Rob, your description of the function of the AI comports perfectly with my own interests. I've played complex wargames all my life -- ASL, Third Reich, now War in the Pacific -- but I've never gotten around to trying WIF. So I really want an AI that can at least give me a basic game so that I can learn the system. WITP's AI does this, and that's all I ask of it -- though I'm pleasantly surprised that the WITP AI is giving me a pretty good run for my money in my first game.
I'm also intrigued and encouraged to hear that AI development is the most "fun" part of designing a game. I guess I shouldn't be surprised, but I'm glad to hear it nevertheless. I
AI vs non-AI
Moderator: Shannon V. OKeets
RE: AI vs non-AI
Since we have someone here who obviously enjoys programming game AI, I have a question: Is a learning AI possible for a game of WiF's complexity?ORIGINAL: RobertCrandall
I do not believe there will be a non-AI version. It will never have a killer AI, but it should have a sufficiently competent one to teach a newcomer some of the basics and to provide useful filler players in a multiplayer game if desired. Beyond that I cannot really say yet.
AI is the most fun part of any game project and I'm sure that there will be continual development over many years to come in the MWIF AI. At least some of it will be exposed so that interested parties can tweak weightings, etc. For version 1, our goals will be quite modest - honestly, there is no other way to do it!
Cheers, Rob.
-
IronManBeta
- Posts: 3918
- Joined: Mon Feb 25, 2002 10:00 am
- Location: Brantford, Ontario
- Contact:
RE: AI vs non-AI
My marching orders are to produce a "faithful adaptation of the lastest Deluxe version of WIF" and 'to use the existing code base as far as possible'. Would it be possible to add a few options that allowed a partial reversion to WIF 5.0-like levels of complexity? Sounds doable to me. You have to realize though that no game in history as ever had more optional rules than this one and they are all but out of control already!
My first job is to do what I was told to do and then I can get fancy and entertain special requests later. Part of the appeal to me of the project initially was that there was a clear, well established and thoroughly tested set of rules. Nearly everyone I've talked to since has asked me to change them one way or the other to suit their personal preferences. Yikes! Our goal is to make everyone reasonably happy and that alone will be a ton of work. All the add ons will be a treat but it will take a lot of work to integrate them properly. I'm particularly concerned about how they would impact the AI we have to develop.
Cheers, Rob.
My first job is to do what I was told to do and then I can get fancy and entertain special requests later. Part of the appeal to me of the project initially was that there was a clear, well established and thoroughly tested set of rules. Nearly everyone I've talked to since has asked me to change them one way or the other to suit their personal preferences. Yikes! Our goal is to make everyone reasonably happy and that alone will be a ton of work. All the add ons will be a treat but it will take a lot of work to integrate them properly. I'm particularly concerned about how they would impact the AI we have to develop.
Cheers, Rob.
ORIGINAL: meyerg
Rob:
IMHO the Wif Rules set was in its prime during 5th edition Wif. Has there been any consideration to making computer Wif based on an earlier version? All the following additions were, in my opinion, unnecessary:
1) motorized and mechanized INF (didn't add much but complexity)
2) Leaders in Flames = highly unbalanced game
3) PIF = many more planes with many different ranges and abilities but no increase in offensive/rebase allowance during land and combined
4) territorials = many more cheap units for allies.
5) the new maps (remember the bright colors of the old paper maps instead of the grey look of the new plastic feeling maps)
I think it sure would be easier to make a game based on Wif5 rules than the latest Wif with all the expansions like cruisers in flames, propaganda in flames, beer halls in flames, feminism in flames and the much awaited for Lord of the Rings Return of the King in Flames
Greg
-
IronManBeta
- Posts: 3918
- Joined: Mon Feb 25, 2002 10:00 am
- Location: Brantford, Ontario
- Contact:
RE: AI vs non-AI
The AI architecture has not been set yet but it will doubtless be some sort of layered hierarchy of finite state machines within a subsumtive framework. (Don't worry if that is Greek to you - it means a fairly conventional job, thats all). Some of the problems posed towards the middle layer of the AI sound like they would lend themselves to a scripting solution so there might well be a 'playbook' aspect too.
Something that I have not done yet in my games but would be fun to do is to give the the fsm agents a 'memory' of what they have done so far. That would really help with 'maintenance of the objective' issues in the short run, and would open up all kinds of possibilities more generally. Once you have a memory you can start to do all kinds of other things, one of which is simple learning.
Will it be fancy? Nope, not to start but it might actually reduce the complexity of the AI in the long run rather than increase it. Instead of working out all the combinations and permutations for all possible action in code it might become possible to do a simple look up of all ~matching instances in a database of memories and estimate probable outcomes based on what it finds - that sort of thing. That makes me really, really interested in pursuing it!
This probably did not answer your question but I liked answering it anyway. Cheers, Rob.
Something that I have not done yet in my games but would be fun to do is to give the the fsm agents a 'memory' of what they have done so far. That would really help with 'maintenance of the objective' issues in the short run, and would open up all kinds of possibilities more generally. Once you have a memory you can start to do all kinds of other things, one of which is simple learning.
