If I'm playing WEGO is the AI playing WEGO or RT?

[altipueri]

I might have to plan my moves and then go and get another beer whilst the time unfolds. Are you sure the AI does the same? (And is drinking my beer).

How do I know when my back is turned that the AI hasn't changed its mind when that little blue bar started creeping across the screen?

[Fuser]

Are you sure it's the AI who it's been drinking beer?

[Michael Emrys]

And is it beer or mezcal?

Michael

[altipueri]

I'm just back after playing a few games of squash/racketball, and consumed a few more beers.

But really, how do you know, playing WEGO, that the AI is "playing fair" too?

The NSA would say "of course it is" but they would wouldn't they?

I can see a situation where back in the early days of WEGO it was a cunning way of giving the AI an advantage.

Just because I'm paranoid doesn't mean they aren't after me.

[Paper Tiger]

But really, how do you know, playing WEGO, that the AI is "playing fair" too?

I can see a situation where back in the early days of WEGO it was a cunning way of giving the AI an advantage.

I'm not sure if this is a serious query or you're just enjoying your Sunday evening. But, if you're serious, the AI doesn't cheat

The AI plays what I'd call 'ONE-Go'. That means that all its moves are planned out from the moment you start the game until the mission ends. It is following one huge WEGO plan and it will not deviate from this plan or adapt it in any way whatsoever regardless of the circumstances that develop in the game.

It is YOU who cheat by changing/adapting your plans as circumstances dictate If you want to play the AI on a level playing field, you should plot all your moves from the start to the end of the mission and then click Go/Start and watch what happens, not interfering EVER if things don't work out the way you planned them to. Now you're playing like the AI plays

[Michael Emrys]

The AI plays what I'd call 'ONE-Go'. That means that all its moves are planned out from the moment you start the game until the mission ends. It is following one huge WEGO plan and it will not deviate from this plan or adapt it in any way whatsoever regardless of the circumstances that develop in the game.

Uh, I think you might be exaggerating there just slightly, PT. Granted, the AI selects a plan from the several that have been written for it and attempts to follow it right on through. However, the TacAI does respond to the conditions that the player imposes on it, and will deviate from the plan as necessary in order to follow whatever prime directives are programed into it, such as survival. This means that among other things, AI controlled units may panic and run even if not "planned" to do so.

Michael

[altipueri]

Thanks for the replies.

I think the AI in wargames these days is excellent. Command Ops and Hearts of Iron 3 are two other WW2 games I play and in both you have the AI operating on both sides and you can micromanage or hands-off pretty much as you want.

My initial question grew out of wondering what does happen when the button is pressed.

If I order a tank straight down a road for a minute it will try to do that for the full minute will it not, subject to, say, retreating if morale drops if one or two crew are wounded? What I can't do is intervene - you know suddenly realising that was a really bad idea and telling him after 30 seconds to go left into a field instead.

But is the AI committed too? There is a difference between reacting to things (e.g. panicking and running) and simply changing one's mind (and once the AI has a mind no doubt things will get very weird).

Is the AI situationally aware - e.g. avoiding or planning encirclements?

I still don't feel I'm expressing this well enough. In Michael's reply above he says "AI controlled units may panic and run even if not "planned" to do so." But will they change their "plan" and go round the field in my example - and if so do it at any moment - i.e. RT response or only after a minute; other than in reaction to something specific happening to them?

PS I am actually sober writing this (I might not have been last night). The ancient Persians, as many of you know, would discuss important state matters twice. First drunk and then three days later sober. If they came to the same conclusion sober as they had when drunk then they would go ahead and declare war or whatever.

[Michael Emrys]

I still don't feel I'm expressing this well enough. In Michael's reply above he says "AI controlled units may panic and run even if not "planned" to do so." But will they change their "plan" and go round the field in my example - and if so do it at any moment - i.e. RT response or only after a minute; other than in reaction to something specific happening to them?

I think not...but that's just my guess, it isn't official. I doubt that the game in its present incarnation has a strategic AI, which is why good plans for the AI have to be designed for each scenario.

Michael

[A Canadian Cat]

If I order a tank straight down a road for a minute it will try to do that for the full minute will it not, subject to, say, retreating if morale drops if one or two crew are wounded? What I can't do is intervene - you know suddenly realising that was a really bad idea and telling him after 30 seconds to go left into a field instead.

But is the AI committed too? There is a difference between reacting to things (e.g. panicking and running) and simply changing one's mind (and once the AI has a mind no doubt things will get very weird).

Read the section on AI scripting in this thread (or the PDF manual that shipped with 2.1) http://www.battlefront.com/community/showthread.php?t=110294

Basically the scenario has a authored plan that his chosen from the available plans at the start of the game. At that point the AI operational layer issues orders to its units just like you do. As the turn runs the AI "watches" just like you do and the TacAI controls the reaction of all the units in the game for both sides. Both your and the AI tank driving down the road will react using the same mechanism and either follow your orders successfully or give up and flee - or somewhere in between. At the next turn you and the AI issues orders again and repeat.

Is the AI situationally aware - e.g. avoiding or planning encirclements?

Nope, it just keeps following its plan the best it can regardless of what is happening around it.

[Skwabie]

once the AI has a mind no doubt things will get very weird

skynet is upon us!

sorry folks carry on...

[altipueri]

@ian.leslie

Thanks for that explanation.

I remember looking at JonS' scenario design thread and realising I would never do one but I didn't get as far as the AI section. Now I definitely know I won't do a scenario but I have ever more respect for those who do have the time and dedication to produce them.

I did once modify the Hofen scenario for Command Ops - Battles From The Bulge - but only to change the objectives from avoiding Monschau to massively attacking Monschau. This was my revenge for getting a speeding fine from the polizei there in 1973.