Matrix Games Forums

Deal of the Week Pride of NationsTo End All Wars Releasing on Steam! Slitherine is recruiting: Programmers requiredPandora: Eclipse of Nashira gets release dateCommunity impressions of To End All WarsAgeod's To End All Wars is now availableTo End All Wars is now available!Deal of the Week: Field of GloryTo End All Wars: Video, AAR and Interview!Ageod's To End All Wars: Video, AAR and Interview!
Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

RE: Will this ship with an AI?

 
View related threads: (in this forum | in all forums)

Logged in as: Guest
Users viewing this topic: none
  Printable Version
All Forums >> [New Releases from Matrix Games] >> World in Flames >> RE: Will this ship with an AI? Page: <<   < prev  1 [2]
Login
Message << Older Topic   Newer Topic >>
RE: Will this ship with an AI? - 5/8/2013 6:22:53 AM   
bo

 

Posts: 2688
Joined: 5/1/2009
Status: offline

quote:

ORIGINAL: rmdesantis

Personally, I don't mourn the fact that the AI will be delayed. An AI would be nice to have, but I can't imagine that it would be anywhere near as good an opponent as a human (even one who plays badly). This game is extraordinarily complex and is challenging for humans to play. A year or two to develop an AI? I think you're off by an order of magnitude.

Once it is released, to your point, there will be bugs to fix. Frustration with the game would only be compounded by any dissatisfaction with the AI as well, and it would prove a distraction to correcting playability issues. Even in your scenario, fixing an AI would be secondary to having a stable platform that could be played by the paying public.

To me, the only advantage to an AI is to help learn the rules to prepare you better to play against human opponents. You can play a few impulses, perhaps a turn or two, but a competent player will likely see little challenge in continuing against it. I would love to be proven wrong, but I think this is the way the smart money would bet.

Mike


Hey Mike I am starting to feel sorry I told you about registrating for net play testing

There was an AI battle two to three years ago that you would not believe, anti and pro and it got nasty. Even though I have to agree with you that an AI in a game this immence could be hairy, but my opinion of this is when I want to play MWIF I can play it without worring about time zone opponets, those who are on vacation in the middle of a match or you young people [any left here?] who have to change a diaper when I am ready to crush them with my Panzers. Only an AI can give me the freedom to play when I want to play.

I believe net play will be difficult at best and pbem next to impossible, again just an opinion, could be totally wrong

I believe with all my heart and soul that Steve will put out a decent AI in the next year or so. But more important Steve is a dedicated WIF player and will not fade into the sunset when issues occur and he will always do his best to keep the game up and running to everyones satisfaction [errrr Steve my vacation days are from nov 1st to the 7th and I know you will gladly let me use your condo]

Bo

(in reply to rmdesantis)
Post #: 31
RE: Will this ship with an AI? - 5/8/2013 12:10:53 PM   
rmdesantis


Posts: 124
Joined: 2/25/2007
Status: offline
Hi Bo

I'm generally a skeptic on the ability of AIs to do a good job of simulating a human opponent. I don't have any concerns with his ability to program, but an AI is very challenging. If it was a game of just tactics, say, then I think an AI can me manufactured to provide a decent adversary. But I am pessimistic that a grand strategic AI could be generated that would provide a reasonable opponent. Will the AI, as German, declare Vichy or try to close the Med? Will a Japanese AI try for a China first strategy? The strategy determines production which in turn limits choices further into the game, etc. I suppose that it can be pre-determined what the AI will do (check a box for the strategy that the AI will take - close the med, for example) and the choices would be set - but the fun part of this game is unpredictability of people who will take chances in trying things that others may not have considered.

I recall that conversation, and I think I said pretty much the same thing back then. I would love for a great AI to be created. If the AI was somehow user programmable, though, then it might be able to be crowd-sourced and with contributions from may people we might get a decent computer opponent.

So I am hoping for the best, but I don't expect it. For me, I'd prefer PBEM over netplay (only because I really like to take time to think things through and don't want people to stare at a screen for an hour while I muddle through things), and netplay over AI.


Mike

(in reply to bo)
Post #: 32
RE: Will this ship with an AI? - 5/8/2013 3:01:50 PM   
pzgndr

 

Posts: 1655
Joined: 3/18/2004
Status: offline
quote:

Thing is - by the time the AI arrives there are going to be lots of people who will just view it as an "old game" and therefore not worth investing in...

They will look at it at release. See that it has no AI. And will write it off. Will they wait two or three years for the AI?


