We hear the phrase “If it’s not broken, don’t fix it” often in the MMO genre, and sometimes you have to ask “Do we know what broken is?” I’d like to throw out that I believe mob AI, in particular agro rules, are in fact ‘broken’ in many of today’s MMOs, but they are accepted because they don’t come out and scream “BROKEN!” like falling through the world or other obvious errors.
Let’s start with the most basic of mob rules: If I see you, I attack you. We know this as ‘agro radius’, and we simply accept it. Why? Why is every creature you encounter outside of a squirrel immediately interested in nothing but running up to you and bashing you until someone dies, especially when 99% of the time it’s the mob running towards its inevitable death?
I get that some creatures are brainless and only interested in violence, so maybe that line of thinking works for them, but what about ‘smart’ monsters? Why would a might dragon be interested in attacking a clearly weaker player just passing by? Beyond just a lack of realism (which should only be considered when it helps gameplay, not when it hinders it), a dragon programmed in that way is basically griefing everyone who passes it, killing them only for the sake of killing them, over and over if given the chance.
I think part of the problem is many of the design decisions in an MMO come from their solo player RPG history. In a tightly scripted and paced RPG, mobs rushing you when they see you usually makes sense. The story is designed to have you invade an orc lair at point A, defend a village at point B, and slay a dragon at point C. But while that works in a scripted single player game, it does not translate perfectly to a virtual world (themeparks, especially solo-hero based ones, are somewhat of a gray area here). The designers don’t know when you will be slaying that dragon, or with who, yet the same rule of “if I see you, I attack” still applies.
The traditional solution to this problem is to place only the monsters you want a player to fight in a certain zone/area. If you are level 50, you go to the level 50 zone and fight level 50 monsters. Obviously in a seamless virtual world this gets tricky the more freedom you give your players, but again the solution only goes so far as to say ‘fight what you can, move on if you can’t’. It’s the reason starter areas are populated by weakling mobs, and only the highest areas get the really cool stuff.
What if you placed the hardest mob in the game right in the starter area?
Everyone joining the game would get bind-camped by the impossible mob and quit shortly after, but only if we play by today’s rules.
If you change the dragon’s agro rules to only attack when someone attacks him, or when someone enters his lair, you can place him in that starter area and allow new players to see him flying around without being ‘griefed’ by such a powerful mob. The mob is ‘real’, so anyone can attack him, and high-level characters will eventually make the trip back to challenge him (with new players able to watch the show), which gives the area some multi-purpose. New players on the other hand will quickly learn to not attack him and to avoid his lair, all while enjoying seeing him flying above, perhaps even occasionally lighting a weak goblin on fire or swooping down for a snack. As a new player, one of your eventual goals (get strong enough to slay the dragon) is put directly in front of you, connecting you more directly with the game, all while giving you some nice eye candy as you bash goblins. Plus now all that time and effort put into creating and animating the dragon is enjoyed by almost everyone, even those who never make it to the ‘end’.
The important point to remember here is that mobs exist to be killed, rather than to kill the player. Of course challenge is important, and in the right situation dying can be more interesting than rolling over countless creatures, but at the end of the day a mobs job is to die, so you don’t want to make them frustratingly hard or behave in a way that makes players avoid the ‘hassle’ of killing them. At the same time, keeping a mob interesting to kill is a great way to keep a player coming back, especially if it’s not a given that the mob will die every time, or at least in the same way.
Hardcore Casual 
EQ2 had elite dragons in corners of Antonica and the Commonlands. They weren’t end-game mobs, but they would leave you alone if you left them alone….
The original EverQuest quite often had raid mobs mixed right in with lower level areas; Gorenaire in the Dreadlands, for instance, would chase low level players around (but with enough warning so they could get away) until a raid showed up to take her down. There were many such examples.
So it’s been done before, but this is something not often done today, and I don’t know why.
