It's definitely and infinitely easy to add some more 'land' tiles into your world and say: "Wow, now it takes 15 minutes to walk from one end to another." BUT! Why do I want to walk through a large area consisting of randomly placed look-alike trees, a few rocks, some plants and a cemetery? How many times have YOU thought, "Great, I get to explore the flora" instead of, "OK, I gotta wander through these plains and woods to get to the next town, where the action is. I'm kinda weak still so I'm gonna do a running version of 'Piper of Hamelin' to evade the monsters." Look at Sacred, that's a beautiful game. A lot of detail and you can zoom right in on it or look at it from afar. BUT! It has a huge world and everything not within a city is boring. It's some kind of a necessity for new games (esp. sequels) to have a world y times bigger than before. It's one thing to figure out how you're gonna store this large world but then you need to fill it with something too, imagine that. If your game calls for a big world then it's not enough that you just make a big piece of land, you have to have content. Morrowind was big but most of it was just cliffs, trees and rocks on an otherwise FLAT, BARREN GROUND. Yeah wow, you textured it with a grassy image and placed a few rocks there by the tree, that's great detail. In nature everything grows wherever for a reason and usually you can see hints of what that reason might be. Here all you can say is, there's some land and so trees grow here, yes. So our computers get faster and we can show more polygons, but the developer still has to come up with ideas to populate the world with. Forget about making a large world if all you're gonna do is spray it with trees and move on to design towns, "'cause that's where the action is."

I agree. Most important part of games for most of us, is the gameplay. When you have to do 90% of repetative or boring things, it ruins the 10% that is fun. I have found that traveling through game worlds are the most boring things to do, especially when you have to go back and forth between the same places.
Although, in my opinion, it does make sense. Too much of everything is bad. So one should not say "Stop making large worlds", but more like "Balance out the gameplay!"

IMHO.

Hmm... This is a little bit tricky. I think that wilderness can be part of the game. Look at the Deer Hunter-esque games. There's some boring, repetitive land, with little to differentiate one area from the next, but it's presented as a sort of arena in which you compete with your pray. This tree is like that tree is like the tree six miles away, but there are some scratch marks on it that tell me I'm on the right track.

So, huge, boring terrain would be conducive to gameplay in which anomalies in terrain are important. For the perfect example of empty terrain, look at a flight simulator. Ten thousand square miles of sky, and only the odd cloud or airplane to break it up. Yet, such games can be thrilling. The air isn't just a void between airfields and targets, it's the essence of the world. You fight, and scout, and otherwise act in that area. Sure, you spend a lot of time cruising from place to place, but you're watching the radar, checking your charts, and using landmarks to navigate. It's terrific.

So I agree that excess space is a waste of everyone's time, and that game content can be lost in a sea of mapping. But there are simple aspects of gameplay that can swell to fill any space. Imagine if you could be a cartographer in an MMO. Would you ever think there's too much map to find?

Don't overcompensate, and leave yourself with a small map that players backtrack through again and again. Seeing the same tree forty times is just as bad as forty identical trees.

This is stuff people already know and try to design around. Generally in modern games if you must travel through an area, then there is a reason for it and they change up the art from area to area enough to make it all consistent to the area, but new and not boring.

A good game study of how to be interesting while people travel through your world is Final Fantasy X-2. While the game obviously had travel areas (reach the top of the mountain, etc.), I never felt bored with the areas or felt that the fights came too often, lasted too long, or that the peices of the story were too far apart.

This is an easy obstacle to design around and I see this in modern game design all the time.

Fun NOW, not fun later.

Personally, I'm not opposed to large environments, especially in RPG's, action, adventure, or racing games. However, I will agree with some of the aforementioned comments that a designer should not create an environment larger than can be supported by the content available for the game. Additionally, I believe that a talented and creative level designer can instill interest in the environment no matter where the player is at if they are aware of a couple factors. These factors are largely determined if the gameplay is catered to "immersion maps" or "points of interests maps," which I'll cover shortly.

I also tend to believe that there is a "sweet spot" concerning placement of iterative elements in level design. On one hand, players won't enjoy passing 100 elements that are all the same again and again, but on the flipside a player also will lose interest in a unique design element if they have to pass it 100 times as well. I know this was pretty much already said, but it's worth reiterating.

The so-called "sweet spot" will be largely determined by the gametype. In the Metroid Prime games for example each room and environment element is fairly unique and there is very little use of repetitive elements. This is largely because the game is exploration-based and the player is expected to revisit the same areas more than onece; multiple times if necessary. Games that are less focussed on exploration and more dedicated to action and player interaction however, such as the Battlefield games have less of a demand for non-iterative elements as the maps require a large scale to accomodate planes, helicopters, tanks, jeeps, etc. This is not to say that these types of games wouldn't benefit from more detailed unique elements to populate the environment, but the cost-benefit ratio would be significantly less as there are many areas of the map that may never be explored by the player. Instead the player is directed to areas of the map that demand their attention. These types of maps are what I prefer to call "Points of Interest" maps. The previously mentioned map type exemplified by the Metroid Prime games are what I like to consider "Immersion maps". Either one of these level design approaches can be engaging for the player as long as they are paired with the correct type of gameplay.

Obviously, not all levels will fall completely into one category. Very few things in game design are ever black and white. In conclusion, there should be no reason to exclude large scale environments in the process of level design as long as they are created with gameplay in mind and don't exceed the limitations of the content available and the ability to create an engaging experience.

I must point out that I'm fed up of small worlds. I like exploring, and looking for interesting things. Obviously there needs to be these interesting things hidden among the terrain, but I don't agree that the same number of things in a smaller area is better. The feeling of exploring a vast territory adds a lot to the feel of a game for me.

Sorry for double posting.

I just wanted to say that I agree with infrmtn that it is bad design to make game worlds larger just because you can or simply for the sake of extending gameplay length. I'm currently working on a level in UnrealEd for a personal project that is less than 1 square mile in size, but because of its layout and in-game events the level will provide 40-60 minutes of gameplay when completed. The other interesting property of the level is that no area will have to be visited more than twice unless the player decides to backtrack. I've also made sure that there is no more than twenty seconds of travel time without encountering something to make the gameplay interesting, new, or challenging. There are repetitive scenarios in some cases, but they occur in specifically designed situations. The level is what I would consider an "immersion level" but with "points of interests" elements to drive the player forward. I might post it for download when I'm completed, since I believe its a good example of making a level seem large in scale when it's in fact not very large at all. Unfortunately, I still think I have about 2 months of work on it so it might be a while before I post it.

Have a look at Zelda: Wind Waker on how to do this right. The world is huge but filled with treasures, games, hiden items and all sorts of things to do. It's a blast to explore and if you really don't want to you eventually gain the ability to instangly go to a point on the map.