I have to chuckle here. WiF already IS an "old game", released in 1985. Methinks after 28 years, those who are somewhat unsatisfied with actually playing the full-up boardgame face-to-face with real human beings or exchanging volumes of emails playing the VASSAL version, will continue to be patient for even a post-release AI. I suspect Steve will be well rewarded for his efforts over the years. And players will be well rewarded for their patience.

(in reply to bo)
Post #: 33
RE: Will this ship with an AI? - 5/8/2013 8:51:20 PM   
shunwick


Posts: 1706
Joined: 10/15/2006
Status: offline

quote:

ORIGINAL: paulderynck


quote:

ORIGINAL: shunwick

quote:

ORIGINAL: paulderynck

Let's see. 50% of the revenue now and 50% later, versus no revenue now.


Paulderynck,

Thing is - by the time the AI arrives there are going to be lots of people who will just view it as an "old game" and therefore not worth investing in.

I do not agree with it but the gaming community (as a whole and including some wargamers) tend towards being ageist in this way.

They will look at it at release. See that it has no AI. And will write it off. Will they wait two or three years for the AI? Chances are that many will have forgotten about it by then.

Best wishes,
Steve

It's funny how many of these forgetful people keep dropping in once or twice a year to whine.


Ah, but the whiners are not the forgetful people. Please don't confuse the two.

Best wishes,
Steve

_____________________________

I love the smell of TOAW in the morning...

(in reply to paulderynck)
Post #: 34
RE: Will this ship with an AI? - 5/8/2013 9:02:31 PM   
shunwick


Posts: 1706
Joined: 10/15/2006
Status: offline
quote:


- The game will be coming out.
- It will have no AI in its original release. That decision has been made, having been weighed up by the stakeholders in the project - you know, the people who actually invest money in a project and so have real cash on the line.

If that decision works for you then buy it, if not then don't, or simply delay the purchase until you get the AI you want. Simples.


warspite1,

It should be that simple and I really hope it is. This game deserves to be a huge success. That would be a fitting reward for Steve and Slitherine and everyone who has supported the project through its development.

Best wishes,
Steve

_____________________________

I love the smell of TOAW in the morning...

(in reply to warspite1)
Post #: 35
RE: Will this ship with an AI? - 5/8/2013 9:11:20 PM   
shunwick


Posts: 1706
Joined: 10/15/2006
Status: offline

quote:

ORIGINAL: pzgndr

quote:

Thing is - by the time the AI arrives there are going to be lots of people who will just view it as an "old game" and therefore not worth investing in...

They will look at it at release. See that it has no AI. And will write it off. Will they wait two or three years for the AI?


I have to chuckle here. WiF already IS an "old game", released in 1985. Methinks after 28 years, those who are somewhat unsatisfied with actually playing the full-up boardgame face-to-face with real human beings or exchanging volumes of emails playing the VASSAL version, will continue to be patient for even a post-release AI. I suspect Steve will be well rewarded for his efforts over the years. And players will be well rewarded for their patience.


Pzgndr,

No. This release will be a new game. You are arguing that Hearts Of Iron III is an "old game" because there was Hearts of Iron II. It does not work like that.

But I do agree that this game deserves to be a huge success.

Best wishes,
Steve

_____________________________

I love the smell of TOAW in the morning...

(in reply to pzgndr)
Post #: 36
RE: Will this ship with an AI? - 5/9/2013 9:24:25 AM   
Joseignacio


Posts: 924
Joined: 5/8/2009
From: Madrid, Spain
Status: online

quote:

ORIGINAL: pzgndr

quote:

Empires in Arms was bug infected? I have been thinking of buying it, lately... Is it bug-free now?


EiA had some issues, for sure. The last patch resolved a lot of it and it plays pretty well. There are still a few gameplay issues but not bad. The AI still needs work, and hopefully the developer will return to it someday.

quote:

To me, the only advantage to an AI is to help learn the rules to prepare you better to play against human opponents. You can play a few impulses, perhaps a turn or two, but a competent player will likely see little challenge in continuing against it.


I disagree. The nirvana of having a computer opponent for a PC wargame is to have a ready willing competent adversary whenever it is convenient for you, not some pbem or NetPlay opponent someplace. I don't mind waiting for a post-release AI. And I don't mind waiting a while for Steve to continue to improve and enhance the AI over time. You can be a naysayer all you want, but I remain optimistic. The Strategic Command series has an outstanding AI, but it's taken years to get it where it is now. And the effort needed to script the AI and game events is considerable, but well worth it and proves the point that it IS possible to have a formidible AI for a complex grand strategy WWII wargame. Be patient.


Thanks a lot!

