Author |
Topic: Essay by Christian Freeling on inventing games (Read 522050 times) |
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender: 
Posts: 804
|
 |
Re: Essay by Christian Freeling on inventing games
« Reply #1110 on: Jan 2nd, 2013, 1:54pm » |
|
I'm a sucker for first sentences- All children, but one, grow up.
- God made Man, like Himself, lonely.
- If I had cared to live I would have died.
- Behalve de man die de Sarphatistraat de mooiste plek van Europa vond, heb ik nooit een wonderlijker kerel gekend dan de uitvreter.
to cite a few.
Now I came over this one- I’ve fallen asleep thinking about abstract games every night for more than a decade, and I’ve designed several hundred such games, the rules for which fill out bundles of notebooks I keep in my garage.
I think it's brilliant, mystery pervaded, suggestive! Why, of all things, abstact games? Why the garage? Why every night?Doesn't that lead to deplorable interference, occasionally? We're obviously about to meet an interesting case of 'method to the madness' here. I can't wait to see the whole novel pubished. It might bring the author a lot more than inventing abstract games  .
|
| « Last Edit: Jan 11th, 2013, 1:30pm by christianF » |
 IP Logged |
|
|
|
SpeedRazor
Forum Guru
Skepticism is the Immune System of Science
Gender:
Posts: 72
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1111 on: Jan 4th, 2013, 6:17am » |
|
• Man was evolved from an ape-like creature.
• God created Man in his own image.
Where am I going with this?
The Universe is Great  Are abstract games somehow not as 'good' as the sometimes more fun, but often a little (or a lot), more arbitrary 'popular' games which may use dice, rolling bones, etc?
I say no Christian! Spend your next Century coming up with more abstracts, perfecting your myriad masterpieces!, or just ... critiquing, the entire genre. You are an Icon, Mr Freeling! Never stop!
Happy New Year, Christian!
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1112 on: Jan 4th, 2013, 7:27am » |
|
on Jan 4th, 2013, 6:17am, SpeedRazor wrote:Are abstract games somehow not as 'good' as the sometimes more fun, but often a little (or a lot) more arbitrary 'popular' games which may use dice, rolling bones, etc?
I say no Christian! Spend your next Century coming up with more abstracts, perfecting your myriad masterpieces!, or just ... critiquing, the entire genre. You are an Icon, Mr Freeling! Never stop!
Happy New Year, Christian! |
|
Thank you and the same to you. However, my next century will for the larger part be subject to the motto "I don't think, therefore I am not", so I might not be able to oblige there .
Regarding your observations, popularity is an issue that many consider important. There's a lively discussion going on at Nick's Blog in response of his addressing the subject once again. If I were a cynic, I'd have no difficulty noticing that what it is, isn't nearly as interesting as how well it would sell. But I'm not a cynic .
|
|
IP Logged |
|
|
|
NickBentley
Forum Senior Member
Arimaa player #6704
Gender:
Posts: 44
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1113 on: Jan 4th, 2013, 1:00pm » |
|
on Jan 2nd, 2013, 1:54pm, christianF wrote:
Now I came over this one- I’ve fallen asleep thinking about abstract games every night for more than a decade, and I’ve designed several hundred such games, the rules for which fill out bundles of notebooks I keep in my garage.
I think it's brilliant, mystery pervaded, suggestive! Why, of all things, abstact games? Why the garage? Why every night?Doesn't that lead to deplorable interference, occasionally? We're obviously about to meet an interesting case of 'method to the madness' here. I can't wait to see the whole novel pubished. It might bring the author a lot more than inventing abstract games . |
|
!!!
Unfortunately, nice sentences for me only happen in moments of rare dumb luck, so there will be no novel. And yes it does lead to deplorable interference. It's also sad that I'm not a better game designer, given this problem.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1115 on: Jan 4th, 2013, 1:49pm » |
|
on Jan 4th, 2013, 1:04pm, NickBentley wrote:
I glanced through it, and I'm glad that it's not a subject I'm very interested in, because chasing commercial success seems like a recipe for frustration. Nor does it mean much, if games like Abalone can have it. I believe good games at first need a long time to emerge and then a much longer time to submerge, but without envisioning any 'centuries' or 'universe' crap. Even Chess in its current form will come to an end sometime, as did its predecessors. Commercial success has little to do with it, as far as I can see. The public at large will usually go for the tactical game. Strategy games imo. reveal their secrets to slowly to have any immediate appeal.