I generally think this simple AI is the foundation of most MMO’s because essentially people are far inferior to computers when it comes to min/max behavior in a structured environment…. as “Big Blue” demonstrated even grand master chess players can be beat by a computer operating without human limitations… Balancing AI would be even harder than class balancing most games … new players almost certainly require basic steps they can learn for victory when learning a game for the first time. WOW and other games have never changed the AI because it was simply easier and less time consuming to increase the difficulty through making the tolerances for error in encounters lower. The players love it since memorization and execution is easier than adaptation on the fly. I suspect better AI would appeal to PVP players looking at PVE more than it would to the PVE players themselves.
Syncaine,
I don’t mean to hi jack this topic but I would LOVE to hear your thoughts on something regarding darkfall:
http://forums.darkfallonline.com/showthread.php?t=228861
This thread is about the Sea tower fight from last night. It is rife with the chronic zerg vs us arguments so common on DF.
What I’d love to hear are your thoughts on the whole: skill/stats vs leadership/organization vs ‘Zerg’/small force dynamic.
This topic really only applies to what I see as the two most complex sandbox games out there: Darkfall and EVE. The complexity of PVP in these two titles allows for the above mentioned dynamics to play a significant role.
Sorry for the de-rail. I hope you find some time to post thoughts about how these dynamics effect the MMOs they occur in.
V/R
Dia
I missed that battle last night (I don’t play on Mondays), sounds like it was a blast. I know VAMP/NEM have been planning to hit the south tower for a while now, since no one really fights us over the north one.
I know this, the guys in my alliance are highly skilled, have jacked characters (only a few are crazy strong, but most are above average), and followed orders better than a real world army. An SG only has to ask vent to go silent once, ever. If someone asks to clear vent because of some PvP, it gets said once. When orders are given, people don’t question or debate them, you go where you are told, and more often than not the orders given are the correct tactical decision thanks to EU experience. I know for a fact most alliances don’t operate like that, and it shows on the field.
When we bring 40, I can see how others feel like we brought 100. We feel confident going into a 20v50 against most. And thanks to some dedicated farming (fire dragon and such), if we need, we can bring out the big toys and really bring the pain.
So it’s not one thing (zerg, skill, toons, gear), but a combination of all of them that makes the alliance deadly. That said, I know nothing would make leadership happier than getting a good fight. We don’t enjoy a slaughter any more than the dead do.
Ugh reading all that crap and watching that video makes me want to resub. That and the DF subforum on the Murder Turd forums is locked to me cus I’m not actively subscribed :(
Okay well I imagine a big dragon like that would have AE attacks amirite? What happens when one of those hits all the little newbies? What about level 1s griefing by pulling the dragon over to newb town for a nice AE rapefest? Well you could make it so it only hurts people who have attacked it (or healed people the dragon is attacking) but then you’ve got to add a whole nother layer to the roll teh dice system, sure it would be right on top (do i hate this player? yes. did he dodge/block/resist me? no. how much dmg do i do? 300. how much does he resist/mitigate? 27, etcetc).
Now I don’t know shit about how adding something like that would actually affect the whole system, but I know someone doesn’t want to add it to an already finished game.
I was half expecting you to go more into agro radii along the lines of, if that mean orc is over there and i can see him and hes looking right at me, why isn’t he attack? Aka huge agro radii.
One thing that was kind of cool about DAoC near the end there was when they changed the starting areas because “the dragons attacked”. In midgard there was ice all over vasudheim and occasionally, iirc, the mid dragon would fly over top. Of course at this time Vasudheim was no longer a starting area, but it was just on the other side of Jordheim, which was next to the new catch all starting area.
Tipa is on the money. Most of what you describe was in Everquest ten years ago. EQ even had creatures that only hated one race (Aviaks in South Karana were neutral to all player races except Trolls, for example, which caused some hysterically funny moments in groups at the Birdhouse), or others which were only “almost always” non-aggressive (elephants in South Karana). Then there were the monsters that loved you when you looked like you but hated you when you used any illusion, or a specific illusion, or vice versa.
