This topic is mainly addressed to the developers, but any of your opinions are just as welcome as long as we can build a conversation on what are the main ingredients for a great game world.

To the developers: Even if your opinions or ideas don't reflect what we will see in Skyrim, I would like it if you posted them anyway as a personal idea on game design.


These are the questions.

1. What's your opinion on open world games? How important is a virtual world for you?

2. Which game do you think is the best example of having a believable world?

3. On what idea/element (AI, story exposition, dungeon crawling, etc.) should a open world game focus on?

4. What would be your ideal/dream game of this type?


1. Very important. Not used enough or not used in the right way since a lot of games are happy to just slap a open world on a linear game like GTA clone.

2. I find that the games with the most believable worlds are strategy games. Namely Victoria: Empire under a Sun and Shogun: Total War. The details of everyday life are abstracted to fit the viewpoint of a general and it's all governed by gameplay rather than just storyline as is the case with RPGs. I think Darklands had the best atmosphere in the RPG crowd.

3. A balance between random generated content and hand placed environments. It's hard to do and not a lot of developers are experimenting with this subject anymore.

4. Right now I'm thinking of the old game that Morrowind was supposed to be. The one that was supposed to be 20.000 square miles of generated content plus the hand placed unique areas that we received in the final game. That some new gameplay features.



These are the elements I think are related to designing a believable game world:

-Scale

*Quantaty. Pretty straight forward. The size of the virtual world, not just landmass, but dungeons, quests and NPCs too. The bigger it is the more stuff you can do in it.

*Organization. Remember when you had to furnish the new flat you bought a long time ago (maybe during college or when you got married)? http://www.nassaucountyny.gov/agencies/filmcom/locations/HHM1/Malcolm%20house%20empty%20room.jpg looked discriminatingly small, but once you http://www.mulberry-inn.com/images/gallery/D/room_2.jpg, it suddenly seemed a lot bigger, despite the fact that you had less space to move about. The brain judges space simply a way of pathfinding. Once you put an obstacle that you have to go around, the space will seem and feel bigger.

Adding natural obstacles like mountain ranges or deadly critters that make it hard to advance at low levels, these things make a huge difference. Consider the path that a player will take in the environment you've made and design it to more interesting then a straight road with the occasional fork. A balance in design is needed, because too many detours and obstacles will end up feeling like restrictions rather then challenges. It's not invisible walls you're looking for, the player needs to have free access no matter what level he is, but that doesn't mean it should be easy.

- Depth

The main point of depth doesn't necessarily consist of spamming tiny details here and there. The main point of depth is to create a believable world to play in. Using a lot of details can create such a world, but it's not the only way to do it. Presenting an solid concept through a subtle or abstract way can leave enough room for the player to fill in the gaps with his own imagination. Diablo isn't a very detailed game, but its Gothic atmosphere is solid and simple enough to be immersive.

*Descriptive detail - Books are nice. Lore being presented through dialog and gameplay is a lot better in my opinion. Daedric alphabet or cultural insignias are always a nice touch. After a while (and these games take more than "a while" to play), people start noticing patterns and appreciate them.

Detailed NPCs are very important and there are two ways to treat this. The Fallout/BG way, where only the important NPCs have things to say, while the majority of people aren't interactive and only serve as a background. Bloodlines managed to be the most immersive game of this category. The other way is the Ultima way, in which every NPC has back story or at least a few lines to set them a part, but as a result there is less depth of dialog with the main characters.

*Visual detail - Morrowind is the best example to look at. Every house had its own broom, plates and cutlery, beds or bedrolls, books, pottery. It doesn't seem like much untill you play a game that doesn't give interiors such detail and you're suddenly struck by the uncanniness of the whole thing. It's like a theater play that expects you to believe that an empty table is enough to represent a market or a court house.

