Mode of Play

[pzgndr]

I think there is alot more to the advantage of a computer game than just an AIO.

I agree, and that's why I specifically mentioned the AI in addition to PBEM and netplay capability. They are ALL important. I support ALL of them. I would not presume to tell another player how he should play his game, nor try to deny another player features that are beneficially to all of us. And yet, there are others with an attitude about something who persist in trying to limit or restrict a computer game's development. It's a silly thing, and thankfully Matrix continues to plod forward committed to trying to satisy all gamers, regardless of whatever mode of play they choose. [:)]

[Shannon V. OKeets]

ORIGINAL: Greyshaft

ORIGINAL: Jagdtiger14
Regardless, there will be an AI...and I hope it will be nothing more than a minimal tutorial strength AI.

I hope the AI is far better than tutorial strength.

An AI will be extremely useful when there are insufficient players for the full Global War and no-one wants to play (say) Russia.

An even better use for the AI is when three players take one side and they let the AI take the opposition. In that case there is no possible game-generated animosity between the players but instead there is a real opportunity to enjoy fighting a game for the Common Good... (until 1945 when another player gets too close to the last victory hex - that's when the gloves come off).

For the initial release, the AI Opponent will play one side against a single human player. That's all. I am also focusing on only 4 scenarios: Barbarossa, Guadalcanal, Fascist Tide, and Global War. The remaining 7 scenarios will be post release (at no additional cost).

[Anendrue]

ORIGINAL: Jagdtiger14
I hope it will be nothing more than a minimal tutorial strength AI.

The AI is going to be in the game so I don't know why you would want such a thing! I hope you realize a bad or minimal AI will hurt reviews and the potential customer base from expanding to new market segments. The more successful MWiF is the more likly we are to get expansions and additions to the game. Of course if you are a competition plant or just a negative type of individual we will just disregard your statement as one more for the trash can.

[arehb]

Will the AI be updated after release, or based on scripts that can be edited? Wouldn't be very cool if you could discover holes in the AI strategies/tactics and use those over and over again.

It would be nice if one eventually could get complete control over what countries are played by AIs and what by players and mix them as one sees fit.

[Shannon V. OKeets]

ORIGINAL: arehb

Will the AI be updated after release, or based on scripts that can be edited? Wouldn't be very cool if you could discover holes in the AI strategies/tactics and use those over and over again.

It would be nice if one eventually could get complete control over what countries are played by AIs and what by players and mix them as one sees fit.

My contract with Matrix Games states that I will support MWIF for at least 18 months after its initial release.

[brian brian]

I really look forward to the Barbarossa AI. My very first computer wargame was such a let-down ... Atari's "Eastern Front", circa 1983. That AI always wins by continually increasing the strength of it's pieces. With an MWiF AI, it also won't be so hard to find someone to play the Russians in that scenario.

[gridley]

ORIGINAL: arehb

Will the AI be updated after release, or based on scripts that can be edited? Wouldn't be very cool if you could discover holes in the AI strategies/tactics and use those over and over again.

There will always be holes in any AI. It's up to you to make your own house rules not to exploit them.

[Jagdtiger14]

abj,

You dont know me, but there are a few here that know me personally...Steve Balk for one. I am an avid WiF player and am in no way a competitor plant. For the record I am a very optimistic guy. As for the AI...it all depends on what your meaning of "good AI" is...for me a fantastic AI is one that is tutorial strength...if Steve can pull that off, he is a hero. Demanding an AI better than that AND if those demands move the release date back, then I consider you a greedy f*#%. You are getting your AI...be happy...now dont make this an impossible release...let us play the game. If you want a more advanced AI, wait for the later add ons.
C

ORIGINAL: abj9562

ORIGINAL: Jagdtiger14
I hope it will be nothing more than a minimal tutorial strength AI.

The AI is going to be in the game so I don't know why you would want such a thing! I hope you realize a bad or minimal AI will hurt reviews and the potential customer base from expanding to new market segments. The more successful MWiF is the more likly we are to get expansions and additions to the game. Of course if you are a competition plant or just a negative type of individual we will just disregard your statement as one more for the trash can.

[pasternakski]

ORIGINAL: Jagdtiger14
As for the AI...it all depends on what your meaning of "good AI" is...for me a fantastic AI is one that is tutorial strength...if Steve can pull that off, he is a hero. Demanding an AI better than that AND if those demands move the release date back, then I consider you a greedy f*#%. You are getting your AI...be happy...now dont make this an impossible release...let us play the game. If you want a more advanced AI, wait for the later add ons.

Okay. So at least three-quarters of the potential customers for this game are greedy whatever-that-is.

See, those who play against the AI don't demand much. They never have, clear back to the old SSI days.

They buy because they want to play against the computer, not against people. That was what computer wargame companies promised to deliver clear back in the infancy of computer wargames. They have always accepted crap AI and computer cheating and don't care. They get by, like the same proportion of cardboard-and-paper gamers who devised (and devise) ways to play against themselves.

Nothing against hotseat, TCP/IP, PBEM. I play a lot of that, and it's fun.

It just isn't what the hobby promised. If you want a computer wargame business that caters exclusively, or even primarily, to the human-vs.-human crowd, you had better be prepared for low sales and eventual insolvency.

It ain't that way because I say it is, it's that way because that's how it is.

[coregames]

ORIGINAL: pasternakski

They buy because they want to play against the computer, not against people. That was what computer wargame companies promised to deliver clear back in the infancy of computer wargames. They have always accepted crap AI and computer cheating and don't care. They get by, like the same proportion of cardboard-and-paper gamers who devised (and devise) ways to play against themselves.

