AI Extra Cost When It's Released? (Full Version)

All Forums >> [New Releases from Matrix Games] >> World in Flames



Message

norvandave -> AI Extra Cost When It's Released? (11/21/2013 2:22:07 AM)

Steve, my apologies in advance if this was covered in another thread (searched but no luck). A couple of questions please on this masterwork..

Will there be an add on extra cost when the AI is released?

Do you have a rough timeline for the AI deployment?

Thanks. Looking forward to getting MWiF in the future...




Arnir -> RE: AI Extra Cost When It's Released? (11/21/2013 2:30:01 AM)

From what I have read:
extra cost: yes. No idea what.
Timeline: None that I have seen.

Sorry that I can't be of more help.



Shannon V. OKeets -> RE: AI Extra Cost When It's Released? (11/21/2013 2:52:18 AM)


quote:

ORIGINAL: Arnir

From what I have read:
extra cost: yes. No idea what.
Timeline: None that I have seen.

Sorry that I can't be of more help.

Extra cost, but amount unknown.

As soon as I get this version all spiffed up (in 2-3 months), I'll be working full time on the AI Opponent. But that will take me some time. I'll continue my monthly posts on what I've done on MWIF for the previous month, so you can follow my progress as I go.



Gizuria -> RE: AI Extra Cost When It's Released? (11/21/2013 3:02:58 AM)

I'm very happy to play the game solo as that's the way I had to play the board game. Once I get the two one-map scenarios to play, I'll be in heaven. But full global war? I'm not quite so sure as I tend to get really focussed on one aspect of the war, say the invasion of Russia or the war in the Pacific and would likely feel a pinch having to do the fringe theaters before focussing on the big ones. Once AI comes along, it will be possible to turn China and the Commonwealth over to the AI and play as Germany/Italy and Japan, or play as USA/Commnowealth/China.

And I suspect the AI opponent we get might be a tad more challenging an opponent than some of us are expecting [8D]



HansHafen -> RE: AI Extra Cost When It's Released? (11/21/2013 5:57:07 AM)

I think the AI version will take at least 2 years of focused and devoted work. So two years from now plus all the time Steve has to devote to the current release.

The AI is really going to be a challenge. I don't think there is an AI out there that has to do what it will need to do in this game. (I could be wrong, if so please let me know of it.)

I think Steve has said that the AI development is what he has really been looking forward to. I am very interested to see how it is developed and works in game.



captskillet -> RE: AI Extra Cost When It's Released? (11/21/2013 8:36:45 AM)

I'd have to say the AI for WITP-AE has alot going on also.



pzgndr -> RE: AI Extra Cost When It's Released? (11/21/2013 1:27:41 PM)


quote:

ORIGINAL: Shannon V. OKeets
I'll continue my monthly posts on what I've done on MWIF for the previous month, so you can follow my progress as I go.


Thank you; I was wondering if you would keep up on these.

quote:

ORIGINAL: Fascist Dog
And I suspect the AI opponent we get might be a tad more challenging an opponent than some of us are expecting [8D]


I agree.

quote:

ORIGINAL: HansHafen
The AI is really going to be a challenge. I don't think there is an AI out there that has to do what it will need to do in this game. (I could be wrong, if so please let me know of it.)


The Strategic Command series AI has proven to be quite capable for handling WWII games (ETO, PTO, and Global) at the grand strategy level involving diplomacy, production, research, etc. WiF certainly has a lot more interactive actions during turn impulses to deal with but that's nothing special. I agree the AI will be a challenge to program and optimize, but it's doable as long as Steve and his team stick with it for the long haul. I'm not worried about it, just patiently looking forward to it.



Klydon -> RE: AI Extra Cost When It's Released? (11/21/2013 1:29:50 PM)

One of the things I would caution as far as the time table on the AI and the quality of the AI.

First, Steve is not working on a AI from scratch. There has been a lot of work done on the scripts for the AI already and I imagine there has been other work in progress as well. How much time remains to be done on it is still quite a bit (or they would have finished it and released with the original game).

As noted in other posts, while it may be daunting for an AI to run this game, one of the biggest factors in programming an AI is how good/experienced the programmer is with the game in question in terms of playing ability. Steve knows WiF and is not a rookie at it. That is a lot more than can be said for many programmers who don't know how to play the games they are programming AI's for very well. In addition, Steve has had a lot of input/help from many good and experienced players.

I will be interested to see how it all turns out as I am one of the ones very interested in the AI project since I doubt I can play much with net play, etc. When it is done, I expect a good solid AI based on everything I am seeing in terms of the work put into it by the community along with Steve's comments and experience with the game.



76mm -> RE: AI Extra Cost When It's Released? (11/21/2013 1:53:28 PM)

Has Steve mentioned whether the AI will "cheat" to give itself an advantage, or will it be a straight-up AI?



