Tobold's Blog
Tuesday, February 14, 2012
Designing an impossible game

Although I just renamed the blog to remove "MMORPG" from the title, I will still talk about MMORPGs whenever the mood strikes me. There was some interesting discussion in yesterday's thread about how the people who hate all current MMORPGs are all clamoring for the ultimate sandbox game which totally revolutionizes the genre and is totally different from all successful current game design. My proposal that successful game design is more likely to come from evolutionary changes was dismissed as lack of ideas. So in this post I would like to list my revolutionary ideas, although I am convinced that it is impossible to turn them into a successful game at the current time. At best MMORPGs could slowly evolve in that direction.

The main design problem of MMORPGs in my opinion is repetition. I understand the cost reasons behind it, but I also see the negative effect on gameplay. For comparison, look at a pen & paper roleplaying game: Every single encounter is unique and new to the players. Thus a series of unique encounters makes for an exciting adventure. In a MMORPG there is much less of that uniqueness: Most monsters react in extremely similar ways, quests have similar structures, and dungeons resemble each other in functionality as well. Furthermore if you "fail" at some content, you have the opportunity to try again, and again, until you succeed.

This repeating structure has as consequence that MMORPGs are not games of discovery, adventure, and exploration. Instead they become games of execution. You learn what happens when you pull a mob, what buttons to press, what special abilities the boss mobs have, and so on. Your success is not based on being able to manage a unknown situation, like it would be in a pen & paper game, but on how well you know what will happen, know what the best predefined solution is, and know how to execute it.

My impossible MMORPG would be very different: Monster spawns would not be static, thus just because there are wolves in this part of the forest today, it doesn't mean they will still be there tomorrow. Monster locations would in part change randomly, and in part in response to player actions. And yes, I know that Ultima Online tried that and failed, but they obviously just underestimated the speed with which players can kill mobs. A better system would increase respawn rates in response to player overkill, and make more dangerous monsters appear if the players empty a zone of monsters.

Secondly in my impossible MMORPG the monsters would be far more different from each other. Players would not be able to rely on "knowing" their "aggro radius", because not all mobs would even use that game mechanic. There should be mobs which attack if they "see" you, not just stare right through you if you stand one step outside the aggro radius. Vision-based aggro would also mean that it matters from which direction you come. Classic aggro radii could be smell- or sound-based. And there should be some random chance of unusual behavior, a mob running away at first sight, or running for help, and not a predictable attack that is always the same.

Mobs should have a wider range of attacks, and there should be some randomness in those as well, especially for boss mobs. It should not be possible to "learn" how to beat a boss with predefined moves. Instead boss abilities should be animated to give players an idea what might be coming, and players would be forced to react to what the mob does.

My impossible MMORPG would still have quests, but the tasks given to players would not involve specific monsters. Instead they would for example be asked to retrieve an item from the end of a cave, without knowing what lives in this cave. And the inhabitants of the cave could change from one day to the next, so there would be no strategy to look up in some database.

Besides a far more dynamic world with changing monsters, dynamic spawns, and full of the unknown and unknowable, my impossible MMORPG would also have a more dynamic system of powers for the players. Instead of always having the same buttons to press on, every available power would disappear once used, and replaced by a random other power. But that new power would not be totally random, but just randomly selected from a set which is predetermined by the players. Thus just like deck building in a tradable card game, players would have to select a set of powers they believe will be able to overcome the challenge they face. And that includes the possibility of finding yourself in an unexpected situation with the wrong set of powers, and having to retreat and change your set of powers to better fit the challenge.

My impossible MMORPG would not have levels. Instead it would start with a very low degree of complexity in the powers on offer, simple attacks that work moderately well against everything. Over time players would learn more specialized powers, which are stronger against certain kinds of enemies or in certain situations, but weaker in others.

Overall this would result in a virtual world full of dangers and adventure, and constantly changing. Certain points would be permanently safe, others permanently dangerous, and large areas would be dynamic and changing in response to the actions of the players. And it would be in these dynamic areas that players could build houses or guild structures. Leave such a structure unattended too long while no other players are keeping the area pacified, and you'll find your house occupied by orcs or worse. Clever algorithms would match mob activity with player activity on the server, so that less populated servers would not be unable to keep the monsters at bay, while overcrowded server would not run out of mobs to kill.