|
| « Last Edit: Jan 5th, 2013, 4:26am by christianF » |
IP Logged |
|
|
|
NickBentley
Forum Senior Member
Arimaa player #6704
Gender:
Posts: 44
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1116 on: Jan 4th, 2013, 2:16pm » |
|
My interest comes mainly from a desire to make stuff swirling around in my head better match the way I spend my days. It's uncomfortable having to fight off these impossibly persistent thoughts so that I can focus on doing something entirely different for pay each day.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1117 on: Jan 4th, 2013, 2:39pm » |
|
on Jan 4th, 2013, 2:16pm, NickBentley wrote:| My interest comes mainly from a desire to make stuff swirling around in my head better match the way I spend my days. It's uncomfortable having to fight off these impossibly persistent thoughts so that I can focus on doing something entirely different for pay each day. |
|
My way to handle thoughts is to consider them like clouds, not important, transitory, only worth following if there's something interesting about them (which usually isn't the case). Admittedly this is only possible because I don't have to work for a living .
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1118 on: Jan 6th, 2013, 5:42am » |
|
This was the cause of a power outage at some 22.000 households, mine among them, that started on Saturday Jan. 5th at 3.20 pm local time and lasted for some ten hours. With no electricity, no heating and the mobile network down, we were literally in the dark about what was going on. Fortunately we still had gas so after a couple of hours, when the temperature in Kobus' room became somewhat critical, I could put some buckets of warm water around his favorite spot. Not that I felt completely comfortable doing so, in the dark, knowing that his infrared sensors could follow my movements far better than I could follow his.
|
| « Last Edit: Jan 6th, 2013, 8:33am by christianF » |
IP Logged |
|
|
|
browni3141
Forum Guru
Arimaa player #7014
Gender:
Posts: 384
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1119 on: Jan 6th, 2013, 4:54pm » |
|
on Jan 6th, 2013, 5:42am, christianF wrote:
This was the cause of a power outage at some 22.000 households, mine among them, that started on Saturday Jan. 5th at 3.20 pm local time and lasted for some ten hours. With no electricity, no heating and the mobile network down, we were literally in the dark about what was going on. Fortunately we still had gas so after a couple of hours, when the temperature in Kobus' room became somewhat critical, I could put some buckets of warm water around his favorite spot. Not that I felt completely comfortable doing so, in the dark, knowing that his infrared sensors could follow my movements far better than I could follow his. |
|
We had a nasty power outage here in Michigan recently also. It lasted over two days and affected around 50,000 people served by Great Lakes Energy. I don't know about other companies. It was caused by a bad winter storm that caused a lot of damage. Luckily I don't have any cold blooded friends to take care of, but my grandmother lives downstairs.
We stayed warm by running our gas stove.
|
|
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1120 on: Jan 8th, 2013, 3:18pm » |
|
A ripple of creativity followed by a ripple of lethargy. I'm so bored that I even picked up posting at rga. Tickle me  .
...
...
...
(A couple of days later)
Ok, that was fun, my lethargy is waning, tomorrow the final round of the CodeCup takes place. Results will pour in during the evening, local time (GMT +1). Stand by.
|
| « Last Edit: Jan 11th, 2013, 6:21am by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1121 on: Jan 11th, 2013, 6:48am » |
|
We've put Luis Bolaños Mures' game Ayu in the ArenA.
Why?
I've played Ayu several times now, deplorably losing to the inventor. My initial impression of the game has only been strengthened. Ayu is quintessential (in that all rules are necessary and sufficient) and highly original. Tactics and strategy are intertwined in a way that makes you feel clarity can only come with experience, but also that it will come with experience.
Ayu suffers from the strategy game syndrome: it requires an investment on the player's side before any reward (in terms of growing insight and improved play) can be harvested. It's not a game you can play 'on the side', at least not at any significant level.
|
| « Last Edit: Jan 11th, 2013, 11:42am by christianF » |
IP Logged |
|
|
|
hyperpape
Forum Guru
Arimaa player #7113
Gender:
Posts: 80
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1122 on: Jan 11th, 2013, 6:41pm » |
|
A few thoughts on symple, based on recent play. My game against Christian illustrates them well, I think (http://www.mindsports.nl/cgi-bin/Arena/Serve.cgi?file=Symple1357161278.h tml).
