Let's return again to the well-rounded design space base that we started with, and let's call this base Unto Caesar. Remember, this base is essentially a collection of all the gameplay elements needed for a great game but no levels created to actuate this content. Instead rending the design space so that the gameplay falls apart; instead of dividing the design space into different slices to create classes or groups of player abilities; and instead of doling out core abilities over time, what if we start with the core and build upon this base? What if we take this well-rounded, set of competent "normal" player abilities and add a few more on top by way of power-ups, upgrades, and new abilities? To rend means to tear down, now we'll consider our options to render or build up the conveyance of a design space through level design.
As designer, we first have to consider what kind of experience we're trying to convey and what kind of meaning is conveyed in the process. What is it about a well-rounded design space that's worth conveying? Or perhaps the better question is, what kinds of messages are conveyed from interacting with a well-rounded gameplay system? To put it simply, the strength of non-linear, non-straightforward gameplay lies in emergent possibilities; how different actions are possible from a given point and how the scenario branches due to these actions. Because there are so many ways to play, there are a lot of different scenarios for the player to consider. Knowing the nature of emergence, games tend to have so many different possibilities that effectively coneying ideas, developing meaning, and teaching the player greatly benefits from a clear and strong filter. Goals work extremely well as filters (read more on goal-setting theory here). Because games have goals or goal-like structures, all of these possibilities and options are measured according to the value scale of victory.
Good games use goals to set player expectations, begin to shape the squeeze, and to steer players closer to better understanding the systems at work. It's important to understand the way a game is designed to teach players works in parallel with the conveyance of ideas. In other words, players can't grasp the complex interactive ideas in your game until they understand the rules at play. So to better teach players the rules at play, one of the most effective ways is to gradually build on a set of essential core concepts. Another method that helps is to tune all the challenges in a way that presents, tests, and reiterates concepts. And finally, removing challenges that convey ideas, rules, or lessons that are not relevant to the core experience or that are too advanced for the player relative to their skill level is a great way to keep the experience focused. Most of the best Nintendo games are designed like this. Super Mario Bros. The Lost Levels was not designed with such careful tuning and most players's experience with the game clearly reflects this.
Teaching and Conveying a Design Space
Designing a game around a well-rounded core of mechanics is a great way to create quality gaming experiences. Well-rounded design communicates to the player, "these mechanics and gameplay elements are core to this game. Learn them. Trust them. And use them to calibrate all of your understanding of this game." Level design that focuses on well-rounded design communicates "don't worry about what you're missing. You have everything you need. Keep your eyes open. Listen to what the challenge asks of you. And breathe." When each mechanic and gameplay element has a clear purpose that's distinct, it is much easier for developers to convey this message (especially through a clean gameplay presentation).
A pro/con list is how most players understand the function of their mechanics and abilities. The simpler the list of pros and cons for a particular mechanic or gameplay element, the easier players can fit it into their understanding of the design space. This is key property of well-rounded design that works hand in hand with the relative efficiency of well-rounded design. With fewer elements in the design space that cover a wide range of gameplay interactions, it's easier for players to create a "mental map" of a game's design space. Though the gameplay elements can feature many more subtle nuances, their basic position in the design space is clear.
Take Super Mario Bros 2D platformers for example. The newer versions of these games feature many different power-ups from Ice Flowers, Fire Flowers, Golden Fire Flower, Raccoon Suits, Blue Koopa Shells, Mini Mushrooms, Mega Mushrooms, etc. For most of the power-ups, Mario can only use one at a time. This kind of design helps keep the pros and cons of each power-up distinct because players can't have it all. You'll miss the offensive power of Fire Mario when playing as Raccoon Mario, and you'll miss the gravity defying flying powers of Raccoon Mario when you're tossing around fire balls. The give and take of the pros and cons is great for gameplay of interesting choices. But this kind of design also makes the game more dynamic because each pro becomes a "con" when switching between power-ups. If you've ever played a modern Mario you know exactly what I'm talking about. You're running around as Fire Mario when you come across a Super Leaf power-up. And for a moment you think to yourself, do I really want to give up my projectile powers for floatiness? Depending on the level, your playstyle, and your skills you have an interesting choice to make.