(in reply to pzgndr)
Post #: 37
RE: Will this ship with an AI? - 5/10/2013 12:37:07 PM   
Ranger5355

 

Posts: 476
Joined: 9/12/2007
From: Michigan
Status: offline
I have played WitP by email and I had a good time, but after playing my turn and sending it to my opponent I was in the mood to play so I was able to switch over to my AI game and play on. For me an AI is criticle but if I was conviced the one would be added in a reasonable time I might buy the game to learn the rules and maybe try an email game.

(in reply to Joseignacio)
Post #: 38
RE: Will this ship with an AI? - 5/12/2013 2:01:52 AM   
brian brian

 

Posts: 1655
Joined: 11/16/2005
Status: offline
I think the AI will come along fine eventually, but there is really no other way to get one than the present course. When you walk into a "Gaming" store and encounter blank stares when you ask if there are any hex-based wargames for sale, you realize how esoteric this hobby has become. And that despite a vibrant line of WWII miniatures in production and widely available.

Release to the public for human & solitaire play will put the perfect icing on what sounds to be an already smooth game engine. I do hope that perhaps there could be a way for Matrix to accept uploads of saved game files from the players, to use in AI development and to generate statistics on what is happening in the game. Data storage is so cheap now... That would be quite a project that could also help further improve the game system as a whole in years to come. Perhaps too much to hope for.

But WiF has a built-in way to learn to play the game - the Barbarossa & Guadalcanal scenarios. I think an AI good at playing Guadalcanal will be the more challenging one, as in the naval system there is more potential for feints than in the my-tanks-go-your-tanks-go land campaigns with zero Fog of War to consider. Once the AI can handle navies, the Decline & Fall scenario is the best to play as a third choice (for humans as well), as Grand Strategy is already mostly set for each side and country, but world-wide operational skills still need to be mastered. I predict that the AI for the Barbarossa scenario will be very challenging for a human opponent, and I look forward to playing it.

(in reply to Ranger5355)
Post #: 39
RE: Will this ship with an AI? - 5/12/2013 2:15:53 AM   
Shannon V. OKeets

 

Posts: 18281
Joined: 5/19/2005
From: Honolulu, Hawaii
Status: offline

quote:

ORIGINAL: brian brian

I think the AI will come along fine eventually, but there is really no other way to get one than the present course. When you walk into a "Gaming" store and encounter blank stares when you ask if there are any hex-based wargames for sale, you realize how esoteric this hobby has become. And that despite a vibrant line of WWII miniatures in production and widely available.

Release to the public for human & solitaire play will put the perfect icing on what sounds to be an already smooth game engine. I do hope that perhaps there could be a way for Matrix to accept uploads of saved game files from the players, to use in AI development and to generate statistics on what is happening in the game. Data storage is so cheap now... That would be quite a project that could also help further improve the game system as a whole in years to come. Perhaps too much to hope for.

But WiF has a built-in way to learn to play the game - the Barbarossa & Guadalcanal scenarios. I think an AI good at playing Guadalcanal will be the more challenging one, as in the naval system there is more potential for feints than in the my-tanks-go-your-tanks-go land campaigns with zero Fog of War to consider. Once the AI can handle navies, the Decline & Fall scenario is the best to play as a third choice (for humans as well), as Grand Strategy is already mostly set for each side and country, but world-wide operational skills still need to be mastered. I predict that the AI for the Barbarossa scenario will be very challenging for a human opponent, and I look forward to playing it.

Acquiring saved games from owners of MWIF is a good idea for improving the AI Opponent. It would only take a few lines of code to change from a different mode of play in a saved game to AIO.

_____________________________

Steve

Perfection is an elusive goal.

(in reply to brian brian)
Post #: 40
RE: Will this ship with an AI? - 5/14/2013 8:49:29 PM   
Ingtar

 

Posts: 170
Joined: 8/1/2004
Status: offline
I'll cheer at the release, even sans AI, for this brings the day closer when I will consider purchasing the fully armed and operational game with the AI.

(in reply to Shannon V. OKeets)
Post #: 41
RE: Will this ship with an AI? - 5/16/2013 2:54:18 PM   
Neilster


Posts: 2247
Joined: 10/27/2003
From: Hobart, Tasmania, Australia
Status: offline
For anyone who hasn't, I recommend skimming through the AI threads. A tremendous amount of work on the AI has already been done by many very experienced WiF players. Mostly what remains is to encode what has been thrashed out. I believe Steve already has a framework worked out for that.

In my opinion, and that is speaking as someone who has a computing and maths degree with a considerable AI component, the AI should be at least competent and shouldn't take too long to develop after release.