More than that, the presence of silt striders and docks, along with all the simple shops like pawnbrokers or clothiers provided a sense of a world that is being lived (even if it never happened). The stables and churches of Oblivion, the banks and court houses of Daggerfall. All these things add up to a subconscious approval of detail, the imitation of a world that's built for more reasons than just dungeon crawling and adventuring. It's all about keeping the fourth wall intact, which is why city streets need to have more then five people walking about.

- Gameplay

First of all: What kind of game is Skyrim? What's its blueprint. Arena was pretty much a dungeon crawler. Daggerfall followed the same design, but while also trying to simulate a giant interactive world. Morrowind was more of a hiking game than a dungeon crawler. Oblivion, while it had a lot of similarities with Morrowind, seemed to return back to its dungeon crawler roots.

The problem with Oblivion is that although it had Morrowind's blueprint (a hand placed world with goodies to explore) its gameplay didn't address this as much as it should. While compass pointing at your goal, fast travel and scaled loot would've worked with a game like Daggerfall, in Oblivion these things took away the enjoyment of exploration. Most of my time in Morrowind was spent pathfinding, following directions and locating objectives. Okay, it wasn't stellar gameplay, but it worked well with what the game had to offer and it was an original idea.

There's lots of room to improve on. Fast Travel isn't a bad thing, but like people pointed out, it needs to be more restrictive. Realms of Arkania series had numerous survival elements to its travel system, New Vegas made the same attempt at adding dehydration and other things. Go further, add more missions based on travel and make the traveling itself more than just backtracking between objectives (The Witcher is a bad example of this).

- Interaction

The game world needs to be flexible. Not too much, but enough for the players impact to be seen. What's my status with this class or group? How does my reputation affect the game? Will people treat me the same after this event? It's so easy to try and represent this through a scale of some sort or through a fame/infamy slider. But other then affecting prices, these scales rarely seem important in the game. You can either detail most of the NPCs to react to your actions or you can make the reputation scale much more punishing. Not being able to do most of the quests, or not being able to buy a bed in a hostel, or talk to most people or even being kicked out of town by the guards. If it's more punishing then it makes the interaction much more solid and noticeable.


- Storytelling

If all else fails then a good story can go a long way in building up a believable world. Games like Mafia or Gothic focus more on the story then free roaming. The Dark Brotherhood questline in Oblivion was enjoyable. And one of the best things about New Vegas was its progression.


*Make Lore more important in the game - Books are nice, but it would be better if lore was more involved in the gameplay (Camoran's book in Oblivion and the books that mentioned the master trainers in Morrowind were a nice touch). http://en.wikipedia.org/wiki/King_of_Dragon_Pass is perhaps the best example of lore being used in gameplay. Memorizing myths, or laws or the customs of different clans or the weaknesses of invading enemies mattered a lot, but you could still finish the game without knowing most of it. That's the secret. Being able to finish the game without knowledge, but also gaining a big advantage if you are aware of the lore.

*Conciseness in writing - Don't spend three sentences to tell me you're a guard and don't trust me and need to be convinced to let into town. One sentence will do just fine. I'm not suggesting that there be less dialogue, just make sure what dialogue you do have conveys useful information in an efficient matter or allows for some roleplaying. Doesn't state the obvious just to achieve dramatic effect. I hate when games feel the need to tell you that the station is going to blow or that the town is under attack when you can tell just by things going on in the game.

*Relateable Characters - Don't go for cliches like the main character a sibling or a spouse. Characters that start out as strangers need to grow on us the more we play. Even if there are just a couple of characters that will grow on us, that's still enough to make the game memorable.

*Don't force the player - Even if there aren't a lot of real choices to make in the game, at least provide an illusion of choice. Never put words in the player's mouth."It was a ruse so I decided to attack them" is a bad way to progress the story. Make sure we know when we're getting into a situation we don't like and are at least given the chance to refuse. Chances are we will still proceed through the story, but if it feels like hand holding it's not fun.