I agree that MWiF needs a sufficient AI in order to succeed to the degree the franchise deserves to succeed. The issue seems to be what qualifies as sufficient. I look forward to playing against the initial release of MWiF, as do many of us I'm sure, with the idea that we can suggest improvements to the AIO. Since Steve is on board for 18 months after release, I believe MWiF will have a surprisingly strong AIO within that timeframe. This is not to say it will consistently give a strong experienced player a run for his or her money in a global war scenario.

Remember that WiF has a bidding mechanic which would prove helpful in providing the computer with a handicap if the player's strength merits it, and this does not require affecting dice rolls, providing extra production, or changing the unit characteristics in any way. This is not cheating, as humans often use bidding to determine which country each player gets to control, which we have found does add a balancing factor, especially when introducing a new player to the game.

Still, for the maximum challenge in a long scenario, I do believe we will discover that actual strong players are the only option. If the Japanese can't program a computer to play the game of go at a professional level, the variables involved in WiF seem to limit the potential strength of the AI.

Strange as this may sound, fortunately, I have made a lot of mistakes when playing WiF over the board, and I look forward to making plenty more against the computer. The program doesn't have to play champion-level World in Flames to challenge me, or most of those who will play this game. I'm sure it will be a great game, and all this anxiety about the AI will be replaced with joy once the release is upon us.

[Shannon V. OKeets]

ORIGINAL: coregames

ORIGINAL: pasternakski

They buy because they want to play against the computer, not against people. That was what computer wargame companies promised to deliver clear back in the infancy of computer wargames. They have always accepted crap AI and computer cheating and don't care. They get by, like the same proportion of cardboard-and-paper gamers who devised (and devise) ways to play against themselves.

I agree that MWiF needs a sufficient AI in order to succeed to the degree the franchise deserves to succeed. The issue seems to be what qualifies as sufficient. I look forward to playing against the initial release of MWiF, as do many of us I'm sure, with the idea that we can suggest improvements to the AIO. Since Steve is on board for 18 months after release, I believe MWiF will have a surprisingly strong AIO within that timeframe. This is not to say it will consistently give a strong experienced player a run for his or her money in a global war scenario.

Remember that WiF has a bidding mechanic which would prove helpful in providing the computer with a handicap if the player's strength merits it, and this does not require affecting dice rolls, providing extra production, or changing the unit characteristics in any way. This is not cheating, as humans often use bidding to determine which country each player gets to control, which we have found does add a balancing factor, especially when introducing a new player to the game.

Still, for the maximum challenge in a long scenario, I do believe we will discover that actual strong players are the only option. If the Japanese can't program a computer to play the game of go at a professional level, the variables involved in WiF seem to limit the potential strength of the AI.

Strange as this may sound, fortunately, I have made a lot of mistakes when playing WiF over the board, and I look forward to making plenty more against the computer. The program doesn't have to play champion-level World in Flames to challenge me, or most of those who will play this game. I'm sure it will be a great game, and all this anxiety about the AI will be replaced with joy once the release is upon us.

Joyful release?[X(]

[SamuraiProgrmmr]

Ain't Sublimation Wonderful
 
<run, duck, cover>

[BeastieDog]

I play AI initially until I get skilled or bored. If I like the game I study it using hotseat and then move on to PBEM. I play WitP. Waiting for WitP-AE. I've never played the board game, but if MWIF gets good opinions at release, I'll give it a try.

I also find AI helpful to test different strategies.

[coregames]

ORIGINAL: Shannon V. OKeets
Joyful release?[X(]

If this is the build-up to a joyful release, then we all put Sting to shame with our staying power! Does that mean that Chris Marinacci's beta was foreplay?

[brian brian]

ORIGINAL: Shannon V. OKeets

For the initial release, the AI Opponent will play one side against a single human player. That's all. I am also focusing on only 4 scenarios: Barbarossa, Guadalcanal, Fascist Tide, and Global War. The remaining 7 scenarios will be post release (at no additional cost).

I was thinking about this the other day. I suggest working up the necessary LAIO inputs to run the Decline & Fall scenario. Just as people (and the AI) use Barbarossa and Guadalcanal to learn the game, working on that scenario with the AI would be an excellent test of how well it all works when the AI is in charge of the maximum number of pieces at the end of the game, rather than the minimum number of pieces at the start of the game. For testing this would be a lot easier than playing Global War or Fascist Tide all the way to 1944 (very, very time-consuming).

I need to play through some Decline & Fall sometime myself, to get in some practice running one of the trickiest military operations of all-time: Overlord.

[Shannon V. OKeets]

ORIGINAL: brian brian

ORIGINAL: Shannon V. OKeets

For the initial release, the AI Opponent will play one side against a single human player. That's all. I am also focusing on only 4 scenarios: Barbarossa, Guadalcanal, Fascist Tide, and Global War. The remaining 7 scenarios will be post release (at no additional cost).

I was thinking about this the other day. I suggest working up the necessary LAIO inputs to run the Decline & Fall scenario. Just as people (and the AI) use Barbarossa and Guadalcanal to learn the game, working on that scenario with the AI would be an excellent test of how well it all works when the AI is in charge of the maximum number of pieces at the end of the game, rather than the minimum number of pieces at the start of the game. For testing this would be a lot easier than playing Global War or Fascist Tide all the way to 1944 (very, very time-consuming).

I need to play through some Decline & Fall sometime myself, to get in some practice running one of the trickiest military operations of all-time: Overlord.

Adding to my task list is not on my task list.[:D]