Cheers, Neilster

(in reply to Ingtar)
Post #: 42
RE: Will this ship with an AI? - 5/16/2013 5:11:59 PM   
pzgndr

 

Posts: 1655
Joined: 3/18/2004
Status: offline
quote:

ORIGINAL: Neilster
For anyone who hasn't, I recommend skimming through the AI threads. A tremendous amount of work on the AI has already been done by many very experienced WiF players. Mostly what remains is to encode what has been thrashed out. I believe Steve already has a framework worked out for that.

In my opinion, and that is speaking as someone who has a computing and maths degree with a considerable AI component, the AI should be at least competent and shouldn't take too long to develop after release.


I am optimistic too and tend to agree that a basic functional AI shouldn't take 'too long' once the various parsers and such are up and running. But I'd still expect a rather lengthy period of multiple iterations to fine tune the AI and, more importantly, resolve issues with the many combinations and permutations of game options. Getting the AI to perform optimally under any and all options may take a while.

But hey, I don't mind waiting. Getting me to perform optimally in this game, with or without advanced options, will likely take a while longer!

(in reply to Neilster)
Post #: 43
RE: Will this ship with an AI? - 6/2/2013 4:45:45 PM   
Lingering Frey

 

Posts: 43
Joined: 2/16/2012
Status: offline

quote:

ORIGINAL: pzgndr

quote:

Thing is - by the time the AI arrives there are going to be lots of people who will just view it as an "old game" and therefore not worth investing in...

They will look at it at release. See that it has no AI. And will write it off. Will they wait two or three years for the AI?


I have to chuckle here. WiF already IS an "old game", released in 1985. Methinks after 28 years, those who are somewhat unsatisfied with actually playing the full-up boardgame face-to-face with real human beings or exchanging volumes of emails playing the VASSAL version, will continue to be patient for even a post-release AI. I suspect Steve will be well rewarded for his efforts over the years. And players will be well rewarded for their patience.


Good gracious. If you are exchanging "volumes of emails playing the VASSAL version", get yourself set up with Skype and arrange to play at the same time.

(in reply to pzgndr)
Post #: 44
RE: Will this ship with an AI? - 6/8/2013 9:04:52 PM   
bo

 

Posts: 2688
Joined: 5/1/2009
Status: offline
quote:

ORIGINAL: Neilster

For anyone who hasn't, I recommend skimming through the AI threads. A tremendous amount of work on the AI has already been done by many very experienced WiF players. Mostly what remains is to encode what has been thrashed out. I believe Steve already has a framework worked out for that.

In my opinion, and that is speaking as someone who has a computing and maths degree with a considerable AI component, the AI should be at least competent and shouldn't take too long to develop after release.

Cheers, Neilster


Hi Neilster

Read your post back in May and put it aside because it was my birthday, one of too many, but now I could use some of your expertise about how an AI would be competitive in MWIF

This not a trick question I am really concerned how this would play out.

European theater: I believe a decent AI could be constructed in that area because of the mostly land warfare. In Barbarossa the AI could and should be very competent, even though it is only five turns long, it could be a great teaching tool for when the player would like to take on Global War or Fascist Tide.

Guadalcanal Pacific theater: A whole different kettle of fish IMHO. As a beta tester Barbarossa is a cake walk compared to Guadalcanal. Even though in Fascist Tide there can be some sea warfare I think it palls compared to the Pacific naval and land warfare.

If you care to take this on and explain to me how the AI can be competent in the Pacific I would deeply appreciate your insight, and just maybe we can get a conversation going here from all posters how they feel about this whether or not they like or dislike and AI opponent.

I am not looking for a discussion whether anybody here will like an AI or not, been there done that, just a discussion about how an AI might run it's war campaign. Maybe Steve does not need any insight about this but my feelings are it cant hurt and it could start us thinking about the AI's pitfalls or bonuses.

Bo

< Message edited by bo -- 6/8/2013 9:10:31 PM >

(in reply to Neilster)
Post #: 45
RE: Will this ship with an AI? - 6/8/2013 10:06:44 PM   
Greyshaft


Posts: 2234
Joined: 10/27/2003
From: Sydney, Australia
Status: offline
Deep in the bowels of my brain (if you'll forgive the lack of anatomical accuracy), there nestles a thought about a two-layer WiF game.

* The top layer is the meglomaniacs like myself who spend little time looking at the details on the map but instead issue strategic commands such as "Invade England" or "Build 2 submarines per month for six months".
* The lower layer consists of the players who need to carry out these instructions.

This would duplicate to some extent the angst of the German general staff who tried to win a war while following Hitler's sometimes imbecilic commands. I don't know it would be much fun for everyone but it might be interesting to watch.