First, I think that Luis's estimate of the proper number of groups to start in the beginning is not accurate. Luis's estimate would work if the only way that you could control space on the board was by filling it, and if all groups filled space at the same rate throughout the game. Neither assumption is reasonable.
First, groups can control space by making it in to territory (as a go player would see it) as well as by filling it. My own player, as befits a go player, focuses heavily on territory, so this game shows that well. A corollary of this point is that you can effectively force your opponent to grow by threatening the split of territory. A major part of maintaining the balance of territory is to keep access to the sides. Thus, in the corner, it's a huge threat to threaten to play under another stone on both sides.
. . . .
. O X .
. X X .
. . . .
(great for Black).
The second point is that particular groups may be forced to connect, and once they connect, they lose half their growing potential. This is an argument for a higher group penalty (>= 8 ). Indeed, I think Symple with penalty 4 is just a bad idea.
In any case, those theoretical points aside, there's a lot related to the endgame in this game. I'm not confident either of us played correctly, but there's a lot going on.
** Move 20
After white plays 21, it will be time for a detailed count and endgame. Now that I know that he isn't connected at O9, it will be substantially easier.
J16 is a really appealing point, but placing at G11 or H11 is also quite a nice idea.
** Move 21
Decided to defer the analysis, and just take J16 and A13. I had a hunch this would be optimal regardless of the analysis (which gets simpler with every move taken towards the endgame), so might as well wait and see. Next move should almost certainly be J17 or G11/H11.
** Move 22
*** Analysis
The score is 66-38 at the moment. White should make three connections, and Black should make two, turning that into 96-58.
Black territory 20 + 14 + 16 + 9 + 12 = 71
White territory 20 + 6 = 26
Score of 122-129 counting the territories, but not contested central areas. That's clearly close enough that we have to consider both how those territories will divide, who will be forced to make placements.
Both sides have largest groups with 20 internal spaces. In White's favor, his group has 9 dame, but in its disfavor, it is currently three disconnected groups, so it will consume those dame faster. Furthermore, Black has several internal points that can be played to slow filling. White has S10, M19, G11 and perhaps M10.
Combinatorially, it's too difficult for me to calculate exactly how things go, but given the closeness of the endgame, I believe playing at G11 will pay off by helping force White to play in my territory later. The cost is 10 points, plus up to 4 dame, minus 6 (3*2) points I'll take from White's territory.
** Move 23
I am rather confident that White's R10 is dominated by R11. It may not matter on this board, because White may want to fill more than two more times now, but better to preserve the internal placement at S10--it would probably not be good for Black to take that point with penalty of 10, the G11 group on the board, etc.
** Move 26
Currently, each turn I spend expanding existing groups gives me two dame and deprives Christian of one. With penalty 10, that means it could be up to a 7 point gain to play within my own territory if it later forces him to make a placement in mine. So it probably would have been better to play within my own territory than grab a few dame this turn.
** Move 27
It's hard to decide the best location for this move. The lower left is shapeless, so a white move at B3 would take a lot of my interior space.
** Move 33
Now, the largest interior area for Black is in the lower-middle, with 11 spaces. White has 10 spaces on the top, and Black has at least two opportunities to make placements in his own territory.
|
| « Last Edit: Jan 11th, 2013, 6:43pm by hyperpape » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1123 on: Jan 12th, 2013, 8:43am » |
|
on Jan 11th, 2013, 6:41pm, hyperpape wrote:
I'll add some comments later Hyper, at the moment (2.30 pm GMT) more than half of the games at the CodeCup Challenge have been played.
It's not that lonely at the top anymore it seems.
4.30 pm GMT:
Final results after 28 rounds (614 games)
1 Abdessamad ELKASIMI (31,MA) --- dyfficult9 --- C --- 7276
2 Dan Banica (25,RO) --- Easi1 --- Java --- 7054
3 Shadeel Han (23,TW) --- newcomer --- C++ --- 6936
4 Remco Bloemen (25,NL) --- 5m --- C++ --- 6873
5 Bertrand Lunderer (30,FR) --- Engine --- Java --- 6703
You can play against the #s 1 and 5 at mindsports.nl. Bertrand Lunderer's 'Engine' plays under the name B_ot.
