PrimeJunta

Then why don't you like this system? They're very similar, only this one is simpler to implement.

I was, but I removed them from my "at its simplest" example. No reason you couldn't add some, though, e.g. have the AI default to an aggressive script if it determines that it's more likely to work against your hitherto-observed tactics. Yup. It's known as "armed recon." You make contact, then scram once you've found out what the enemy response is. What's your opinion on reputation systems, party influence systems, and alignment/karma systems, by the way?

Funny, I strongly preferred AUJ to DoS, but not for those reasons. I thought DoS was overcooked. I would've enjoyed it more if they had cut most of the action sequences and left only the highlights; as it is, what should have been climactic scenes just got lost in the general jumble of it. AUJ had more quiet moments which made the actiony bits stand out more and ultimately have more impact. Don't worry about LotRO, I haven't ever played an MMO and have no intention of starting.

@Valorian: (1) This system would have no impact at all on combat AI. It would simply determine the composition of the units that appear in combat. The simple crushy/shooty/zappy model would already introduce variety that wasn't there before. Naturally you can take it much further, if you like, with finer-grained profiles for example. (2) In the sketch I proposed, there would not be a separate subsystem for intercepting spies. Instead, we have a quest with bog-standard triggers feeding into party_profile, and an end-state check for encounters without witnesses. Both are dead easy. You don't have to like the idea, but at least do try to base your objections on what's actually in it rather than something you made up on the spot, m'kay?

I haven't played LotRO and don't intend to. I'm pretty sure mopey and whiney isn't what Tolkien intended either. Tragic is hard to pull off well; it's all too easy to turn it into maudlin or unintentionally comic. I'd still like somebody to film the Narn i Hîn Húrin though. Or hell, even the entire freaking Silmarillion from the point of view of Maedhros -- from the rape of the Two Trees and the vow of Fëanor and his sons, to his final crime and suicide after the fall of Melkor. As to the unearthly beauty, it can be done on film, although IMO it hasn't been done much in these cynical times we live in. I won't name any films because they're personally very meaningful to me and I don't want to have them shredded here as someone surely would, tastes and the Internet being what they are.

It is an invitation to game the system. That's the point of a gameplay system. Jeez! And yes, I do contend it would be a relatively low-hanging fruit. At its simplest: (0) Categorize units as crushy, zappy, and shooty. (1) Track who shows up at each battle, how they're equipped, and what abilities they use. (2) Adjust party_profile based on this data (e.g. 20% crushy, 40% shooty, 40% zappy). This happens incrementally over time, and does not automatically track the latest encounter. (3) Adjust enemy_encounter_profile to counter party_profile (e.g. 20% zappy, 40% crushy, 40% shooty). Each encounter would have a pool of shooty, crushy, and zappy units to draw from, and final composition would be determined by the current party_profile. This would already be an interesting-enough system to play against. To make it more interesting, add intelligence-gathering subsystems the player can mess with. For example, if (encounter is in remote location) and (encounter has no survivors) then (don't adjust party_profile). Or write up a quest where the questgiver is also an enemy spy, will have observers following the party as it performs the quest, and will try to pump the party about what they did afterwards. The party's actions will determine how party_profile is adjusted based on this. And so on. But no, this isn't hard, compared to standard stuff like, oh, I dunno, writing branching quests, or coding up a character class with talent trees and what not -- and it will make replays a good deal more interesting as the encounters will be different.

...yet they're the ultimate failures. They see all their works come to naught, and they themselves are doomed to fade away. That's kind of the point.

@JFSOCC I would've made an even smaller intervention: I would just have spread out the questgivers geographically more, with the tougher quests available from people who are further away from the areas you first visit. There were suitable locations; there's no compelling reason IMO so many of those things had to start in the Copper Coronet or the marketplace.

^^^ This is why I play most games a few years after they came out. The ones that are still around are worth playing, and they cost peanuts.

I wasn't actually thinking of scaling at all. I was thinking of changes in enemy tactics and group composition. Suppose you have three categories of units, crushies, shooties, and zappies, where shooties counter zappies, zappies counter crushies, and crushies counter shooties. If you've been fielding lots of shooties, then you'll start seeing the enemy group composition shift towards having fewer zappies and more crushies. They'd still be the same overall strength. The strategic deception you could pull off is to convince the enemy e.g. that you don't have any zappies, then break them out at a critical moment. Yes, that would be pretty lol. That wasn't me. That was AlO3. I would do that differently; instead of scaling the boss encounter in power, I'd change it in composition. If the boss encounter is with Baron Invidius, <minion> and <bodyguard>, I'd have <minion> be Archmage Necrosius, Champion Invictus, or Master of the Hunt Oriol, and <bodyguard> consist of a mix of elite shooties, crushies, and zappies, depending on what you had done up to that point.

Guilty as charged. I do have trouble grasping the concept that strategic deception is not strategy if it produces the desired result. Especially as this strategy involves a meaningful trade-off -- you're making things harder for yourself now in order to gain an advantage later. Same thing as, oh, dual-classing in BG2. You commit to a stretch of seriously underpowered play in order to get major advantages later. Do you consider that exploiting a flaw in the AD&D character system, or playing strategically? If the latter, how is it different from playing against our hypothetical strategic AI?

Keeping your most powerful weapons secret until they're needed is not strategy. OK, glad we got that cleared up.

See post #85.

:sigh: Playing a system designed to be played is not degenerate. It's called "playing the game." What we're discussing here is a gameplay system -- a layer of strategic intelligence to the enemy AI, with associated intelligence-gathering subsystems. It is designed to introduce a new element into the game. It is intended to be played. A strategy is only degenerate if it involves exploiting a flaw in a system in a way to gain a massive and unintended advantage. For example, in the IE games, setting a trap and then luring the enemy into it is not degenerate. It's tactics. However, exploiting a flaw in pathfinding which stops the enemy from reaching you so you can plink it to death with arrows is degenerate. Whether the layer of strategic AI allows degenerate strategies or not depends on how well it is designed and implemented.