_____________________________

/Greyshaft

(in reply to bo)
Post #: 46
RE: Will this ship with an AI? - 6/9/2013 3:23:49 PM   
Numdydar

 

Posts: 1802
Joined: 2/13/2004
Status: offline
I though we could watch all of that already on the History Channel

(in reply to Greyshaft)
Post #: 47
RE: Will this ship with an AI? - 6/9/2013 4:01:48 PM   
bo

 

Posts: 2688
Joined: 5/1/2009
Status: offline

quote:

ORIGINAL: Numdydar

I though we could watch all of that already on the History Channel

You know I thought we could get something going here a little constructive towards what will make a good AI or a so and so one or something in between GUESS NOT.

Bo

(in reply to Numdydar)
Post #: 48
RE: Will this ship with an AI? - 6/9/2013 4:47:26 PM   
Neilster


Posts: 2247
Joined: 10/27/2003
From: Hobart, Tasmania, Australia
Status: offline

quote:

ORIGINAL: bo

quote:

ORIGINAL: Neilster

For anyone who hasn't, I recommend skimming through the AI threads. A tremendous amount of work on the AI has already been done by many very experienced WiF players. Mostly what remains is to encode what has been thrashed out. I believe Steve already has a framework worked out for that.

In my opinion, and that is speaking as someone who has a computing and maths degree with a considerable AI component, the AI should be at least competent and shouldn't take too long to develop after release.

Cheers, Neilster


Hi Neilster

Read your post back in May and put it aside because it was my birthday, one of too many, but now I could use some of your expertise about how an AI would be competitive in MWIF

This not a trick question I am really concerned how this would play out.

European theater: I believe a decent AI could be constructed in that area because of the mostly land warfare. In Barbarossa the AI could and should be very competent, even though it is only five turns long, it could be a great teaching tool for when the player would like to take on Global War or Fascist Tide.

Guadalcanal Pacific theater: A whole different kettle of fish IMHO. As a beta tester Barbarossa is a cake walk compared to Guadalcanal. Even though in Fascist Tide there can be some sea warfare I think it palls compared to the Pacific naval and land warfare.

If you care to take this on and explain to me how the AI can be competent in the Pacific I would deeply appreciate your insight, and just maybe we can get a conversation going here from all posters how they feel about this whether or not they like or dislike and AI opponent.

I am not looking for a discussion whether anybody here will like an AI or not, been there done that, just a discussion about how an AI might run it's war campaign. Maybe Steve does not need any insight about this but my feelings are it cant hurt and it could start us thinking about the AI's pitfalls or bonuses.

Bo

I'd have to read back through all the AI threads because I can't remember how we worked out how the AI was going to turn strategic directives into operational and tactical decisions. It was a long time ago but the issues you've raised were considered. Steve should be able to provide you with a quick summary.

Cheers, Neilster

(in reply to bo)
Post #: 49
RE: Will this ship with an AI? - 6/9/2013 4:50:20 PM   
bo

 

Posts: 2688
Joined: 5/1/2009
Status: offline
A little guessing here but I think it was a good twenty years ago I saw a DOS game about the invasion of Russia. First war game I ever saw on the computer, for the life of me I cannot remember the name of the game, but a comment by the game programmers I do remember "it's a game you cannot win"

What I do remember very well was the 24th Panzer division in the game. In the first few moves it wiped out every Russian unit in front of it. After that it could not beat any unit Russia had and every game turned into a nightmare of losses. I was crushed because I had visions of myself being Erwin Rommel or Heinz Guderian but alas that fell by the wayside with my warrior ego. About a year later I read somewhere that the AI cheated and you could not win. Oh well!

I know Steve would not think of cheating but I do not think he would mind beating most of us with his AI

If anyone here has played the Strategic command series from the Battlefront gaming company please let me know.

When the German AI in Strategic command attacked Russia after destroying Poland and France you had to play your best game without mistakes to stop them. I could not until I found that one famous stop gap that hurts all computer AI's. A trick within a trick.

While Germany was defeating France I [Russia] built a string of forts along rivers in front of Moscow from Rostov in the south to Vologda in the north, by the time the Panzers reached the line of forts I had been able to put a large amount of artillery just behind the forts. For the first time in the game the AI struggled with the situation. Attacking piece meal instead of concentrating in one crucial area with a lot of its units.

I personally thought the AI was very adequate for a game constructed five years ago. Their Pacific game fell short with the AI IMO. I found that given enough power the AI could be very strong even against better than average players.

I am guessing here But Steve could have levels of ability for the AI, not cheating in any way but increasing each AI's unit with either never being disrupted, never OOS, higher attack factors etc. To the purist that's a nono and I understand that but I would like to play against a competitive AI and I do not care how it's done to make it competitive.

I would appreciate some constructive comments from people who appreciate this game whether it be the board game people or the CWIF people. The comments above belong some where else even though I deeply appreciate your comments Not referring to you Neilster.