Shannon V. OKeets -> RE: AI Extra Cost When It's Released? (11/21/2013 4:32:41 PM)


quote:

ORIGINAL: 76mm

Has Steve mentioned whether the AI will "cheat" to give itself an advantage, or will it be a straight-up AI?

No artificial advantages for the artificial intelligence opponent.[:-]

We'll see how this goes as it develops, but presently my thinking is: "Be afraid, be very afraid."[:D]



pzgndr -> RE: AI Extra Cost When It's Released? (11/21/2013 4:43:26 PM)


quote:

ORIGINAL: Shannon V. OKeets
No artificial advantages for the artificial intelligence opponent.[:-]


Players would still be able to toggle various optional rules on/off for play balance, as desired.



henri51 -> RE: AI Extra Cost When It's Released? (11/21/2013 7:48:08 PM)


quote:

ORIGINAL: Shannon V. OKeets

We'll see how this goes as it develops, but presently my thinking is: "Be afraid, be very afraid."[:D]


I am already afraid and I don't even have an opponent!...[X(]

Henri



Greyshaft -> RE: AI Extra Cost When It's Released? (11/21/2013 7:57:20 PM)


quote:

ORIGINAL: Shannon V. OKeets
quote:

ORIGINAL: 76mm

Has Steve mentioned whether the AI will "cheat" to give itself an advantage, or will it be a straight-up AI?

No artificial advantages for the artificial intelligence opponent.[:-]

We'll see how this goes as it develops, but presently my thinking is: "Be afraid, be very afraid."[:D]

I still favor an option for the player to tune the AI up and down to reflect their own skill level. This is an advantage when learning the game.



Dabrion -> RE: AI Extra Cost When It's Released? (11/21/2013 8:07:40 PM)

I believe this will be a rules based AI, which makes it rather hard to formulate a linear scale of [easy...hard]. But for learning something like a full-defense/sitting-duck AI could be available. So that essentially the aggressiveness of the opponent is scaling.



Neilster -> RE: AI Extra Cost When It's Released? (11/22/2013 1:16:31 AM)

Allowing AIs to "cheat" is a bad idea. It warps the game. Better to use optional rules for play balance.

Cheers, Neilster



pzgndr -> RE: AI Extra Cost When It's Released? (11/22/2013 1:16:43 PM)


quote:

ORIGINAL: Neilster
Allowing AIs to "cheat" is a bad idea. It warps the game. Better to use optional rules for play balance.


Ideally, if the AI is really good, then the optional rules should provide sufficient flexibility. But whatever rules are selected, they apply for all players. Since MWiF could have up to 6 players, there could be multiplayer games of less than 6 where some players could be set to AI or human players get replaced by AI. In these cases, depending on how the AI turns out, some specific bonuses and/or difficulty levels for individual AI players may be needed or desired. I don't necessarily see cheats as a bad idea, they're just compensation for AI weaknesses. And for some newbie human players in multiplayer games, this might also be a nice feature to have.



Centuur -> RE: AI Extra Cost When It's Released? (11/22/2013 1:24:44 PM)

A player can choose a lot of optional rules which favours the Allies and don't choose the one which favour the Axis, thus balancing the game towards the allied side. Now, if the AI will be so smart to say: "hey: you want that one? Than I want that one...", than you would be right of course. I don't know if Steve will go as far as to code that...



Omnius -> Artificial Ignorance Not Worth Extra Money (11/23/2013 2:40:03 PM)

Shannon,

Sorry but I never met an Artificial Ignorance worth a squat and for such a complex and complicated game such as MWiF I certainly don't see any chance of creating an AI worth a squat, much less one to pay for. I'm glad that MWiF got released sans an AI, forces everyone to learn how to play solitaire and learn each country. Yes it's time consuming to play but with all of the many bookkeeping functions automated the game is much easier to play than the old board game.

As far as being afraid of your AI all I can do is LMAO. [:D] Unless you give the AI serious advantages in pumping up it's production multiple it won't be a challenge. You'd have to script complex moves like taking Norway or the many Pacific amphibious invasions of Japan early on, or the US march westwards through the Pacific. Unfortunately scripting can be disrupted by unusual events like Japan taking Pago Pago early on.

I'd much rather see time invested in bringing us America in Flames and Patton in Flames additions to MWiF, now those would be worth paying extra for.

Omnius



Numdydar -> RE: Artificial Ignorance Not Worth Extra Money (11/23/2013 3:28:26 PM)

I find it absolutely hilarious that the game is only out a little over two weeks and people are already complaining about an AI and what it might or might not be capabile of [:D]. There are many people that were saying similar things about the game itself. That it would be crap when it came out, unplayable, etc. Just browse through the When thread and you can quite a few of these. And yet here we are with a pretty polished game and one that is definately NOT crap and unplayable. So now that people cannot complain about the game itself, lets start whining about the AI.