I don't think we even have the technology to make all this possible, and the money even less. But at one point in the future I would like to see a virtual fantasy world which is dynamic, and uncertain, full of adventure instead full of checklists of things to do to "finish" a zone.
Quite a few of these ideas are things that have been used already but which were toned down or removed because players complained about them and/or avoided games/zones/mobs that used them.

Older MMOs often had (and still have) mobs that attacked on line-of-sight, for instance. Everquest had lots of them and so did DAOC. Even EQ2 has them. Try walking across the hills of Enchanted Lands towards Runneyeye and see how far you get before an Evil Eye spots you from several hundred yards away!

Older MMOs also had numerous hidden mechanics relating to agro. If there were a dozen mobs in a dungeon room you could not predict which might or not attack you without knowing each of their individual agro ranges (which would vary), which were sociable with which, which you had faction with and how good the faction, and even which might have a particular dislike for your race or class. Some mobs even had a random element to their aggressiveness and could be passive sometimes and aggressive at others, switching from one to the other without warning (elephants in Everquest used to do this in the very early days).

What happened over time was that players voted with their feet against unpredictability and randomness. When it was all that was on offer they complained about it almost non-stop but continued to pay their subscriptions because it was that or nothing. As soon as MMOs began to offer more predictable environments with more consistent outcomes that's where players chose to go.

Most of the older games then revamped in the same direction to compete. Even the older MMOs that are still around now feature much less random, unpredictable environments than they once did and offer players a much higher chance of a successful outcome.

Complexity, such as it is, ceased to be found in the underlying interactions of environmental factors, with the inherent unpredictability and surprises that created but instead became vested in increasingly baroque scripts, which could, with much time and effort, be learned and performed reliably and repeatedly.

I doubt there's any technical reason why your "impossible" game couldn't be made right now. It could probably have been made a decade ago. I think the issue is that it couldn't be made profitably because there isn't a sufficiently large audience for it.
What happened over time was that players voted with their feet against unpredictability and randomness.

But that same argument would also kill the idea of a sandbox PvP game. The attraction of PvP is that it is unpredictable. And a predictable sandbox game is an oxymoron.
This isn't too far from what I myself posted not too long ago in a comment. I managed to find it and will paste it here again.

I don't know if it's the right way to go, but I'm dreaming of a fully dynamic huge world where what I do will actually make some difference. Maybe just for me (and perhaps a few friends/the guild) but still some difference. The easiest for me to imagine it right now would be Skyrim online as that's the closest I've come so far to such content, but I think even that might be a bit limited in the long run. Seeing merchants travel with their carts along the road and having highway robbers set up ambushes. Orcs patrolling the perimeter of their camp being attacked by a dynamically created group of centaurs that decided that they had to leave camp to hunt for meat. Kingdoms going to war because of a gold mine on the border between them. Battles rolling back and forth. One of the aforementioned kingdoms actually capturing the mine and it's attacked by trolls that want to use it as their lair. In the middle of this the player characters could be employed by factions depending on what their current needs are.

I know that some of this content I've described is announced in GW2, but somehow I doubt that it will be as dynamic as announced and even then it's still a bit too bound by a certain set of rules or time tables. Time will tell though. But that's the type of stuff that I'm dreaming of. It would be hell to program no doubt, but I think that who ever makes this happen in such a large scale would probably have the next WoW on their hands. At least if it isn't a (&%#(@/ niche PvP game and is polished/relatively bug free.

Great ideas, and in my humble opinion possible to develop, with good marketing and of cource quality it could get same success as vanilla wow back in the days.
"But that same argument would also kill the idea of a sandbox PvP game. "

Sandbox PvP games haven't had the breakout success that WoW did. EVE is the most successful, but it doesn't seem to be a model that people are rushing to copy. And when they have, like Perpetuum, it hasn't been a breakout success.
I am not sure whether I like your ideas (i.e. I'm not sure whether I'd like to play such MMO should it be made).