Bo

< Message edited by bo -- 6/9/2013 4:54:42 PM >

(in reply to bo)
Post #: 50
RE: Will this ship with an AI? - 6/9/2013 6:05:23 PM   
Klydon


Posts: 2156
Joined: 11/28/2010
Status: offline
I am not a fan of AI's that bend the rules for themselves, but not for the players so I would be against changes in the rules to a point on certain matters of unit strength, supply factors, never being disrupted, etc. That is one of my issues with WITE and the Russian admin setting. Set it high enough and it gets unlimited admin points, which totally destroys the game feel.

Having said all this, the search to make AI's more competitive does deserve examination. Some areas I think would be helpful would be ignoring gearing limits, varying amounts of extra resources, perhaps a positive modifier for combat (no more than +1). One of the major weakness of most AI's is they are very, very predictable and can often be baited into doing certain things in response to a player "setting the AI up". Of course players do that to other players as well, but AI's are more predictable when falling into a trap like that. Adding some randomness may help in terms of sometimes the AI may fall for it and sometimes they may not, but they consistently won't do one thing or another to make it far less predictable that if player does x, AI will do y.

(in reply to bo)
Post #: 51
RE: Will this ship with an AI? - 6/9/2013 10:15:05 PM   
SirWhiskers

 

Posts: 30
Joined: 12/27/2009
Status: offline

quote:

ORIGINAL: bo

If anyone here has played the Strategic command series from the Battlefront gaming company please let me know.

When the German AI in Strategic command attacked Russia after destroying Poland and France you had to play your best game without mistakes to stop them. I could not until I found that one famous stop gap that hurts all computer AI's. A trick within a trick.

While Germany was defeating France I [Russia] built a string of forts along rivers in front of Moscow from Rostov in the south to Vologda in the north, by the time the Panzers reached the line of forts I had been able to put a large amount of artillery just behind the forts. For the first time in the game the AI struggled with the situation. Attacking piece meal instead of concentrating in one crucial area with a lot of its units.

I personally thought the AI was very adequate for a game constructed five years ago. Their Pacific game fell short with the AI IMO. I found that given enough power the AI could be very strong even against better than average players.

Bo


This sounds like Strategic Command Global Conflict, for which I've only played the demo. The Original SC has a fairly limited AI, which is hard-coded with certain grand strategies. The ability of the AI varies according to the level of the game - tactically, it can be quite lethal; operationally it can be competent; strategically, rather inept. It constantly wastes build points moving units back and forth via SR. It doesn't really know how to handle unusual situations, such as an Allied invasion of Italy in 1939/40. It throws away the German u-boats early by keeping them in range of British airpower.

Some of us are old enough to remember the original SSG (Strategic Studies Group) that published some very good early computer games, such as Reach for the Stars and Carriers at War. Roger Keating, the designer known for his excellent AI's, made the very important point that most game AI's fail because they are tacked on at the end. Per Keating, proper game design requires designing the AI first. This serves to guide the designer, so the game code takes the AI into account from day one. If the designer realizes that he needs several new variables for the AI only after the code is 90% complete, that's too late. In designing the AI first, the game designer also learns what features should be excluded because they can't be properly handled by an AI. Keating also made the point that converting an existing boardgame to computer is the most difficult situation for a game designer - not only are the game rules already written, but they were originally designed for humans, not machines.

I want MWIF to have an AI. I want it have a competent AI that doesn't cheat. I'm keeping my expectations low not because I believe Steve and others won't put in the necessary effort (they will), but because I recognize that designing this game's AI will be incredibly difficult. I also expect, like the original SC, the AI will be better at the lower tactical levels than the higher strategic levels.

(in reply to bo)
Post #: 52
RE: Will this ship with an AI? - 6/9/2013 11:41:54 PM   
Shannon V. OKeets

 

Posts: 18281
Joined: 5/19/2005
From: Honolulu, Hawaii
Status: offline

quote:

ORIGINAL: SirWhiskers


quote:

ORIGINAL: bo

If anyone here has played the Strategic command series from the Battlefront gaming company please let me know.

When the German AI in Strategic command attacked Russia after destroying Poland and France you had to play your best game without mistakes to stop them. I could not until I found that one famous stop gap that hurts all computer AI's. A trick within a trick.

While Germany was defeating France I [Russia] built a string of forts along rivers in front of Moscow from Rostov in the south to Vologda in the north, by the time the Panzers reached the line of forts I had been able to put a large amount of artillery just behind the forts. For the first time in the game the AI struggled with the situation. Attacking piece meal instead of concentrating in one crucial area with a lot of its units.