There are a lot of AIs that cheat, or are not very good. However, there are many things in the world today that were thought the same thing of, like long distance travel, home movies, etc. Yet today these are much better and vastly improved over the originals. As technology advances, including in developing an AI, saying that the AI is going to be terrible before it has even been released is making the assumption that the way AIs are now are the ways the AI will always be.

This is not the way these things work. AI development is occuring at a rapid rate and I fully expect that at some point, in specific cases, AIs will exceed us. Eventually AIs may surpass us in the general as well.

Will this happen for WiF? Doubtful. But I have a lot of confidance that the AI for WiF will be better than the AIs for games that were released this year. The main reason for this confidance is that, unlike other AI development, the AI for WiF will be developed in partnership with expert WiF players and the designer of the game itself. To me that along should make the AI a lot better than your typical game.

Until the AI actually is released and can see for ourselves how it does, we will just have to disagree on how good it will be. Of course if you REALLY want an awesome AI, then join the beta team that will be helping to test it. If you don't and then complain later about how terrible the AI is, you only have yourself to blame. So it is up to the WiF community as a group to develop an awesome AI. So you can be part of the problem or part of a solution. Your choice.



brian brian -> RE: Artificial Ignorance Not Worth Extra Money (11/23/2013 4:34:15 PM)

I think that the more optional rules are left truly 'optional', the harder it will be to program a good AI. Not in terms of being impossible, but in terms of how long it will take to create one. And the longer it takes, the more unhappy people will be. So the best route will be to make some fixed decisions on which optionals to use or not use in the AI. Which will make people unhappy.

I would also think a "Learning Level" AI setting would be entirely possible and easy to create in the two introductory scenarios, which is where people learn to play the game. After that I'm not sure adding that to the To Do list would gain that much.



Rasputitsa -> RE: AI Extra Cost When It's Released? (11/23/2013 5:00:10 PM)


quote:

ORIGINAL: Shannon V. OKeets
quote:

ORIGINAL: 76mm
Has Steve mentioned whether the AI will "cheat" to give itself an advantage, or will it be a straight-up AI?

No artificial advantages for the artificial intelligence opponent.[:-]

We'll see how this goes as it develops, but presently my thinking is: "Be afraid, be very afraid."[:D]

Steve is the only person in a position to make a judgement at this stage. Wait and see. [:)]



Dabrion -> RE: AI Extra Cost When It's Released? (11/23/2013 5:25:24 PM)

I would freeze the AI project and concider put it into the public domain. That will decrease development time drastically, for MWiF and for a computer opponent. The gained dev-time could be spent on improving the boardgame sandbox and UI/UX!



Shannon V. OKeets -> RE: Artificial Ignorance Not Worth Extra Money (11/23/2013 5:27:09 PM)


quote:

ORIGINAL: brian brian

I think that the more optional rules are left truly 'optional', the harder it will be to program a good AI. Not in terms of being impossible, but in terms of how long it will take to create one. And the longer it takes, the more unhappy people will be. So the best route will be to make some fixed decisions on which optionals to use or not use in the AI. Which will make people unhappy.

I would also think a "Learning Level" AI setting would be entirely possible and easy to create in the two introductory scenarios, which is where people learn to play the game. After that I'm not sure adding that to the To Do list would gain that much.

As for "learning level", that can be partially accommodated by probabilities of different strategies & tactics. For instance, the more predictable the AI Opponent is, the easier it is to defeat. Continuing that thought, if the AI Opponent always chooses an inferior tactic, then it will be weaker. A simple example: setting the odds ratio for what is a good attack by the Germans in Barbarossa can make the Germans overly aggressive or overly fearful.

Note that none of the above requires anything more that changing the value of a couple of parameters.

As everyone should know, the true difficulty in writing the AIO is making it play well. Dumbing it down won't be a problem.



Shannon V. OKeets -> RE: AI Extra Cost When It's Released? (11/23/2013 5:42:55 PM)


quote:

ORIGINAL: Dabrion

I would freeze the AI project and concider put it into the public domain. That will decrease development time drastically, for MWiF and for a computer opponent. The gained dev-time could be spent on improving the boardgame sandbox and UI/UX!

If you were going to build a 30 story condominium, would you want hundreds of untrained people to help you? For the AIO, some helping hands would be nice, but involving everyone who has a stray idea to write a script would be foolish. Sorry, but that's a solid truth in my opinion - see the next paragraph.

There must be an underlying structure and organized implementation layered onto that structure. That means the group has to work as a team, not a mass of individuals with each working on what he thinks is a good idea. All of that implies each person has an understanding of the structure and his role in the overall implementation. There will be a need to do some learning before writing scripts.