However aren't you contradicting yourself? On the one hand you want players to be able to react to unexpected -- which in my mind includes power sets that are able to handle the unexpected (thus generalistic set) -- and this it something I'm very much "for" -- I much prefer to play true hybrid builds as much as possible.

On the other hand you talk about "complex power that work better against certain monsters or in certain circumstances" which seems to imply "go wipe and learn what powers you need" (and which, of course, assumes that you'll be able to re-create conditions reliably for the powers to work).
Would a game similar to D&D pen and paper be successful? I mean, using the same tropes translated directly to digital. I don't that exists today and would be an interesting deviation from the norm and satisfy the open world seekers

The most similar game that exists today that would fit this would be Little Big Planet (a PS3 game). This game has a built in mission system to get a feel for what you can do, but the real game is in designing levels, including the challenges and rewards found therein. Each level becomes like its own game. And the creative force is amazing. There are a lot of amazing designs out there, and of course you can create your own. For example, I created a level that is somewhat cyclical, giving my toddler (who doesn't altogether know what's going on) a great deal of entertainment in the same LBP universe.

I could see a translation of that into some sort of MMORPG, with a more literal translation of the components of the D&D P&P game. (What's the right short hand for "Dungeons and Dragons pen and player" game?)
I agree with Bhagpuss - none of these ideas are completely new or indeed tremendously difficult to implement, and most of them and probably all of them have been tried.

A game with these ideas sounds good to me too. Except for the 'choose your powers' bit which can make a good game, but that game will not be an MMORPG. Examples of games using a similar are Guild Wars and Diablo III. The 'use skills that appear from a list' idea has been tried too, though I cannot remember the name of the game.

The problem with unpredictability is that it is bad for balance and it is also bad for bad players. Casualisation of MMORPGs is IMO the main reason companies such as Blizzard have been working so hard to expunge unpredictability from their game for the most part. There is a reason that roguelikes are a niche market.

I dream of an MMORPG that plays more like a roguelike, and perhaps some indie game will deliver that some time. But I don't think the mainstream is evolving that way. It's evolving in a direction I don't like. It is the nature of evolution to take its own course.
The basis idea here is that fun in a game comes from a busy mind (somewhere between boredom and frustration).

One very important method to keep a player's mind busy is learning (exploring in the broadest possible sense).

WoWlikes are comparatively(!) easy to learn. That's why what would help WoWlikes are more complex rules which scale with the player's capabilities.

The human brain loves learning about the environment in a empiric way. A good game is just like some parts of real life: you look at something, create a model in your head about how it works. The model works well enough.

A little while later the challenge level increases slightly. You need to have another look at the model in your head and you realize that it wasn't the whole truth. You adjust the model.

This is what brains evolved to do - and that's why we usually have a lot of fun doing this - as long as the challenge level doesn't rise too fast or too slow!

The best way to achieve a good pace of challenge are dynamic challenges, instead of a fixed script which is the same for every player.

If MMOs turn more social (and not less social, like in recent years) they become more interesting even without developers doing anything...

One thing which is happening right now, however, is the opposite. The developers basically gave up on creating well balanced challenges for every player, because they realized how very different players - even of the same age - are.

They now try to keep your mind busy with emotion and execution alone. They thus ignore logical and social activities. However, a game which successfully uses all four ways to keep a mind busy would be tremendously successful.

But, yes, it's not easy to design the rules for such a game. And it certainly won't be done by an indie studio. We need AAA money for this.
I didn't see anything in your list that would seem to be that technically challenging. Instead of the set spawn formulas, add a few variables to add surprise.

WoW already has a LOS check; just use to to trigger aggro.

I think your list is very possible. Even easy.

My impossible MMO:

1) instead of maps that are essentially flat planes with a few polls stuck in a few places, real terrain that you have to use to defeat enemies. Trenches, climbable trees, etc.

2) Player driven content. I would like it if instead of buying items placed in the AH, players could instead make a standing offer to purchase a given item at a certain price. Once bought, the item has to be delivered. Instead of just hearthing back to storm wind someone needs to go travel through dangerous territory to deliver it. If they run into bandits, the item is lost. So then you might put out an ad for a few guys to travel with you to protect you. And another player could hire a few guys to help him rob travelers. That sort of thing. There would be a lot of potential for that system to go haywire, but it would be pretty awesome.