I personally thought the AI was very adequate for a game constructed five years ago. Their Pacific game fell short with the AI IMO. I found that given enough power the AI could be very strong even against better than average players.

Bo


This sounds like Strategic Command Global Conflict, for which I've only played the demo. The Original SC has a fairly limited AI, which is hard-coded with certain grand strategies. The ability of the AI varies according to the level of the game - tactically, it can be quite lethal; operationally it can be competent; strategically, rather inept. It constantly wastes build points moving units back and forth via SR. It doesn't really know how to handle unusual situations, such as an Allied invasion of Italy in 1939/40. It throws away the German u-boats early by keeping them in range of British airpower.

Some of us are old enough to remember the original SSG (Strategic Studies Group) that published some very good early computer games, such as Reach for the Stars and Carriers at War. Roger Keating, the designer known for his excellent AI's, made the very important point that most game AI's fail because they are tacked on at the end. Per Keating, proper game design requires designing the AI first. This serves to guide the designer, so the game code takes the AI into account from day one. If the designer realizes that he needs several new variables for the AI only after the code is 90% complete, that's too late. In designing the AI first, the game designer also learns what features should be excluded because they can't be properly handled by an AI. Keating also made the point that converting an existing boardgame to computer is the most difficult situation for a game designer - not only are the game rules already written, but they were originally designed for humans, not machines.

I want MWIF to have an AI. I want it have a competent AI that doesn't cheat. I'm keeping my expectations low not because I believe Steve and others won't put in the necessary effort (they will), but because I recognize that designing this game's AI will be incredibly difficult. I also expect, like the original SC, the AI will be better at the lower tactical levels than the higher strategic levels.

I expect the reverse.

WIF has some phases that are clearly strategic (e.g., production planning and production). Programming those won't be much of a problem because the search space is relatively small. Likewise decisions in the DOW phase are pretty straight forward, assuming that strategic plans have been defined for that purpose - just select one and stay with it unless something drastic happens.

Operational decisions could be difficult to make (e.g., where to send reinforcements, how to allocate air and naval units around the world and within a theater.

But tactical decisions can be very difficult since they are mostly related to the hex grid and sea areas. Geography is difficult for an AI Opponent. Players glance at a map and know where threats and opportunities exist (assuming the player is any good at the game). But for an AIO, examining the map is tedious and potentially a source of errors. Tactical decisions also involve sequencing actions. While the AIO can do an exhaustive search, elapsed time is a consideration. Again, the hex grid opens up an enormous range of possibilities. Comparing WIF to chess, chess is a closed ended search space with few moves available to either side. WIF has hundreds of units on the map and a world of hexes to explore. For example, there are 64 squares on a chessboard, and 83 sea areas in MWIF.

_____________________________

Steve

Perfection is an elusive goal.

(in reply to SirWhiskers)
Post #: 53
RE: Will this ship with an AI? - 6/10/2013 8:07:07 AM   
paulderynck


Posts: 4170
Joined: 3/24/2007
From: Canada
Status: offline
The day an AI makes an attack that puts the flipped defenders of its next attack OOS - I'll be impressed.

_____________________________

Paul

(in reply to Shannon V. OKeets)
Post #: 54
RE: Will this ship with an AI? - 6/10/2013 9:02:32 AM   
Numdydar

 

Posts: 1802
Joined: 2/13/2004
Status: offline

quote:

ORIGINAL: bo

A little guessing here but I think it was a good twenty years ago I saw a DOS game about the invasion of Russia. First war game I ever saw on the computer, for the life of me I cannot remember the name of the game, but a comment by the game programmers I do remember "it's a game you cannot win"

What I do remember very well was the 24th Panzer division in the game. In the first few moves it wiped out every Russian unit in front of it. After that it could not beat any unit Russia had and every game turned into a nightmare of losses. I was crushed because I had visions of myself being Erwin Rommel or Heinz Guderian but alas that fell by the wayside with my warrior ego. About a year later I read somewhere that the AI cheated and you could not win. Oh well!

I know Steve would not think of cheating but I do not think he would mind beating most of us with his AI

If anyone here has played the Strategic command series from the Battlefront gaming company please let me know.

When the German AI in Strategic command attacked Russia after destroying Poland and France you had to play your best game without mistakes to stop them. I could not until I found that one famous stop gap that hurts all computer AI's. A trick within a trick.

While Germany was defeating France I [Russia] built a string of forts along rivers in front of Moscow from Rostov in the south to Vologda in the north, by the time the Panzers reached the line of forts I had been able to put a large amount of artillery just behind the forts. For the first time in the game the AI struggled with the situation. Attacking piece meal instead of concentrating in one crucial area with a lot of its units.

