Has AI been implemented?

[crazydave1066]

I've been following this game for a while but have been wanting for AI to be implemented before purchasing. I see for sure that head to head has been implemented but can't seem to find any updates on AI. Can someone help me?

[davidachamberlain]

ORIGINAL: crazydave1066

I've been following this game for a while but have been wanting for AI to be implemented before purchasing. I see for sure that head to head has been implemented but can't seem to find any updates on AI. Can someone help me?

Not really.

There are some AI-ish features to simplify how the game is played, but this is really rules enforcement. There is some considerable design work that has been done, but the actual AI features will not be delveloped for a while yet. There are other priorities including addressing the most important defects, some optional rules, remaining single map scenarios.

You can see the details on current plans at:

tm.asp?m=4388273

I expect that before any firm commitments are made, there will be further discussion on how it will need to work. Since WIF is a VERY VERY VERY complex strategy game, it will probably need to be incremental.

Until that time, it is still a good game for 2 players (or teams with multiple players) or solitaire which is a very challenging way to play the game.

Dave

[crazydave1066]

Thanks for the quick response! I might just cave and buy the game anyways. How exactly does playing solitaire work out? I can't see how it would be nearly as much fun without the element of surprise.

[davidachamberlain]

ORIGINAL: crazydave1066

Thanks for the quick response! I might just cave and buy the game anyways. How exactly does playing solitaire work out? I can't see how it would be nearly as much fun without the element of surprise.

There are several threads on this.

The game is based on board game, which used dice, but uses the same feature.

Random elements include at least the following:
1) initiative rolls
2) weather rolls
3) surprise rolls (for naval combat)
4) combat rolls (air, ground, naval)
5) turn ending rolls

On top of that, when playing a larger, more complex scenario, you add the complexity and human nature to the mix.

There are 3 major Axis and 5 major Allied powers. It can be really challenging to remember your plans. Often, you will miss something that was important or even less important and have the opportunity to "catch yourself" with that error and exploit it with the other side.

Since the weather can vary every other impulse and the turn can go on for a long time or suddenly end, the strategy becomes much less fluid.

After playing the game for more than a year with both a 4 player Netplay (internet long distance) and solitaire games, I still find it very challenging.

I find it easier to play the multiplayer games since I only have to focus on 1 or 2 major powers from the same side than dealing with 8.


Dave

[AlbertN]

I warmly suggest you buy Matrix World in Flames as game.
It teaches you how to play the game in a way that to try it on Vassal can be tiresome. Because things are enforced. And because the order of play is strict here.

But then in the future I suggest you move on to play on Vassal with the latest edition of the game that is coming out.

[Grotius]

I love playing solitaire, and yes, I routinely forget what the "earlier me" was doing when I was playing another side. Especially if I take a couple days between Major Power turns, as I sometimes do.

It's a great way to learn the game system and basic tactics. I've ordered the new physical Collector's Edition, so I'm trying to make myself predict how MWIF will handle a rule beforehand. Of course MWIF uses RAW7, not RAW8, but the core of the rules is the same.

All that said, I worry that my solo play is instilling bad habits, especially on the strategic level. I constantly read and see things I hadn't thought of here. I may end up trying RAW8 on Vassal, or MWIF with Netplay. But for now, solitaire is awesome!

[Auchinleck]

Picture playing Chess against yourself. Pretty hard to fool yourself using strategy, considering you're your own opponent. An AI is what's desperately needed to make the game worth the expense.

[paulderynck]

Nowadays the chess AIs beat all the humans so why bother with an AI?

[Neilster]

ORIGINAL: paulderynck

Nowadays the chess AIs beat all the humans so why bother with an AI?

Because they can be set for any level of play. As one improves, you can increase the AI's ability. They can also be set to play aggressively or defensively or use unusual openings etc.

Cheers, Neilster

[Neilster]

ORIGINAL: Auchinleck

Picture playing Chess against yourself. Pretty hard to fool yourself using strategy, considering you're your own opponent. An AI is what's desperately needed to make the game worth the expense.

