumb- either erect a wall yourself, or
- keep a safe distance
Here's another primary observation. Purple moves second and has a slower start in terms of 'growth per turn' because he has less low-capacity cells in the immediate vicinity to work with: Yet he has already won! Red could have progressed much further on move_23, but it would have meant getting too close (see the argument against that above), allowing Purple to eject a wall that he doesn't even have to occupy (though he can in some variants). The best Red has at move_23 is to hold back, but that's enough too for Purple: Red has dug his own grave in the initial positioning of the segments, and now he can rest in peace So this is a major strategic disaster on Red's behalf, and an important idea on Ed's account. The lay-out phase must be reconsidered, and the starting segment's position must be put before the speed of growth it may generate. The center may be slow on growth, but it's fast on controlling key areas. That's a whole new ballgame, and one that may make any turn-order advantage even more difficult to determine. P.S. Of course these new insights beg for immediate application so we've started a new one .
|
« Last Edit: Aug 26th, 2012, 12:43pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #952 on: Aug 30th, 2012, 10:55am » |
|
In new games, all players are beginners and there's no-one to explain strategy, so tactics and short time considerations dominate. In retrospect new insights may hardly deserve the name "paradigm shift" because the initial approach to the game was totally naive. This game was a clear demonstration of "position" over "speed" when it comes to negotiating a good starting segment. "Speed" derives from a low-capacity starting vicinity, in particular single segments at the end of 'tubes'. As it turned out however, you don't need much speed if you start in the center, provided you can block the opponent's tube or direct him to an unprospective area in terms of territory. Actually it's so obvious that the question "why didn't we see that before" presents itself rather emphatically. So what are the provisional variables now, regarding the quest for a good starting segment:- A position near to the center or at least with fast access to it, depending on the second point
- No presence of a nearby opponent's segment that would be particularly suited to block its development
- Low-capacity vicinity
Of course you can't have it all you way, but these are things to consider. Given the magnitude of the effects of a bad starting segment, it is clear that placing and clearing them is a highly strategic phase. This magnitude also dwarfes considerations about theoretical turn-order advantage, while still keeping in mind that it is the second player who can place the last segment. So how did that pan out in our current game? This is the situation after purple_15: ten placements and five removals each. Purple's trump card is #6, near the center and very well situated against Red's #1, 2, 3 and 4. Previously Purple had 'shielded' this segment from further extensions by placing the now empty segment to the SE of it. Red's segment #5 however, is very well situated against Purple #6, but its high-capacity profile works against it. Using the immediate adjacency to start with a capture could lead to a one- or two-men exchange, which would seem to favor Purple, because he has access to two capacity-3 cells for starters. Whether or not Red would start with a capture, the question is: can Purple grow fast enough to inflict some initial damage? If Red #5 can block Purple to the NW, Purple would have to go around the other side of the lake, through a single connection bottleneck. That he would get the SE section (the tube with Red #4) in the process, is hardly enough compensation if Red is fast enough around the topside, which would almost certainly be the case. So Purple's alternative is #9, or even #8, depending on Red's further removals. Update: Ed didn't like the looks of it and removed #5 so I got my prefered starting segment. Note: This is all velox, did I mention I like velox? Nevertheless, the consideration regarding segment placement and the removal of starting options would appear very similar in levis.
|
« Last Edit: Aug 31st, 2012, 8:11am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #953 on: Sep 2nd, 2012, 10:52am » |
|
Velox is a bit like sumo wrestling: if you're caught on the wrong foot, there's an immediate collapse. Red 27 was a death blow from which Purple coudn't recover. "Position over speed" would seem to need some nuance too. In the previous game Purple was close enough to block Red's tube-exit. In the latest one this wasn't the case, and Red's move_25 gave the emerging wall a wrong turn, from Purple's perspective. That was the strategic pivot, 'the wrong foot' for Purple, so to say. Characteristically, for better or worse, Mu is completely devoid of specific opening theory. Strategy is ruled by as yet immature considerations, tactics reveal themselves by example. Velox has a powerful positive feedback: more growth, more segments, more combinatory power, more growth. That too may be considered for better or worse. I'm all for deep strategy and sublte tactics in a well known and largely explored context, but every now and again a bit of opportunistic streetfighting is refreshing .
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #954 on: Sep 6th, 2012, 6:26am » |
|
Organic design is a rule in nature, and as far as animals go, there are sometimes tamed variants (dogs, cats, rabbits, homing pigeons, horses) and untamed ones (wolves, wild cats, hares, wood pigeons, horses). So it isn't quite out of the ordinary that an organic game would show a similar division. Mu_velox may appear unbalanced, judging from the games so far. I could argue that a beginner on a bike does likewise, but that doesn't quite cover it. Velox will as far as I can see always guarantee a large amount of sweeping victories. It's the suma wrestling syndrome: once off balance you tumble into the gorge. Whether or not the balance will in the long run favor the first or second player is a tough question. Mu_levis is much more a strategy game. Combinations aren't possible in a single turn and thus will meet with the intermediate replies of the opponent. So the resolution of the effects of action and reply is much higher. As is predictability: in Levis the result of one move, in terms of a subsequent chain reaction, is easy to 'read'. In Velox, the combinations that may be possible in a single turn, are not. Strategy must be based on experience and intuition, tactics are fierce and unforgiving as our last game less than subtly illustrated. Here's a new one, wish me luck .