3) Crafting driven itemization instead of loot. I would like crafting to be a complicated thing, where the logistics of it are complicated enough that you have to hire other players to help. To craft an item you might need logs from the forest of sorrow, mushrooms from such and such, gold from this and that, and so on. The demand you create with these items generates more player driven content as people aggregate the demand and start their own businesses to meet the demands of others. Since all items come from crafting rather than loot, crafting becomes really important instead of the consumables side show it is in WOW. This helps drive dynamic player driven content generation.

4) multi-layered loyalty. In most MMOs, you have your faction. You couldn't betray this faction if you wanted to. You're just alliance or chaos and that's that. It's lame. I want people to be able to cross over. That might be a really epic quest line, but it should be possible. It should also be possible to sell your faction out (and get away with it if you are smart). Also, I'd like people to have their guild, super guild (basically an official alliance of many guilds) but also a religious, national, and city affiliation, and there is content based around people with the same affiliation doing things together. So if their guild decides to attack the Temple of Whatsahoozits, all the whatsahoozitians in the guild will be upset and suffer consequences, and all their whatsahoozitian friends will like them less. Lately I've been thinking it would be cool to have a family in the game. You have to take care of your family. If you make money you buy them a mansion; if you don't they live in a studio apartment. So they live in a certain place. They bring certain benefits and have lives apart from yours. (Quest: Your Niece got Knocked Up; Go Find the Bastard and Make Him Marry Her) And if the Orcs besiege the place, you are personally motivated to help save them, because if they get killed you suffer serious consequences.

Basically I want a game where your activity matters in a semi-permanent way (if you burn a town, it stays burnt until someone rebuilds it), and the game presents you with conflicting loyalties.

I would also like to see a NPC faction that is run by a dedicated employee of the company. This faction is small but powerful, and it's sole purpose is to keep the drama at high boil to give the players the chance to be heroic.
"A better system would increase respawn rates in response to player overkill, and make more dangerous monsters appear if the players empty a zone of monsters"

UO tried that too and it failed. You will end up with areas full of dangerous monsters no one will enter, leving an angry mob rage quitting the game cause its too difficult, while the players who did it sit in their corner stroking their fat chests full of food.

"My impossible MMORPG would not have levels. Instead it would start with a very low degree of complexity in the powers on offer, simple attacks that work moderately well against everything. Over time players would learn more specialized powers, which are stronger against certain kinds of enemies or in certain situations, but weaker in others."

Players want clearly visible progress. Otherwise they don't feel empowered. Also don't underestimate the casual horde which don't get your game unless you have levels.
I love the idea of NPC's and regions that react to large scale player behavior. In low level areas, it'd be something that encouraged interaction between low level players and assured general zone safety would be pretty assured.

In higher level zones, it'd be something like a tug of war with the environment. The more players win and make their areas safer or just more farmed, the harder the zone would "fight back" with dynamic/harder monster spawns. Essentially, I'd love the idea of an NPC driven EVE. High risk/high reward areas that change based on player participation/success in those areas.
I'd have thought Syncaine would be in saying how darkfall does these things? Certainly it does line of sight aggro.
I had a snarky post about how we already have the random abilities game and it's called Poker... and other observations.

But I'll just leave it as - Tobold your blog has been an experience. I have had a few moments of unique thinking reading your blog and I may pop in from time to time. But I will not read frequently.

I truly hope you find a game that can live up your expectations. I don't think the ideas are too elaborate or too uneconomic. Quite honestly after reading your blog for a long time... I don't know what you are other gamers who frequent here want.

I my modest opinion maybe we all are seeking novelty in one form or another. And further when we see small changes around central game genre some are driven to seek fault and not revel in the familiar.

I have to say for myself, at this moment, my gaming cup runneth over. I feel blessed to play online games in this age where people will risk astronimical sums of money in the mere hope of entertaining me. And they ask but a pittiance of my money with little promise that I will stay around. (Try that kind of deal at the local health club boys!)

In any case:

May the virtual road always rise to meet you.
Early sidenote, @Tobold: Ryzom did the adaptive mob ecologies, where overhunting a predator resulted in that predator's prey overpopulating, or the predators migrating elsewhere. It worked pretty well until the high level players would get together with low-level players on crafting excursions, which was the ecological equivalent of setting off a thermonuclear device.