Here are the games between these five:
Abdessamad ELKASIMI - Dan Banica (89-70 white win)
Dan Banica - Abdessamad ELKASIMI (75-78 black win)
Abdessamad ELKASIMI - Shadeel Han (85-80 white win)
Shadeel Han - Abdessamad ELKASIMI (76-77 black win)
Abdessamad ELKASIMI - Remco Bloemen (56-91 black win)
Remco Bloemen - Abdessamad ELKASIMI (91-44 white win)
Abdessamad ELKASIMI - Bertrand Lunderer (76-89 black win)
Bertrand Lunderer - Abdessamad ELKASIMI (50-67 black win)
Dan Banica - Shadeel Han (64-80 black win)
Shadeel Han - Dan Banica (91-62 white win)
Dan Banica - Remco Bloemen (66-81 black win)
Remco Bloemen - Dan Banica (88-71 white win)
Dan Banica - Bertrand Lunderer (87-48 white win)
Bertrand Lunderer - Dan Banica (63-84 black win)
Shadeel Han - Remco Bloemen (92-73 white win)
Remco Bloemen - Shadeel Han (67-86 black win)
Shadeel Han - Bertrand Lunderer (67-86 black win)
Bertrand Lunderer - Shadeel Han (73-62 white win)
Remco Bloemen - Bertrand Lunderer (87-72 white win)
Bertrand Lunderer - Remco Bloemen (80-91 black win)
|
| « Last Edit: Jan 12th, 2013, 12:29pm by christianF » |
IP Logged |
|
|
|
christianF
Forum Moderator
Forum Guru
Arimaa player #4019
Gender:
Posts: 804
|
|
Re: Essay by Christian Freeling on inventing games
« Reply #1124 on: Jan 14th, 2013, 8:06am » |
|
on Jan 11th, 2013, 6:41pm, hyperpape wrote:A few thoughts on symple, based on recent play. My game against Christian illustrates them well, I think.
First, I think that Luis's estimate of the proper number of groups to start in the beginning is not accurate. Luis's estimate would work if the only way that you could control space on the board was by filling it, and if all groups filled space at the same rate throughout the game. Neither assumption is reasonable.
First, groups can control space by making it in to territory (as a go player would see it) as well as by filling it.
...
The second point is that particular groups may be forced to connect, and once they connect, they lose half their growing potential. This is an argument for a higher group penalty (>= 8 ). Indeed, I think Symple with penalty 4 is just a bad idea. |
|
You're probably right that Luis' reasoning is more of a signpost, as he himself confirms at the end of his post at RGA. For clarity I've copied his reasoning:
Quote:Let's assume that:
p is the penalty,
n is the total number of intersections on the board (for a 15x15 board, n = 225), and
g is the number of groups for each player that ensures the fastest filling of the whole board.
Christian has explained on MindSports that:
g = sqrt(n/2)
So, on a 15x15 board: g = sqrt(225/2) = 10.6
In practice, g = 11.
Let's now see how this relates to strategic planning at two opposite stages of the game:- Right after g groups have been placed on the board, each group has a potential growth of g. Connecting two groups costs the player g - p points (it substracts him g potential points and gives him p points). If p = g, we can consider that connecting two groups has no raw effect on the score. This would mean the decision to connect or not to connect is hardest at this stage of the game, since it has to be entirely made under different considerations than its raw effect on the score.
- Close to the end of the game, each group has a potential growth close to 0. If p = 0, we can consider that connecting two groups has no raw effect on the score. As before, this would mean the decision to connect or not to connect is hardest at this stage of the game.
It seems convenient to try and place the hardest decision point midway between the start and the end of the game so that there's a progressive increase in tension up to that point, along with the increasing complexitiy of the board position, and a progressive decrease from that point onwards as the game tree approaches its end. This is achieved by setting p midway between g and 0, i.e.:
p = g/2 = sqrt(n/2)/2.
On a 15x15 board, since g = 10.6, p should be 10.6 / 2 = 5.3 = 6 (rounding up to the closest even number). Actually, for all boards between 15x15 and 19x19, the rounding up always gives 6 as the ideal penalty, which is nice.
This is all speculation on my part, but I think it makes some sense. |
|
Part of the arbitrariness of the argument is in the placement of the 'hardest decision point' midway the game. I can see what is meant by it, but believe me, there are hard decision points throughout a game, in particular in the opening, where the basis for the later development of a game is laid. I also agree with your positional arguments, the importance of corners, the "Go aspects" of the game, and the upsetting role of tactics early on.