Everquest had an agro-faction system that you couldn’t always trust because some monsters actually lied about their feelings toward you (the Ogres in North Karana, for example). And of course it had high-level monsters in low-level zones, guards that hated and would kill player characters they didn’t approve of.
I could go on quoting variations all night.
The reason that you don’t see these things much any more is simply that most players didn’t appreciate them. All the above made Norrath a fascinating place for those of us who appreciate mystery and quirkiness over efficiency, but sadly we seem to be in a tiny minority nowadays.
Subsequent games did everything they could to address what were seen as flaws in EQ’s model. Huuge efforts were made to make things more consistent, better-explained, easier to understand and, above all, less threatening. It’s a trend that still appears to be growing, as almost any danger or inconsistency is now treated as a design error and corrected almost as soon as a player points it out.
There may be room for some cosmetic variety in mob AI in a mass-market MMO, but as soon as it actually inconveniences players in any way you can forget it.
Yea I all but skipped EQ1, went from UO to AC and then to DAoC. But reading the above only increases my WoW (the general term, not that actual game hate+1. GTFO of my genre tourists!
Bhagpuss is correct and it’s one of the reasons why I will liken EQ1 to more of a sandbox and not a theme park. Factions were just a huge part of the game and I loved that my Necro was an Iksar (hated by all races). It let me kill just about anything I wanted to in game, including NPC’s. Not mobs but actual game NPC’s, you know npc’s like vendors, trainers and guards.
Sadly, the days of the complex older MMORPG’s are gone (I’m hoping not forever) as designers make MMORPG’s more like a game of checkers rather than chess.
I think the inconvenience you mention at the end is really long and short of it. I know the common argument is “I work all day, I don’t want to come home and fight with a game/”work” again, I just want to ‘have fun.'” I really am sympathetic to the spirit in which this is said, but I think the “have fun” is the the moment of disconnect, not the “i work all day” part. Fun is so subjective that it can’t really be pinned down, but I think complexity, nuance, subtlety, etc are things that most people, for whatever reason, associate with things that are “hard” rather than “fun,” where as I couldn’t disagree more.
I love doing the NY Times crossword puzzle, I love spending free time listening to lectures on itunesU, and I love to immerse myself in a fantasy (I say that in the most generic way possible, to include sci fi, and just not-real in general) world full of political intrigue, difficult decisions, and so forth. To me that is a thousand times more “fun” than simply seeing my numbers associated with a character go up (experience, stats, damage, whatever). Now, I like mindless fun sometimes too(Torchlight, Team Fortress 2, as current examples), but mindless fun isn’t the only kind of fun their is.
Problem is you’d need to have aggro anyways, otherwise you get “Look at me mom, I’m standing next to the dragon!”
The more imposing a monster, the less safe it should be, and I don’t think most high level monsters like dragons would let anyone get in range of their treasure. They’d surround themselves with deadly lairs and minions.
Plus, with “safe” aggro you’d wind up making the players the aggressors. You don’t want to make the players the bad guys any more than they already are.
What’s wrong with standing next to a dragon as a noob? It’s an intelligent creature, why would it feel the need to one-shot someone who is of no threat or benefit to them? I mean it depends on the lore behind the dragon or whatever creature, but I don’t think we just have to assume dragon=evil=insta-agro. It would protect it’s lair and treasure, but if it’s out just flying around or looking for a goblin snack, it could do so without fireballing noobs.
You have two conflicting purposes:
1. Safe for noobies
2. Should be killed by high levels.
Can’t see how both fit together. If you can walk right up to it and its more or less passive unless provoked, why encourage players to kill it? If it should be killed, why is it safe to walk next?
Unless you want to play something like factions, where a dragon is safe for one faction and an enemy for another, you should really have mobs meant to be killed be aggressive.
World first, game mechanics second. Are dragons aggressive above other things in the setting? Are the smart enough to know the difference between a threat and not. Are they animalistic such that they simply attack whatever they see.