I read these things, and I see pure awesome. Faction changes? Awesome. Weaving of powers? Awesome. Terrain as part of a combat environment? Awesome. GM-run NPCs? Awesome.

The thing is, you really need to check these kinds of ideas against the human element. Developers call it 'TTP' (Time to Penis: How long does it take for players to take any given feature and use it to draw a penis). You need to take any given awesome feature and hire some 12yr olds to figure out how Internet Bastards will break it or exploit it.

Then you have to try and find some kind of design way of dealing with the exploit or antisocial behaviour. Then make sure THAT can't be exploited. I'd say 90% of our design decisions aren't made around what players find predictable or challenging or whatever, but how long it takes for people to be dicks about it.
There's basically no way that a game like this would be a major success. This would be a niche game, with niche players. However, I wouldn't be surprised to find out that this niche was just big enough to pay the bills.. Maybe even turn a small profit.

I would play this Impossible MMO. For ever and ever most likely.
One reason why any game which does things differently is impossible is because if you propose anything new, everybody will tell you how this is impossible. Just read through the comments here. No wonder we only ever get WoW clones.
I find it amusing how you pointed to D&D when giving a counter-example to repetition. How many hours a week do you play D&D? Ten hours? Twenty hours? Do your players play D&D 80 hours a month, like the average MMO? Do you think you could come up with 7000 hours of non-repetitive D&D content over four years (which is how long I played WoW)?

As some other people are mentioning, and you implied by calling it the "impossible game," the problem with the sandbox ideal is that who actually wants to play it? You don't get to have AAA money without AAA subscriptions, and the evidence overwhelmingly suggests that the average gamer enjoys a scripted experience.
I've been playing pen & paper RPGs for 30 years and have never done the same combat twice.
I haven't played a lot of D&D and didn't know what to expect when I tentatively found a DM to play with. He ran an early edition version of the rules, with minimal dice-rolling, and used a fairly traditional module for us to play with - respond to a village's request for aid in quelling goblin raids, by storming their stronghold underneath some ancient ruins.

Along the way, various tropes were subverted.

Stone statues housing precious gemstones in their design had their limbs and weapons smashed to bits prior to gem-removal, in case they came to life.

Approached by innocent-looking 'fellow adventurers', our mistrustful party prepped weapons and provoked them, discovering that the 'adventurers' were actually bandits. The survivors were stripped naked, searignly branded as bandits, and sent on their way.

Approaching the ruins, our party carefully paced out the size of the buildings from outside, to do aural recon and gauge the dimensions of the building to have mental maps prepared for internal exploration.

Sneaking through the stronghold, the goblin mess hall and food stores were located... and poisoned.

A bag of holding and a bounty on monster bodies was exploited mercilessly, every fresh corpse stuffed in between dimensions to later be hacked apart for both claiming bounties and selling parts to alchemists, making us rich.

cheap lantern oil was used to fashion crude but horribly effective incindiaries.

Surviving goblins and leaders emerging - coughing and choking and sometimes on fire - from the smoke-filled warren of subterranean tunnels were struck down by a barrage of crossbow bolts.

...Such is the freedom afforded by nasty, horrible player imaginations. I want the MMO which can mimic that, but by god, can you imagine the design document?
Um, it all sounds very much like S.T.A.L.K.E.R:

"* Living-and-breathing world: virtual monsters and stalkers migrate around the Zone, similarly to the player they are preoccupied with day-to-day things – feeding, fighting, looking for prey, trading, sleeping, taking rest
* Freedom of movement around the storyline, limited only by the player’s will and abilities
* Atmosphere of eerie mystery and constant danger"

But this is offline shooter, not MMORPG. Only thing about playing with other people, I found in your description, was guild houses. And I think, that playing with other people is what make MMORPG players to log on. If only some game would give people something sensible to do every day.
Honestly a lot of the MMO format you see is the result of the technical limitations of having to build an environment that theoretically can be visited by thousands of people at a time. So to a degree a lot of the things we see as being the pressure to conform to the WoW standard (and that's probably a factor) but it's also the pressure to conform to what the hardware is actually capable of. There's a reason so many MMOs launch with serious bugs and problems.