Regarding the value of "p" I must say that "base-15 p-6" worked very well in the CodeCup, that "base-19 p-10" is my personal favorite and that "p-4" is included in the applet to serve maybe a base-11 game. Penalties "2" and "0" are included because Ed hasn't removed them yet .
on Jan 11th, 2013, 6:41pm, hyperpape wrote:In any case, those theoretical points aside, there's a lot related to the endgame in this game. I'm not confident either of us played correctly, but there's a lot going on.
** Move 20
After white plays 21, it will be time for a detailed count and endgame. Now that I know that he isn't connected at O9, it will be substantially easier.
I16 is a really appealing point, but placing at G11 or H11 is also quite a nice idea. |
|
I considered O9 and you're probably right I should have connected. But an invasion at S8 didn't look too good either, and it was one or the other.
on Jan 11th, 2013, 6:41pm, hyperpape wrote:** Move 21
Decided to defer the analysis, and just take I16 and A13. I had a hunch this would be optimal regardless of the analysis (which gets simpler with every move taken towards the endgame), so might as well wait and see. Next move should almost certainly be J17 or G11/H11. |
|
Deferring analysis was a safe bet by that time. You've got all corners and quite a few dame points in the center. Add two certain connections and you're clearly in the lead already.
on Jan 11th, 2013, 6:41pm, hyperpape wrote:** Move 22
*** Analysis
The score is 66-38 at the moment. White should make three connections, and Black should make two, turning that into 96-58.
Black territory 20 + 14 + 16 + 9 + 12 = 71
White territory 20 + 6 = 26
Score of 122-129 counting the territories, but not contested central areas. That's clearly close enough that we have to consider both how those territories will divide, who will be forced to make placements.
Both sides have largest groups with 20 internal spaces. In White's favor, his group has 9 dame, but in its disfavor, it is currently three disconnected groups, so it will consume those dame faster. Furthermore, Black has several internal points that can be played to slow filling. White has S10, M19, G11 and perhaps M10.
Combinatorially, it's too difficult for me to calculate exactly how things go, but given the closeness of the endgame, I believe playing at G11 will pay off by helping force White to play in my territory later. The cost is 10 points, plus up to 4 dame, minus 6 (3*2) points I'll take from White's territory. |
|
In the 66-38 position (after white 22) white has five groups, two of them isolated, so there are only two white connections in the pipeline. That brings your calculation to 112-129 instead of 122-129. You were safer than you calculated .
The vacant cells in one's own territory where an isolated stone can be placed for slow filling have fairly recently emerged as a useful tactics. As in everything in Symple, there are two sides to it: in different conditions these very points may provide the best invasion points for an opponent.
on Jan 11th, 2013, 6:41pm, hyperpape wrote:** Move 23
I am rather confident that White's R10 is dominated by R11. It may not matter on this board, because White may want to fill more than two more times now, but better to preserve the internal placement at S10--it would probably not be good for Black to take that point with penalty of 10, the G11 group on the board, etc. |
|
Right, I didn't at the time consider the use of 'isolated internal placement' yet, and indeed it didn't matter that much here. As I remarked in the comments: "On the face of it I'd say you wouldn't need an invasion."
on Jan 11th, 2013, 6:41pm, hyperpape wrote:** Move 26
Currently, each turn I spend expanding existing groups gives me two dame and deprives Christian of one. With penalty 10, that means it could be up to a 7 point gain to play within my own territory if it later forces him to make a placement in mine. So it probably would have been better to play within my own territory than grab a few dame this turn. |
|
Also a recurring dilemma on which the last word isn't spoken yet. But you got a nice division of territory (five groups with between 11 and 17 vacancies, one connection pending) and you can handle an invasion (as will become clear).
on Jan 11th, 2013, 6:41pm, hyperpape wrote:** Move 27
It's hard to decide the best location for this move. The lower left is shapeless, so a white move at B3 would take a lot of my interior space.
** Move 33
Now, the largest interior area for Black is in the lower-middle, with 11 spaces. White has 10 spaces on the top, and Black has at least two opportunities to make placements in his own territory. |
|
As I said, a decisive victory and an instructive game, not in the last place for me .
|
| « Last Edit: Jan 14th, 2013, 10:46am by christianF » |
IP Logged |
|
|
|
|
Home
Help
Search
Members
Login
Register
Notify of replies
Send Topic
Print