Will it be fancy? Nope, not to start but it might actually reduce the complexity of the AI in the long run rather than increase it. Instead of working out all the combinations and permutations for all possible action in code it might become possible to do a simple look up of all ~matching instances in a database of memories and estimate probable outcomes based on what it finds - that sort of thing. That makes me really, really interested in pursuing it!
This probably did not answer your question but I liked answering it anyway. Cheers, Rob.
ORIGINAL: amwild
Since we have someone here who obviously enjoys programming game AI, I have a question: Is a learning AI possible for a game of WiF's complexity?ORIGINAL: RobertCrandall
I do not believe there will be a non-AI version. It will never have a killer AI, but it should have a sufficiently competent one to teach a newcomer some of the basics and to provide useful filler players in a multiplayer game if desired. Beyond that I cannot really say yet.
AI is the most fun part of any game project and I'm sure that there will be continual development over many years to come in the MWIF AI. At least some of it will be exposed so that interested parties can tweak weightings, etc. For version 1, our goals will be quite modest - honestly, there is no other way to do it!
Cheers, Rob.
RE: AI vs non-AI
My marching orders are to produce a "faithful adaptation of the lastest Deluxe version of WIF" and 'to use the existing code base as far as possible'.
My opinion is that this means we are not starting with a clean sheet of paper, but CWIF will have to make compromises for PBEM.
Would it be possible to add a few options that allowed a partial reversion to WIF 5.0-like levels of complexity? Sounds doable to me. You have to realize though that no game in history as ever had more optional rules than this one and they are all but out of control already!
I believe we will only have one CWIF with optional rules to toggle on/off because the customer base is so small. I worry that a "faithful adaption" leaves no room to remove rules that add little or no gameplay but added complexity does three bad things: 1) added complexity for coding 2) added challenge to AI 3) higher learning curve for newbees. I believe these things make for a longer development cycle with reduced chance of mainstream acceptance (see computer Eastfront for an example of a great computer adaption with mainstream acceptance).
Finally, this makes the rumors of the European map US and Pacific map seem to me alive and well.
Greg
RE: AI vs non-AI
Well, this is somewhat greek; I am a programmer, but specialising in databases rather than AI.
However, this brings me to another point that interests me - logging. In a much erlier post before you joined in these discussions, I raised the question of an option to maintain a full log of every move made by man or machine in a given game so that a game's progress could be saved and replayed, and if desired, play could be picked up at any point in order to try to change the result (creating a duplicate of the log to that point).
Sure, with a game of this complexity, the log could be quite large, but it would provide some players with handy ammunition for bragging.
I would imagine that a full log would be necessary for the AI memory you discussed; the replay and interrupt facility (which I first saw in the original Civilization PC game) would be another bonus.
However, this brings me to another point that interests me - logging. In a much erlier post before you joined in these discussions, I raised the question of an option to maintain a full log of every move made by man or machine in a given game so that a game's progress could be saved and replayed, and if desired, play could be picked up at any point in order to try to change the result (creating a duplicate of the log to that point).
Sure, with a game of this complexity, the log could be quite large, but it would provide some players with handy ammunition for bragging.
I would imagine that a full log would be necessary for the AI memory you discussed; the replay and interrupt facility (which I first saw in the original Civilization PC game) would be another bonus.
ORIGINAL: RobertCrandall
The AI architecture has not been set yet but it will doubtless be some sort of layered hierarchy of finite state machines within a subsumtive framework. (Don't worry if that is Greek to you - it means a fairly conventional job, thats all). Some of the problems posed towards the middle layer of the AI sound like they would lend themselves to a scripting solution so there might well be a 'playbook' aspect too.
Something that I have not done yet in my games but would be fun to do is to give the the fsm agents a 'memory' of what they have done so far. That would really help with 'maintenance of the objective' issues in the short run, and would open up all kinds of possibilities more generally. Once you have a memory you can start to do all kinds of other things, one of which is simple learning.
Will it be fancy? Nope, not to start but it might actually reduce the complexity of the AI in the long run rather than increase it. Instead of working out all the combinations and permutations for all possible action in code it might become possible to do a simple look up of all ~matching instances in a database of memories and estimate probable outcomes based on what it finds - that sort of thing. That makes me really, really interested in pursuing it!
This probably did not answer your question but I liked answering it anyway. Cheers, Rob.
RE: AI vs non-AI
ORIGINAL: RobertCrandall
My marching orders are to produce a "faithful adaptation of the lastest Deluxe version of WIF"
[:D] YEAY [:D]
"Hun skal torpederes!" - Birger Eriksen
("She is to be torpedoed!")
("She is to be torpedoed!")
RE: AI vs non-AI
I think it sure would be easier to make a game based on Wif5 rules than the latest Wif with all the expansions like cruisers in flames, propaganda in flames, beer halls in flames, feminism in flames and the much awaited for Lord of the Rings Return of the King in Flames
Greg
In fact, playing World in Flames Final Edition (WiF FE) in its Classic version is very very very close to playing WiF 5th Edition.
As far as I know about CWiF in the days of Chris Marinacci, it included SiF and PiF from the start and you were not allowed to play without them. But there were a number of people asking for the SiF & PiF counters & rules to be toggled off, to be able to play WiF FE Classic.
Maybe it will be possible with MWiF, so that everyone will be happy.
I, for one, prefer to play with all the lot, except leaders.
Cheers !
Patrice