Whats important is that the mechanic fit the setting you’ve made. Once you’ve established what Dragon’s ARE in your setting, then you can determine their mechanics.
I’d go the other way. Do you want dragons to be semi-aggressive and intelligent? Come up with the mechanics, then create lore to fit them. Stay consistent with the lore, but IMO lore should never determine game mechanics. Same goes for realism. Does something being realistic help the game, if so keep it. More fun to make it unrealistic, create some lore to explain it.
I have to disagree. To me the world is the single most important part of the experience in an MMO. As a sort of example, I think one of the reasons the Lord of the Rings is such a great fantasy story is because Middle Earth wasn’t created for Lord of the Rings, lord of the rings was a story that happened in this larger world of Middle Earth that Tolkien imagined.
It never feels like he’s created just enough of Middle Earth that if you could look over the next hill you’d see 20th century England, instead you’ll see Middle Earth stretching out in front of you, even if it has nothing to do with that main story line.
In game terms, I think you need to take this approach as well, or you risk a very gamey feel. I don’t consider this to be a theme park v. sandbox issue either. A theme park can have a very well realized (but generally unfortunately static) game world, and a sandbox can have a very generic or contrived feel.
If the lore is to be constrained to what game mechanic feels right, I think you are restricting yourself a lot.
LotRO is different though, because of how important the lore is to that game. In that situation I would say lore>mechanics. I think that just means Turbine has to work that much harder to make the mechanics fit AND be fun to play.
But in something like DF/WoW, why would you limit your mechanics because of the lore, especially if we are talking pre-release here. Obviously people know how I feel about the space goats and undead paladins in WoW, but it can go too far as well.
Well, in WoW it seems like they’ve thrown all sense of coherent story/world to the wind and just go “what would be cool to play” and its basically turned into a joke/caricature of itself.
I wasn’t talking about Lord of the Rings Online as much as the story in general, but the game is a good example of how to have a good “world” and still basically be a theme park.
I don’t think you are necessarily “limiting” your mechanics by doing lore first. I think asking yourself “What is this dragon, why is it here, what are its goals” then after determining that, decide the game mechanics to make that happen in game. If you decide you just want “a dragon to be raided” or something, and then come up with some cheap story to justify it you end up with a very artificial character/story/part of the world.
Then again, my “ideal” MMO is much less a game and much more a virtual world. I’m the type that doesn’t even necessarily care for combat all that much as long as there is plenty to explore and see. I think developers should be thinking “What world do we want our players to live in” Nor “what game do we want to create.” Maybe this is not the best way to approach game development, but it sounds good to me at least.
Actually, I do not play Lord of the
Rings online precisely because of the
lore.
When I saw hobbits running all
over Middle Earth adventuring (a very
un-hobbit like activity) and sometimes
wearing shoes,(an even more un-hobbit
-like activity)
and a screen shot of a small party of
nameless adventurers beating down a
Nazgul like it was a WoW Kobold, I said
“this game is not for me.”
Some folks
might call that “niggling” but I see
all of these details as the notes
that breath life into Tolkien’s world.
Games, gamers, and game makers do not
like to be held to that level of
commitment to the lore. The moment you
say “well, they had to let people run
about adventuring as hobbits because…”
or “well, everyone wants to fight a Nazgul”
Well, then you have admitted that game
mechanics come first (players want to
play hobbits or fight Nazgul) and lore
comes second.
That rule is not hard and fast, but
there is always a point where a game
will chose mechanics over lore.
That’s why, for me, the games that have
lore invented for the mechanics, as
Syncaine suggests, work better from a
narrative standpoint.
Of course, once you establish that lore
you cannot change it with a mechanics
update without risking the loss of
credibility. Warcraft lore is a
perfect example of heaps of lore made
to create a game, but then thrown out
the window and retconned (badly) with
each mechanical change.
once you
I wrote a short reply to this on my blog, dunno if you’ve got trackbacks enabled though.
http://theelitist.net/2009/12/29/log/