Traveller Store CotI Features New Posts Mark Forums Read Register


Go Back TravellerRPG.com > Citizens of the Imperium > Random Static > Non-Traveller Gaming

Non-Traveller Gaming A forum specifically for discussing those other games we like to play.

Reply
 
Thread Tools Display Modes
  #1  
Old April 30th, 2016, 08:24 AM
Blue Ghost Blue Ghost is offline
Banned
 
Join Date: May 2001
Location: Foster City, CA
Posts: 7,500
Gallery : 0
Visit Blue Ghost's Blog
Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++
Default Important Gaming question

I am re-watching an old 1970s program called "The Ascent of Man" on DVD. And the host,
Jacob Bronowski, quotes one of his contemporaries Jon Von Neuman as saying that Chess, the game that we know, is merely a calculation exercise that has a correct solution; "a well defined form of computation."

In a sense that would seem to imply that a lot of board games are also exercises in calculation, each with a correct solution, and that REAL games have bluffing, little tactics and second guessing the opposition.

I was tempted to re-up my old SFB account, and ask the geniuses there, largely because there's more activity there, but I thought I might put this thought to the COTI geniuses.

Board games = computer calculations

Real games = human interaction

Does anyone have a thought on that?

It's pretty heavy stuff, but I'm wondering what the opinions are here.

Thanks to anyone who wants to reply.
Reply With Quote
  #2  
Old April 30th, 2016, 08:31 AM
mike wightman's Avatar
mike wightman mike wightman is online now
Noble
 
Join Date: Jun 2003
Location: Newcastle
Posts: 15,448
Gallery : 0
mike wightman Respected Citizenmike wightman Respected Citizenmike wightman Respected Citizenmike wightman Respected Citizenmike wightman Respected Citizenmike wightman Respected Citizenmike wightman Respected Citizen
Default

Super-computer beats world Go champion, and Go is a much harder game to encode than chess.

Human opponents can fall for a bluff, have an off day, or just not be up to thinking so many moves hence.

Pit two super-computers against each other playing Go and it could be interesting.
__________________
The beauty of CT LBB1-3 is that the ref is free to make such decisions for themselves.
Reply With Quote
  #3  
Old April 30th, 2016, 11:49 AM
whartung whartung is offline
Citizen: SOC-14
 
Join Date: Nov 2004
Posts: 1,911
Gallery : 0
whartung Citizen+whartung Citizen+
Default

Go and Chess have perfect information, and no luck. So, at any one point, in theory, there is an "optimal move". The only uncertainty is what move the other players chooses. The problem with the computer players was simply that the magnitude of the data involved in making the optimal move was limited by the computers capacity and performance. I guess in Chess if each player always makes the perfect move, you end up with a draw. I don't know what happens with Go.

Other games, like poker, are filled with uncertainty, but work well with statistics. But there's enough luck that you can put two perfect players against each other, and one will win, and one will lose, and not necessarily the same one each time. The bluffing and "reading" of the other players is the psychological aspect of the game, trying to force the other player to make a mistake. Given two perfect, soulless computer players, they will just play the statistics and one will simply "roll badly" eventually.
Reply With Quote
  #4  
Old April 30th, 2016, 02:17 PM
Blue Ghost Blue Ghost is offline
Banned
 
Join Date: May 2001
Location: Foster City, CA
Posts: 7,500
Gallery : 0
Visit Blue Ghost's Blog
Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++Blue Ghost Citizen++
Default

It got me to thinking though; are games like Car Wars or Star Fleet Battles or even GDW's own "Imperium" simply bloated versions of chess? And if so, are they really games at all, or just highly complex calculations?
Reply With Quote
  #5  
Old April 30th, 2016, 02:19 PM
flykiller's Avatar
flykiller flykiller is offline
Citizen: SOC-14
 
Join Date: Aug 2002
Location: desert
Posts: 6,216
Gallery : 102
flykiller Citizen+flykiller Citizen+flykiller Citizen+
Default