I personally thought the AI was very adequate for a game constructed five years ago. Their Pacific game fell short with the AI IMO. I found that given enough power the AI could be very strong even against better than average players.

I am guessing here But Steve could have levels of ability for the AI, not cheating in any way but increasing each AI's unit with either never being disrupted, never OOS, higher attack factors etc. To the purist that's a nono and I understand that but I would like to play against a competitive AI and I do not care how it's done to make it competitive.

I would appreciate some constructive comments from people who appreciate this game whether it be the board game people or the CWIF people. The comments above belong some where else even though I deeply appreciate your comments Not referring to you Neilster.

Bo


I just found Strategic Command Gold and the actually have Assult on Communism out which is just the Russian campaign (bigger map, more units, etc) and the AI is supposted to be very good. In the Gold version, which I played, I thought the AI was pretty good. Or maybe I am just really bad lol. 90% of my games I play against the AI and I am just happy they work at all, much less great.

(in reply to bo)
Post #: 55
RE: Will this ship with an AI? - 6/10/2013 12:51:23 PM   
pzgndr

 

Posts: 1655
Joined: 3/18/2004
Status: offline

quote:

ORIGINAL: Numdydar

I just found Strategic Command Gold and the actually have Assult on Communism out which is just the Russian campaign (bigger map, more units, etc) and the AI is supposted to be very good. In the Gold version, which I played, I thought the AI was pretty good. Or maybe I am just really bad lol. 90% of my games I play against the AI and I am just happy they work at all, much less great.


The combination of fuzzy-logic generic AI (movements, combats, etc.), campaign-specific AI scripting (plans, production, diplomacy, etc.), and variable event scripting (plus fog of war effects!) result in a very dynamic computer opponent in the Strategic Command series. It just keeps getting better as each of those three areas get improvements and enhancements, often requiring updated game engines.

If you've got the Global GOLD expansion, check out my Advanced Third Reich mod. Both Axis and Allied AI is scripted to be competent and challenging. This is most noticeable in the Med Front, which proved very tricky to script properly to handle many different strategies. Many WWII ETO games don't handle the Med very well at all, for either Axis or Allied reinforcements of North Africa. For instance, French colonies could go Vichy or Free, Axis could invade Spain or Turkey, Axis could break through in Egypt or not, etc. Mentally, we can think through it easily enough; it's another matter to script it all so it works adequately under various conditions. But, it IS possible. I have been routinely surprised by my own AI scripting...

(in reply to Numdydar)
Post #: 56
RE: Will this ship with an AI? - 6/10/2013 6:24:30 PM   
Shannon V. OKeets

 

Posts: 18281
Joined: 5/19/2005
From: Honolulu, Hawaii
Status: offline

quote:

ORIGINAL: pzgndr


quote:

ORIGINAL: Numdydar

I just found Strategic Command Gold and the actually have Assult on Communism out which is just the Russian campaign (bigger map, more units, etc) and the AI is supposted to be very good. In the Gold version, which I played, I thought the AI was pretty good. Or maybe I am just really bad lol. 90% of my games I play against the AI and I am just happy they work at all, much less great.


The combination of fuzzy-logic generic AI (movements, combats, etc.), campaign-specific AI scripting (plans, production, diplomacy, etc.), and variable event scripting (plus fog of war effects!) result in a very dynamic computer opponent in the Strategic Command series. It just keeps getting better as each of those three areas get improvements and enhancements, often requiring updated game engines.

If you've got the Global GOLD expansion, check out my Advanced Third Reich mod. Both Axis and Allied AI is scripted to be competent and challenging. This is most noticeable in the Med Front, which proved very tricky to script properly to handle many different strategies. Many WWII ETO games don't handle the Med very well at all, for either Axis or Allied reinforcements of North Africa. For instance, French colonies could go Vichy or Free, Axis could invade Spain or Turkey, Axis could break through in Egypt or not, etc. Mentally, we can think through it easily enough; it's another matter to script it all so it works adequately under various conditions. But, it IS possible. I have been routinely surprised by my own AI scripting...

I'm not familiar with any of those games.

But I have seen my AIO code (for other games I have written) make good decisions that were a surprise to me. They were all perfectly logical, but quite disconcerting when first seen - after some analysis, the decisions look brilliant. Basically, the aggregation of many small, well made decisions can generate impressive results.

_____________________________

Steve

Perfection is an elusive goal.

(in reply to pzgndr)
Post #: 57
Page:   <<   < prev  1 [2]
All Forums >> [New Releases from Matrix Games] >> World in Flames >> RE: Will this ship with an AI? Page: <<   < prev  1 [2]
Jump to:





New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI

0.132