|
« Last Edit: Sep 6th, 2012, 11:23am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #955 on: Sep 13th, 2012, 5:53am » |
|
The first cars looked like stage coaches without a horse. Likewise, the first online abstracts looked like boardgames without a physical board. I'm no expert at social interaction. I like my relative solitude and haven't touched a physical board in the last decade or so. So I won't comment on what is lost in online play, as opposed to over the board play. But something is won. Playing online is more convenient (unless of course one finds social interaction 'convenient') and you can do a few things that you can't do at over the board play. That being the case, there's no reason to somehow limit these options.- Mu_levis can be played over the board, provided you have the simple materials to do so. Playing online may be more convenient, but the lack of social interaction is somewhat more pronounced if played with more than two players: that may turn into a collusion infested affair, but it's great fun.
- Symple can be played over the board, but one would need special Go stones with a small dot on one side. One doesn't have that problem online.
- Sygo already raises a bigger hurdle if it is to be played over the board. It doesn't have any problem online.
- Mu_velox is the first hybrid of an abstract game and a pinball machine. It can only be played online.
The underlying thought is that abstract games may evolve eventually, like the 'stage coach without a horse', opening fields that were impossible to open before. I can't be anymore specific than a carmaker in 1912 or thereabouts could have been regarding the future of cars. But I'm glad to have been able to give you at least one clear representative of a new breed. P.S. Please consider the above metaphor as such - I'm not implying that say Go is an 'old car'. At the same time I'm not so sure about Chess and very sure about Checkers .
|
« Last Edit: Sep 13th, 2012, 7:10am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #956 on: Sep 16th, 2012, 9:20am » |
|
So Red won the last one, and the first one played in full awareness of the speed versus position dilemma regarding segment placement. Purple valued 'position' maybe a trifle to optimistically because it eventually had to outweigh a double disadvantage: a slower start while moving second. It was clear from the onset that he had to cut through the central string of lakes somewhere. That didn't quite work out, though Red was forced through a very narrow channel, that left him with no straight route to the bottomleft territory. If he hadn't had the one invasion point, a shooting pimple of sorts, to slow Purple down, he might easily have lost the race for it. The final position came after a double pass. Red has 79 cells, Purple 54, with a more or less corresponding count of forces. Red would naturally maybe rearrange a bit and pass. What if Purple had refused to pass? Red has no incentive to invade, nor a profitable way to do so. He must at least enter a 2-column that is pushed out of the pimple only to be captured. The captor next can move out of the line of fire. Invading with 4 or more means occupying 4 cells without making a capture. Lets put it a bit sharper: if red had no man on the wall he wouldn't be able to invade at all. So if Purple keeps moving, he can either pass or make the occasional move, but the game doesn't end. Purple can't invade himself, because he ends on one segment, facing a reply that can make use of at least three. That's a nogo. Of course this is silly, but what if the territory differs by only one or two cells, with about equal forces? That's one bordering on 'tournament rules', and who has the burden of proof. I'm still considering it - Go has had similar issues for as long as it exists, so there's hardly any hurry. Meanwhile the balance of velox still hangs in the balance, so to say, but this much is sure: if velox doesn't derail, then levis is balanced by implication. It's not by chance that we fight with the wild one to see if it can be tamed. So here we go again.
|
« Last Edit: Sep 16th, 2012, 9:57am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #957 on: Sep 18th, 2012, 6:21am » |
|
Maybe it's worth mentioning that the CodeCup 2013 test competitions have started.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #958 on: Sep 19th, 2012, 3:40am » |
|
Also: Ed has finished the AYU applet and this simple intruiging game by Luis Bolaños Mures can now be played at mindsports. 'Simple' refers to its structure - Ayu may well be considered a quintessential 'approach' game.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #959 on: Sep 22nd, 2012, 1:23pm » |
|
So here we go again ended in Sumo style, Red overrunning Purple. With the current opening protocol, the first player may arguably have have an advantage, though Purple probably chose the wrong strategy by investing in an early sidetrack, a segment that seemed promising for quick growth because of its low average capacity. But it took too long to get it started, and when it did, being trapped between the lakes interfered deplorably with speedy development. In case there should be any question about the current protocol as a vehicle to a balanced starting position, Mu can use a pie rule protocol instead. Since it is equally easy to make a starting position in which Red loses than one in which Purple loses, it is also possible to make a balanced position, in which Red's advantage of moving first has been counterbalanced by Purple's better position. So if ...- the first player makes a starting position, and ...
- the second player makes a choice to play either Red or Purple
then the first player bears full responsibility for the measure of balance in the starting position, while the second bears full resposibility for his choice. This being said, we'll leave Velox for the time being, but if any of the obviously large number of viewers likes to try, we've recently lifted the ban on playing . We're entering the quieter realm of strategy again, see the camera follow us in slo-mo .
|
|
IP Logged |
|
|
|
|