Gamasutra is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Gamasutra: The Art & Business of Making Gamesspacer
Game Design Essentials: 20 Open World Games
View All     RSS
June 24, 2021
arrowPress Releases
June 24, 2021
Games Press
View All     RSS

If you enjoy reading this site, you might also want to check out these UBM Tech sites:


Game Design Essentials: 20 Open World Games

September 26, 2007 Article Start Previous Page 8 of 11 Next

13. Dragon Quest III, a.k.a. Dragon Warrior III

The earlier Dragon Quest games usually had a section where players were left to their own ends. This one had the most substantial example.

Developed by Enix

Designed by Yuuji Horii

Platform: Famicom, NES

Length: Very long

Of Note:

Until this game, and unlike the series inspiration Wizardry, the Dragon Quest games gave the player pre-made characters (DQII), or at least those with randomized stats (DQ). This was the first game to use a Wizardry-like system of letting the player actually create his teammates in a simple creation process, giving them names and classes and even letting him swap them in and out of his party at will. The result was a game with greater longevity than many Japanese console RPGs. Players can play with a party with the basic Soldier, Pilgrim and Wizard, or experiment with other unusual configurations like three Merchants or Jesters.

The Game:

Up until Dragon Quest III, the venerable series did have some non-linear aspects. Most of the original game is not really blocked in any way other than by being infested with monsters that can instantly kill an unprepared player. The second and third games have a substantial portion of them taken up in a search for important items scattered throughout their worlds.

One interesting thing that separates these Dragon Quest games from many others is the role bridges play in them. Bridges tend to signify a major change in the difficulty of monsters the player encounters. If the player is still struggling with the monsters he's fighting and sees a bridge, he'll know not to traipse across it. There are exceptions to this rule, and sometimes monsters improve upon crossing invisible boundaries, but the use of some element of the scenery to indicate to the player that difficulty is going up is an idea that far more RPGs could stand to borrow.

While most games use "hard" barriers to prevent player access to later areas, Dragon Quest is more likely to use "soft" barriers, where the player can go where he wants, maybe after an early section that's more linear, but it might be unwise to do so because of the strength of the monsters. In an action game player skill can make up for undue numeric difficulty, but in a turn-based combat RPG the numbers and flags that define a monster's abilities are the challenge, and there's only so much the player can do to get around them.

If you think about it, this poses unique problems when designing a non-linear RPG. To a player with 40 maximum hit points, going from an area where the monsters do 12 hit points of damage to an area where they do 20 is an almost unbearable increase in danger. If the quest path is mostly straight and one-way, then the monster spawn zones can be placed to match.

But if the game isn't linear, then the designer has no idea where the player will be going next. The player might stray into areas intended for far hardier characters than he's had the chance to build, and could get wiped out quickly, causing frustration. I think some frustration isn't necessarily bad in a game, but no one wants to challenge the gods unprepared. But on the other hand, if all the areas have monsters that are a good challenge for characters from the moment they are able to enter them, then growth will be rapid at first and soon the player will cruse through the rest of the game.

The Dragon Quest games, to my eyes, solve the problem by taking advantage of an interesting aspect of RPG character growth, the fact that the effective rate of character growth actually decelerates over time.

Take 1st and 2nd edition Advanced Dungeons & Dragons level growth as an example. Back in those days, a character's starting hit points were determined by the role of a single die. Upon reaching each additional level, players got to roll another die and add that to their total. The result was, at first level the player's hit points could be as low as one, but with each additional level gained the law of averages caused HP to trend towards average, making them more generally survivable against foes of equal level. However, the amount of hit points gained with each level remains constant, even though the player is accumulating more and more of them. A first level mage in 2nd Edition AD&D had an average of 2.5 hit points. Gaining an additional level approximately doubles that figure. But a tenth level mage has about 25 hit points, and gaining level 11 still adds on just 2.5 more. The gain relative to what he already has becomes less and less as the character grows.

More recent versions of D&D even out the randomness a bit more by giving characters maximum hit points at first level, and JRPGs like Dragon Quest work around this by giving characters some extra HP at the start, by subtly increasing growth rate at higher levels, and sometimes by even starting players at higher levels, but the overall principle still applies. As the player's party gains levels, they may grow linearly, or even in a steadily accelerating rate, but the effective benefit from those levels decreases.

Dragon Quest games never start the player, from first level, in a wholly non-linear part of the game; that bit only kicks in once he's gotten some levels under his belt. The result is, that large area with monsters all of about the same level, if made just a little harder than usual, can suffice to provide good challenge for quite some time. A level 21 player may struggle a bit when facing level 26 monsters, but he won't be wiped out as a level 1 player would be fighting level 6 foes.