Quote:
Board games = computer calculations

Real games = human interaction
in go it is known that if one plays on the center point and thereafter simply mirrors each opposing move, this wins. not playing the center point first is considered a simple matter of etiquette. in chess this dynamic is not present but presumably white will win an entirely calculated game.

while each game is in fact limited in possibilities and thus the "perfect" game theoretically is obtainable, in human practice they are not - humans no matter how studied and experienced and memory-capable are not able to approach this "perfect" game and thus experience each game instance as if it were unbounded and thus each game in fact includes human interaction - this is particularly true of chess where grandmasters tend towards high-strung neurosis and almost idiot-savant status. thus the games are "real".
Reply With Quote
  #6  
Old April 30th, 2016, 02:42 PM
timerover51's Avatar
timerover51 timerover51 is offline
Citizen: SOC-14
 
Join Date: Sep 2012
Location: North of Chicago
Posts: 5,163
Gallery : 0
Visit timerover51's Blog
timerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizen
Default

I have played Axis and Allies quite a bit, and on those boards, you have a group of players who have tried to reduce the game into a set series of moves that will always allow the Axis player to win. However, when you examine how they are doing that, it is always a two-player game, one Axis and one Allies. As soon as you add another player to the mix, their systems break down, as then two players have to cooperate with each other, and they may not follow the set pattern, and the player used to straight calculation experiences a break down. They are also the ones totally opposed to any form of weapon development use, as that also upsets the calculations.

Based on my observations of our summer games classes over a period of 13 years, many of the students never having gamed before, as soon as a board game goes beyond two players, it no longer is a set of calculations, and the more the players, the less predictable to game is.

One of the game we use is called Ancient Conquests, covering the Biblical World of the Old Testament. It has four players, each with a list of objectives to gain victory points, with the maximum possible being 40. However, one problem the players have to deal with it that in the process of gaining victory points for themselves, they may also be gaining victory points for another player. Alliances are possible, as it the double cross. That board game can get pretty wild at times, and I cannot see how you could reduce it to pure calculation, as long as you have 4 human players.

Then, you also have to account for the different approach to board games between males and females. They play totally differently, and heaven help the males if the females make an alliance.

It might be possible to reduce a game with only two players to calculations, but if dice are involved, it is going to be extremely difficult. I can see someone trying to reduce a tactical battle to pure math, with is what Dupuy tries to do with his Quantitative Analysis model, but it breaks down even on a tactical level. If you are considering multi-player games, then while math is helpful, it is not going to be the end all.
__________________
Link to Piper Sector: http://www.zarthani.net/ridder-mankind_to_the_stars.htm
Do you have a security clearance? New Blog Post. http://www.youtube.com/watch?v=ElI451TxsTg, 3:24 in.
I march to my own set of bagpipes. Caution: This individual thinks that studying logistics is FUN.

Nail to the mast her holy flag,
Set every threadbare sail,
And give her to the god of storms,
The lightning and the gale!
Reply With Quote
  #7  
Old April 30th, 2016, 03:01 PM
flykiller's Avatar
flykiller flykiller is offline
Citizen: SOC-14
 
Join Date: Aug 2002
Location: desert
Posts: 6,216
Gallery : 102
flykiller Citizen+flykiller Citizen+flykiller Citizen+
Default

Quote:
Then, you also have to account for the different approach to board games between males and females. They play totally differently
I would very much like to see an explication and discussion of this.
Reply With Quote
  #8  
Old April 30th, 2016, 03:04 PM
nobby-w nobby-w is offline
Citizen: SOC-13
 
Join Date: Jul 2013
Location: The High Weald.
Posts: 542
Gallery : 5
nobby-w Citizen
Default

Quote:
Originally Posted by Blue Ghost View Post
I am re-watching an old 1970s program called "The Ascent of Man" on DVD. And the host,
Jacob Bronowski, quotes one of his contemporaries Jon Von Neuman as saying that Chess, the game that we know, is merely a calculation exercise that has a correct solution; "a well defined form of computation."
...
Optimising chess play is theoretically possible but computationally impractical because of the combinatorics - there are just too many possible games to examine them all. Therefore chess software is still dependent on heuristics to evaluate the merits of different strategies without having to exhaustively explore all possible paths.