An AI for MWiF is obviously a good thing and hopefully it will eventually arrive. Until it does, however, apart from hotseat and netplay, MWiF is far more suited to solitaire play than chess as it involves more unit types, more potential moves and includes probabilities.

I've had excellent solitaire games. As mentioned above, it's very hard to predict all the "other side's" potential moves. Also, if I for my current side/power I have more than one seemingly good course of action, I'll flip a coin or roll a die to decide. This mixes things up considerably and makes it difficult to predict what will happen.

If you haven't tried a solitaire game, I highly recommend it. It's also a great way to learn the system.

Cheers, Neilster

[paulderynck]

ORIGINAL: Neilster

ORIGINAL: paulderynck

Nowadays the chess AIs beat all the humans so why bother with an AI?

Because they can be set for any level of play. As one improves, you can increase the AI's ability. They can also be set to play aggressively or defensively or use unusual openings etc.

Cheers, Neilster

I look forward to such an AI for MWiF, but forgive me if I don't hold my breath.

[davidachamberlain]

ORIGINAL: Auchinleck

Picture playing Chess against yourself. Pretty hard to fool yourself using strategy, considering you're your own opponent. An AI is what's desperately needed to make the game worth the expense.

As I mentioned in my earlier post, between weather, initiative, short or long turns, searches, attacks, and just plain forgetting your plan for dealing with 8 major powers every turn, there is quite enough "fog of war" to keep the games interesting - even without AI.

Dave

[Cataphract88]

So, it's more like playing chess with yourself, but with dozens and dozens of chessmen on a ginormous board. [:D]

[wodin]

The main reason I got into computer games four s that I no longer needed to find someone else to play games I like. Also I'm useless at most games so never had issues with AIs though I've never set out to game them is find its weakness as some do, would ruin it for me.

[davidachamberlain]

ORIGINAL: wodin

The main reason I got into computer games four s that I no longer needed to find someone else to play games I like. Also I'm useless at most games so never had issues with AIs though I've never set out to game them is find its weakness as some do, would ruin it for me.

Well, maybe someday. It is on the roadmap for some time after the bugs get fixed.

It will be challenging to develop and keep it playable, but Steve is keen to do that.

In the mean time, I still find it to be a good solitaire game, but I do understand your point about your concern about the abilility to play.

This game does very good at managing the rules so you are not able to do anything you are not allowed to. That helps a lot and allows you to focus more on the strategy. However, as others have mentioned, playing against yourself does eliminate some of the challenge, but I still find that between the impact of roles on initiative, turn length, weather, and the results of naval search and combat rolls, and just forgetting to do something in a turn, there are still enough things to keep it challenging (for me).

Dave

[Nicolas II]

Would be great to have a good AI for this game

[Courtenay]

Yes it would, but it won't happen in this decade.

[Nicolas II]

Next decade ? perhaps [:D]

[Neilster]

ORIGINAL: Nicolas II

Next decade ? perhaps [:D]

I can't find it quickly but there was a thread years ago where lots of work on the AI was done. It's the intention of Steve (the developer) to do an AI but he has other priorities first.

He's developed a quasi programming language for it and WiF experts here have thrashed out almost every conceivable strategy for each power.

Hang in there. This game is a long haul. Welcome to the forum, by the way [:)]

Cheers, Neilster

[Joseignacio]

... in case we don't pass out earlier.

I was thinking Bo was funny when saying so, but he had a certain age and he was half serious half joking.

Now I am not so sure it was such a joke, after the 13? years of the previous developer this project seemed a much shorter program, like 5 or 6 years in the initial estimations, but the project is running from 2007? and not even MP for two opponents is fully ready.

Who knows how many years this can still take to end this fully and develop AI? Some people may have died in this 11 years, and some more may not be here when the AI is done...

Sorry to mention it but not speaking on it will not prevent it to happen.