Which would be a way cool strategy. Also something entirely believable -- keeping your best stuff in reserve and not letting the enemy find out about it makes total sense. Seriously, I want this feature. It would introduce a whole new gameplay element without all that much effort. Of course you can game it -- that's the point. To make it fun the intelligence-gathering methods would have to be believable, and it would have to be possible to interfere with them in many of the ways listed here -- not using your most powerful abilities until they're really needed, not allowing enemies to escape the battlefield, identifying and squishing spies, dealing with recon... perhaps undercover enemy spies would actually try to recruit you to do stuff for them, posing as your usual questgivers, mixed among the other sidequests. Damn, this has possibilities... and it would fit great into an IE-style game.

@Kjaamor I'm not a filmmaker so I'm quite sure anything I would've done would have been miles worse, so I'm not gonna answer that one. I can explain a bit what I didn't like about Jackson's elves though. The crucial thing with Tolkien's elves is their tragedy. Their story is one of fading and loss that spans millennia. They tread the line between the real and the supernatural; they're distant and superhumanly, unattainably, sublimely beautiful. Seeing a Tolkienian elf should invoke the kind of emotion you would get from walking in a stand of giant costal redwoods on a perfect day, knowing that tomorrow Consolidated Lumber will cut it all down. I don't get any of that from Jackson's elves. None of the majesty, tragedy, certain, inexorable, impending loss. Instead I get camera special effects, Cate Blanchett trying to do contralto in a Brünnhilde bodice, a slightly pudgy Haldir with a carefully powdered nose, Arwen snogging with Aragorn, and oh! the hair. The hair! Jackson's elves are way more like D&D elves than Tolkien's elves. Tolkien's elves deserve better. Jackson is great at action movies, and Tolkien's elves are not action heroes, even the ones that act and do heroic things. It would have taken a director with greater sensitivity to do them justice, even in an action film. Edit: for the record, I loved the movies. Jackson does hobbits great, and the focus was solidly on the hobbits. I just think he dropped the ball on the elves.

At least the way I'm imagining this, yes, pretty much, to all of those questions. If you discover and kill their spies and scouts, don't allow anyone to escape from the battlefield, protect yourself from scrying etc., it ought to interfere with their attempts to counter your strategies. Also offer an interesting new gameplay dimension, and encourage you to switch party compositions from time to time. Of course it presupposes a story where you have that kind of antagonist, not random roving bands of... bandits. It would be jarring if those random bands of bandits changed composition for no reason. Thing is, lots of games do have that kind of antagonist.

I'm sure there could be plenty of explanations for large numbers of undead. Maybe there's a mad god who sticks souls into corpses for her unfathomable purposes. Maybe souls of people who died under traumatic conditions linger and spontaneously animate corpses from time to time. Maybe these types of situations bypass some of the stages of undeath and go directly to skeleton, dargul, or something else. Maybe this, that, and the other. Gromnir is right though that the devs haven't given any such explanations, and it would be good if they did rather than just letting it pass without comment.

@Gromnir, you're right, I did miss the qualifier in your first message. I agree with you re elves and the LotR movies btw. Peter Jackson doesn't get elves. Haldir of Lorien especially was so bad that I breathed a sigh of relief when we saw the end of him. Only Elrond was half-okay and that only because he was half-elven and therefore more believable, if you could get past the Matrix thing. I also agree about "don't do it like Tolkien" and boobplate.

But adapting to the encounters if you don't know what they're going to be is even more fun. Your replays with different types of parties would gain even more variety, as the encounters would change too. I really like this idea. Procedurally generated variety, but in an intelligent way. Of course you could always ruin it by overcooking, but that's true of everything.

That... is a cool idea. It wouldn't even be too hard to do. The counter-strategies themselves could be quite simple; just field the appropriate counter-units to your party members. To make it feel fair and interesting, introduce a suitable time delay so that the counters don't appear immediately, and even better, require some way for the enemy to gather intel on you. You might see shady characters spying on you at the local inn, the enemy doing armed recon, retreating quickly after making contact and so on, with the opportunity to thwart these strategies (kill or capture the spies, allow no-one to escape the field of battle). Each of these thwarts would add to the time lag or add errors to the counter-strategy. Huh. Cool!

Only noticed @Posbi's excellent post now, and I gotta say I concur. There's nothing wrong with having disparate elements there but the whole should make sense. It wouldn't even be all that hard to rearrange them for that. Have a village surrounded by a stockade with the craftsmen and services, nearby the manor with the topiary maze, surrounded by a hedge, overlooked by a tightly-packed, walled keep at a higher location. The amphitheater can be an ancient ruin nearby that you can rehabilitate if you want. The hedge and stockade are enough to keep occasional bandits away, the keep is big enough to hold the villagers, lord, and entourage in case of a full-scale military attack, and in peacetime it all works together. Although it probably is far too late for that. If so, a pity; bit of a missed opportunity there, especially considering Josh is a history and warfare nerd.

I wanna play an aumaua ranger with a pet octopus.

Perfect is a bit of a high bar to clear, obviously, but... NetHack. IMO it has both excellent class differentiation and class balance. A Wizard, Archaeologist, Tourist, and Samurai each require a different strategy to play to maximum effect, the differences carry from early game to endgame, and they're not dramatically different in power at any point in the game.

Haha, yeah, I know what you're talking about. I've given up on some games for similar reasons. They do feed the OCD in us...