On the other hand, I am always willing to read other players ideas about how to play WIF well. The existing AI Opponent threads, one per major power plus dozens for setting up the minor countries, attest to that. Writing in those threads does not require any particular skill other than knowing how to play the game. You don't even have to be a genius at playing WIF, since many great ideas have strange points of origin.



Greyshaft -> RE: Artificial Ignorance Not Worth Extra Money (11/23/2013 6:07:37 PM)


quote:

ORIGINAL: Shannon V. OKeets


As for "learning level", that can be partially accommodated by probabilities of different strategies & tactics. For instance, the more predictable the AI Opponent is, the easier it is to defeat. Continuing that thought, if the AI Opponent always chooses an inferior tactic, then it will be weaker. A simple example: setting the odds ratio for what is a good attack by the Germans in Barbarossa can make the Germans overly aggressive or overly fearful.


yes, including that style of tweakable parameter (plus recommending the list of which optional rules favor Axis or Allies) should do the trick.



Dorb -> RE: AI Extra Cost When It's Released? (11/23/2013 8:13:19 PM)


quote:

ORIGINAL: norvandave

Steve, my apologies in advance if this was covered in another thread (searched but no luck). A couple of questions please on this masterwork..

Will there be an add on extra cost when the AI is released?

Do you have a rough timeline for the AI deployment?

Thanks. Looking forward to getting MWiF in the future...





Yes, this has been told up front ... just like the books and the maps!



Greyshaft -> RE: AI Extra Cost When It's Released? (11/23/2013 8:27:04 PM)


quote:

ORIGINAL: Shannon V. OKeets
...
On the other hand, I am always willing to read other players ideas about how to play WIF well. The existing AI Opponent threads, one per major power plus dozens for setting up the minor countries, attest to that. Writing in those threads does not require any particular skill other than knowing how to play the game. You don't even have to be a genius at playing WIF, since many great ideas have strange points of origin.


I would consider it essential to have a few klutz setups that pop up now and again. This would reflect reality...

"Hey Adolf... look over there at the Ardennes ... I think the French forgot to garrison a hex"
"Hey Josef... the front with the Germans is secure. I've got all of our troops adjacent to the border with the aircraft parked in nice neat rows in the middle of the airfields".
"Hey Benito ... lets declare war on the CW now before all of our transports are safe ... who needs merchant ships anyway?"

So even the newest of the noobs can assist by writing a half-baked, strategically-suspect tactically-indefensible setup which Steve might assign a 1% chance of appearing in a game. Just imaging the AAR when it does appear [:D]



Aurelian -> RE: AI Extra Cost When It's Released? (11/23/2013 8:37:45 PM)


quote:

ORIGINAL: Dabrion

I would freeze the AI project and concider put it into the public domain. That will decrease development time drastically, for MWiF and for a computer opponent. The gained dev-time could be spent on improving the boardgame sandbox and UI/UX!


And just who is going to fix bugs when they come up?

Too many cooks spoil the soup after all.



76mm -> RE: AI Extra Cost When It's Released? (11/23/2013 8:46:04 PM)

No disrespect intended toward Steve at all, but I find it difficult to believe that this will be the first commercial wargame ever to have a competent AI, especially when you consider the implications of the various combinations of optional rules which may/may not be activated. Maybe possible OK for shorter scenarios but hardly for the campaigns



celebrindal -> RE: AI Extra Cost When It's Released? (11/23/2013 9:09:37 PM)

quote:

ORIGINAL: Shannon V. OKeets


quote:

ORIGINAL: Dabrion

I would freeze the AI project and concider put it into the public domain. That will decrease development time drastically, for MWiF and for a computer opponent. The gained dev-time could be spent on improving the boardgame sandbox and UI/UX!

If you were going to build a 30 story condominium, would you want hundreds of untrained people to help you? For the AIO, some helping hands would be nice, but involving everyone who has a stray idea to write a script would be foolish. Sorry, but that's a solid truth in my opinion - see the next paragraph.

There must be an underlying structure and organized implementation layered onto that structure. That means the group has to work as a team, not a mass of individuals with each working on what he thinks is a good idea. All of that implies each person has an understanding of the structure and his role in the overall implementation. There will be a need to do some learning before writing scripts.

On the other hand, I am always willing to read other players ideas about how to play WIF well. The existing AI Opponent threads, one per major power plus dozens for setting up the minor countries, attest to that. Writing in those threads does not require any particular skill other than knowing how to play the game. You don't even have to be a genius at playing WIF, since many great ideas have strange points of origin.


I'd be willing to PM a team of folks working on it if you need.

As for AI, i'd prefer to see it going faster than PBEM.. i'm just not satisfied that PBEM lends its self well to something that has so much interaction between folks in various phases...



Page: [1] 2   next >   >>

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
0.71875