A.I. such as Deep Blue uses machine learning techniques to develop computations that do this sort of evaluation. Similar techniques have been applied to Go with some success recently.

Theoretically you could apply deep learning technology to playing any arbitrary game if you could (for example) teach it to infer motive from a players actions or negotiations. This is a step further than current game playing technology as you would have to get it to evaluate human interaction.

There is some research in this field (unrelated to game play), although my google-fu fails me. If you could use a deep-learning network to evaluate people's conversation W.R.T. their actions you could - at least in theory - combine this with game playing A.I.

The limiting factor of this is finding training data for the A.I. Finding go or chess games to examine isn't hard. Creating a big enough training set of footage of players' conversations around games of Risk or Diplomacy would probably not be feasible.
Reply With Quote
  #9  
Old April 30th, 2016, 03:56 PM
JimMarn's Avatar
JimMarn JimMarn is offline
Citizen: SOC-14
 
Join Date: Mar 2013
Posts: 1,095
Gallery : 0
JimMarn Citizen+JimMarn Citizen+JimMarn Citizen+
Default

Quote:
Originally Posted by flykiller View Post
I would very much like to see an explication and discussion of this.
In my first year or two of AD&D, the majority of the players were women. Their attitude was 'We attack the orcs ! No prisoners !'. While some of the men wanted to do that, and others wanted to parley. That was the home group.

Most convention players I ran into could game in either direction, parley, some fighting, some for all out attack. Gender didn't seem to matter much.

Different game store groups gamed similarly to the convention groups, with the occasional player who didn't ant to take prisoners, and about the same number wanted to parley.

Note that out of about 200 players, some with multiple characters, less than 5 human characters. The majority were elf, half-elf, and dwarf player characters. So there would be some animosity there between them and certain monsters they encountered during game sessions.

Even when the player characters encountered non-giant class monsters, they still did the above.
__________________
'through the thorns to the stars...' Ghost in the Shell Anime.
My Traveller site <- 140 planet surveys, 1299 UWPs, and 1299 world maps
my ATU planet survey links <- links to all surveys
Reply With Quote
  #10  
Old April 30th, 2016, 04:02 PM
timerover51's Avatar
timerover51 timerover51 is offline
Citizen: SOC-14
 
Join Date: Sep 2012
Location: North of Chicago
Posts: 5,163
Gallery : 0
Visit timerover51's Blog
timerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizentimerover51 Respected Citizen
Default

Quote:
Originally Posted by nobby-w View Post
The limiting factor of this is finding training data for the A.I. Finding go or chess games to examine isn't hard. Creating a big enough training set of footage of players' conversations around games of Risk or Diplomacy would probably not be feasible.
We also play Diplomacy. Those games can get really nasty at times, and the players do tend to have grudges the next year. I cannot see how you could reduce that to strictly calculation, or computer play.
__________________
Link to Piper Sector: http://www.zarthani.net/ridder-mankind_to_the_stars.htm
Do you have a security clearance? New Blog Post. http://www.youtube.com/watch?v=ElI451TxsTg, 3:24 in.
I march to my own set of bagpipes. Caution: This individual thinks that studying logistics is FUN.

Nail to the mast her holy flag,
Set every threadbare sail,
And give her to the god of storms,
The lightning and the gale!
Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
FFE Web Site: Nothing Important Going On Here Draconian Traveller 5 20 October 24th, 2013 03:55 PM
An important tip. iron warrior Random Static 3 March 31st, 2011 08:39 AM
IMPORTANT (If you don't want to be banned) hunter Citizens' Information Centre 20 January 17th, 2005 06:58 PM
T20 Open Gaming Content Question Zhanken the Snake T20 - Traveller for the D20 System 2 April 26th, 2002 12:07 AM

This website and its contents are copyright ©2010-2013 Far Future Enterprises. All rights reserved. Traveller is a registered trademark of Far Future Enterprises .
Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Copyright (c) 2010-2013, Far Future Enterprises. All Rights Reserved.