Design Lessons:

The non-linear sections of the early Dragon Quest games (IV does not apply!) allow the player some freedom in his exploration, but have to be carefully handled to continue to give the player a sense of danger. But it isn't all that bad if the player's party gets a little strong compared to the opposition during the tail end of these sections. If an RPG couples monster difficulty too closely to PC advancement, then the player never gets the feeling that his wards are really improving. The worst games in this regard are those that multiply by scaling values to directly tie monster difficulty to player level. Final Fantasy Tactics and Final Fantasy VIII are two notable games that do this. Oblivion does it too. If monsters are always slightly stronger than the player, then what on Alefgard/Ivalice/Tamriel are experience levels for?




14. Ufouria, a.k.a. Hebereke

It's like Metroid meets Hello Kitty, true, but it's actually rather a cool little game.

Developed by Sunsoft.

Directed by "Chow-Mu-Sow" and Yoshinori Homma

Designed by "Ucchi"

Platform: Famicom, NES (European, Australian)

Length: Short

Of Note:

This is a rather quirky open world game that, despite being the subject of an infamous Nintendo Power spread, never made it to the United States. The reasons are possibly related to Sunsoft's continued disappointment over Blaster Master. It's actually fairly rare in any territory, and is regarded as a collector's item.

By the way, if you decide to hunt this game up and play it, you should probably be made aware of the primary means of attack. Early in the game, before weapon items have been found, the only way to harm enemies is to jump on them with the down direction held on the control pad. If you jump on them Mario style you'll just get hurt, but holding down causes the characters to present their feet in order to give the monster a good stomp. Once killed, the enemy usually leaves behind a weird blue face that can be used as a weapon against other monsters.

The Game:

The game's map is found early, but is more like the one from Goonies II than Super Metroid, showing the general location of the areas and sometimes it shows the location of objects, but not exits or any detail, and other things only if special detectors are found.

This is another game that starts the player out with only very basic maneuvers and makes him find 90% of the abilities he can eventually have. In this case, some of those abilities come in the form of alternate characters he can control. Unlike Legacy of the Wizard (see below), the player can switch between these guys at any time, which is kind of a weird decision from a design standpoint. Only one character walk on ice and float on water, yet another can sink in water, one can climb walls (once the appropriate item is found), one can destroy bombable walls (again, with the right item)... and another has floaty jumps that let him cover long distances.

There is really no good reason all of these abilities can't be given to one character, especially since they otherwise have similar abilities and share the same damage meter. But on the other hand, a lot of the game's charm comes from its animations, and there'd be a lot less of that to see if the player only had one guy to play.

As far as game structure, there are some rather fiendish secret routes in this one, like doors found on platforms surrounded on all sides by air, and which must be sort of lucked onto from above. There's another place where the ghost character has to jump from cloud to cloud to make it to an important area, but there's a gap between two clouds that looks too far to jump. It is too far, but if the player tries it, a cloud shoots up from below, just as the player passes it.

While the game's character graphics show an undeniable Sanrio influence, the backgrounds seem a lot more detailed than they have to be. Have a look at the water for example, those animated waves look a lot more realistic than they have any right being in a game like this.

Design Lessons:

One thing about Ufouria that could possibly be regarded as a negative in how it takes after Metroid is its health system. Both games start the player with a pitiful amount of health, far less than maximum, and when starting a new life they go back to that minimal energy level. In Ufouria the player can eventually get up to 250 health, but every time a life begins, he's back to 10. At least in Metroid enemies drop health rather frequently! Maybe one in eight kills in Ufouria drop a health ball, and those are only worth 2 energy each!

The only realistic way to refill health is to find Medicine or Water of Life somewhere in the game, which refill 50 and all health respectively, but each can only be found once per life, and neither can the player carry more than one of either at a time.

While I think we can all agree that this is going too far, there is something to be said for the spirit it adds to a game. It means that the player can't ever just bludgeon his way through any areas. Any time he's going somewhere to fight a boss, even if he's going through the areas at the very start of the game, he can't just rush through and take the hits. Whether the damage taken at any point leading up to a boss, or facing the boss himself, it counts the same. The player must pay attention if he's going to survive, and classic gamers tend to appreciate that.




Speed Demos Archive (I'm as surprised as you are)

Article Start Previous Page 8 of 11 Next

Related Jobs

Gameloft Australia Pty Ltd
Gameloft Australia Pty Ltd — Brisbane, Queensland, Australia

Lead Game Designer
Yacht Club Games
Yacht Club Games — Los Angeles, California, United States

Mid-Senior Game Designer
Hinterland Studio Inc.
Hinterland Studio Inc. — Vancouver/Victoria, British Columbia, Canada

Systems Design Lead (Co-Op, Online, New IP)
Moon Studios
Moon Studios — Vienna, Remote, Remote

Senior Designer

Loading Comments

loader image