*Avoid cliches - Being a save-the-world scenario doesn't automatically condemn it to blandness. But there are still smaller places where cliches need to be avoided. We all know what a dasmel in distress usually says, or a mayor of a town, or the old instructor who is your second father. Make certain we are surprised. Planescape: Torment used this where a bloodied girl will ask you to save her sister whose being attacked in an ally way. Do so and you'll be ambushed by the thieves whose trap you fell into.

Well how about a real functional economy?

Farmers grow crops, collect apples, slaughter cows, chop wood etc. Raw materials are taken to the cities where they are refined, sold at the markets. NPCs earn money for their work buy cloths, food, weapons, houses .. depending on their income.

If you rob a person a few times he might lose his house, can′t afford to buy expensive food, having servants etc ... If you kill / rob the apple collectors ... "Why are this goddam apples so expensive?" endless possibilities.

Bethesda needs to bring back the old RAI and tweak it if you want this to happen.

Hey, I'm all for features like this, but I always get the feeling that a functional economy in a RPG sounds a lot better on paper than if applied. The main reason is that a part from fooling around, there's not a lot of gameplay value in a linear game.

Strategy games make us of functional economies. A Thief game with an open world would make use of a functional economy. But unless you design your quests around manipulating the economy, in a TES there's little use for this such a thing.

.. Of course if they added a fully function economy they would make quests that involve it... Lol

RPGs need to have economies that are reasonable. Morrowind was okay, for example. Oblivion is pretty much completely broken as far as economics are concerned. You cannot have a world where apples cost 2 currency, NPCs are running business, NPCs own houses, and NPCs claim "I don't make 5 currency in a year!" If such a claim were true in Oblivion, all NPCs would be dead because they could not afford to live.

Yeah, what was up with that quest? 5 septims a year? :facepalm:

And they're living in houses that would cost 10,000 septims if the player would buy it.

I dont think so. Politics, factions, law enforcement, conflicts of all kind, criminal activities and the whole social life cyrcle around economic factors. Playing a role in an RPG means playing a role thru interactions with given factions in such a world. Without economy conflicts have to be built up artificially and also their meaning or their influence on the society around. The world is sterile and there are minor dependencies between player activities and everything else around.

Imagine a military conflict between two powers. One of them lacks military power to defend itself in the long run caused by bandit activities. There is starvation depending on the bandits success and this leads to an advantage for the other power. In such a situation the player has to learn what happens around him and his actions have an influence on the people. Later on he gains influence on the military balance between the two powers engaged. Economy can be a factor that helps illustrate what happens without scripting and designing every detail once a working economy is invented. This could also change the way the player gets in touch with his quests. He can see what happens around and decide how to interact. He can be a hero without becoming a commanded servant. Dialogues and AI would be easier to develop. Themes would automatically come up and give more feedback to the quest's state.

A well thought out discussion, and I agree with pretty much everything you've said.

With regards to quest marking and tracking, I wouldn't be averse to something similar to what Assassin's Creed 2 did when you're sent on a task: Your mini map will point you in the general direction of your destination (if you've never been there before) and you will have to search the area for the specific thing you're after. Eg, you're sent to slaughter some goblins, but you don't know exactly where the entrance to their lair is. The farmer who gave you the quest was a bit vague, so your compass will point you to the general area (marked by, perhaps, a shaded circle like AC2) and you will have to look for it's specific location. I think that's a reasonable compromise, as it doesn't get you hopelessly lost but also conveys the feeling that your character knows roughly where it is, but not exactly. Kind of like any normal human being when given directions!

Actually simulating the economy is probably a no go, but just trying to put the economy in for appearances would be great for immersion. You should pretty much be able to have some idea of where any crafted item in the game came from (even if the answer is "imported from another province"), and NPCs should actually work in their jobs rather than just hanging around in town all day. More thought should be put into the plausibility of living areas (where do they get water, where do they get food, how does anyone survive if 75% of the people just hang around and train?)