However, I really don't see the problem with more flexible spawn tables. Someone mentioned that some places become overcrowded with tough monsters and no one goes there; well, add a despawn function where the monsters despawn after an hour or two and get replaced with the normal mobs.

As fair as my ideas having serious problems with players abusing them--- yes, absolutely. I'm sure it would be a nightmare and that some clever guys would spend a lot of time trying to figure out ways to abuse their freedoms. I don't know how you'd stop that, and I'm sure there would be plenty of unintended consequences causing major headaches in my dream MMO.
I think the moment game designers try to make sure that nothing bad ever happens to the players, the game becomes sterile and uninteresting. I would like my real life to be safe, and my virtual life to be an adventure. In these days of economic uncertainty and boring game design you're more likely to have it the other way around.
"I think the moment game designers try to make sure that nothing bad ever happens to the players, the game becomes sterile and uninteresting."

Remember the zombie invasion in WoW in 2008? I loved it, and looking back at your posts you did as well, but we are sadly in the minority. This dynamic event just caused torrents of complaints from players who felt that their game play was being unduly interrupted, and as I predicted at the time to my friends it was pretty much the end of "dangerous" world events in WoW. Players were safe by order of the subscribing majority.

You are getting cross at people for saying your idea wouldn't be a mass market triumph, when that is what the market has clearly shown us! Creating an exciting next generation MMO that also turns a profit is a lot more difficult than I think a lot of us armchair developers realize. I think effort would be better spent promoting the idea that a MMO does not have to be the most popular or the most profitable to be a success.
that is what the market has clearly shown us!

No. Because if that was true, Minecraft would not have been a success. The only thing it shows is that if you start offering players permanent safety, and then take it away from them, they will complain. It doesn't show that a game without permanent safety can't succeed.
I guess your impossible mmorpg is very possible. But it wouldn't work, and here's why.

An interesting experience of downscaling open-ness/randomness to make it possible is coop action series Left4Dead.

In L4D, the levels were pretty linear, but the monster spawns, as well as huge zombie horde waves, were not. Mini-boss encounters were randomized too, and when mixed and matched with other mini-bosses and regular enemies in various environments, they felt pretty fresh and unpredictable.
Supplies were mostly randomly placed too.

The difference between L4D and other coop shooters is that you cannot learn and perfect a fixed routine to beat the level (although you can develop a general strategy considering where it would be wiser to hide and where to run). You cannot just learn what buttons to mash in each given part of the level, you had to understand the game world and adapt to it.

My point is, this was implemented on a now eight year old Source engine, without any huge technical (but with many game-mechanical) innovations. Perfectly possible.

Since this has something in common with your "impossible mmorpg", maybe you'll be interested to know why and when players eventually grew bored from L4D.

You see, while the sheer number of objects' position in the game world is very big, number of different tactical situations is not.
The game seem to invariably degrade to one of the situations (run, defense or sweep) with negligible differences.
You may not know when the world will switch state and you'll be forced into one of those situations, but you pretty much know what you'll be doing anyway.

Even if your monsters would be procedurally generated, the players would eventually learn all the significantly different combinations of monster abilities, and how to play against them.
Of course, replayability was much greater (you could play one level for hours no playthrough would be the same), but by no means replayable enough to preserve the sense of exploration and mystery.

So yeah, I'm afraid sheer randomness, even very cleverly applied, doesn't necessarily lead to tactical variation, let alone sense of adventure and exploration, which are fare more fragile.
I like your ideas. Some of this could be addressed by giving players the option (and incentives) to play "npcs", monsters and bosses. Perhaps there could be a system where a player could control, to a limited extent, creatures and encounters in certain areas and dungeons. Basically, giving players the option of being a DM within the game.
To be fair, if minecraft where a triple A mmo, it would be considered a horrific failure.

There's certainly people who want that sort of game. whether there is enough of them to support an MMO that could be considered big in relation to WOW is another question. I don't think you'd like the answer.
Post a Comment

<< Home
Newer›  ‹Older

  Powered by Blogger   Free Page Rank Tool