Because Mario can only have one power-up at a time (excluding the Starman, Gold block hat, and Yoshi power-ups), players can have very different experiences playing through levels based on their power-up choices or mistakes made. But no matter what, even as Small Mario, players have all the core abilities to beat the game. More importantly these core abilities are still very capable, dynamic, engaging, and interesting. This is the difference between designing a game around a well-rounded core and merely doling out abilities over time.
Why you shouldn't... Stack It Up!
Designing power-ups or additional abilities that have clear pros and cons is the best way to compliment a well-rounded core design. On the other hand, allowing the player to stack power-ups and abilities is the best way to undermine a game's design space. Good gameplay or gameplay of interesting choices requires a balance of power so that many tactics and strategies are viable. For any game, if you give a mechanic or element too many pros and too few cons, the balance of the entire game is put at risk. Likewise, stacking power-ups and abilities almost always results in unbalanced gameplay.
Consider that nearly all player mechanics have more pros than cons. Mechanics in games wouldn't be good for much if they were bad at more than they were good at. And certainly all power-ups have more pros than cons otherwise they would fail to help the player undermine challenges and therefore would fail to be power-ups. So, with the design space of these moves biased toward pros, stacking these mechanics and power-ups would result in player abilities that are overpowered. Stacked abilities stress the interplay design of the system by making counters less obvious, less straightforward, and more nuanced. If a player uses a mechanic with 99 pros and 1 con, simply trying to influence that player in ways to expose, target, and exploit that 1 con will be quite an effort. Matchups can get pretty rough in fighting games that slice up their design space via its characters, but stacked powers are worse. A similar balance stressing effect happens when cancels aren't carefully designed into a game.
Not only does stacking make games harder to balance, but it makes games harder to maintain clean feedback. Consider Pokemon for a simple example. In this game Pokemon can perform moves in battle and hold items that increase their speed. In battle, it can be difficult to impossible to tell how fast an opposing Pokemon is. Since there are a lot of factors that affect the speed stat and this stat is mostly used to determine which Pokemon acts first, there may not be enough feedback in the game for players to determine why the result is the way it is (see video here). Was it the EVs, IVs, the moves it used, the movies I used, or the hold items that made the difference? Or was it merely a speed tie? You may never know.
Stacked abilities also smooth out wrinkles in a design space. Remember, wrinkles in a design space are created when properties are not applied to all elements within a set in a game's design space. It's unique that the Plasma Pistol in Halo can disable vehicles and shields. It's unique that the attack Feint in Pokemon is one of the only moves that can hit through the ultra defensive move Protect. It's unique that Mini Mario can run on top of water and up walls in New Super Mario Bros. U. But if these games allowed players to stack these abilities with other weapons, moves, powerups, etc., these unique qualities would become a lot more common. Instead of give and take, players would simply take. And when player abilities are more similar in this way, interesting choices are harder to develop because dominant strategies (where players stack up all of the best abilities) are more likely.
It's great that you can only have one power-up at a time in Mario, or carry two weapons at a time in Halo. Similarly, Spelunky has a great, dynamic way of limiting the use of certain items. Since players can only carry one item in their hands at a time, they cannot combine the power of the shotgun and the webgun, for example. To use both, players have to learn to juggle items throughout a level by picking them up, dropping them down, and sometimes throwing items across the stage. Since juggling engages the dynamics of space and increases the likelihood of a deadly chain reaction of events, the increased difficulty balances out the increased item power. And ultimately at the end of each stage, players can only carry one item through the door. Tough decisions are sometimes made like taking a golden key with you instead of a shotgun.
At the same time, Spelunky let's player stack just about every other powerup in the game. At one time player can have the jetpack, pitcher's mit, climbing gloves, spring shoes, spike shoes, spectacles, paste, compass, crysknife, parachute, Kapala, Udjat eye, Ankh, Hedjet, Book of the Dead, and Vlad's Amulet. That's a lot of stacked abilities! And as I explained in detail in my review of Spelunky, this stacked power allows players to undermine much of the challenge in the game rather than embrace it and play through it.
Kid Icarus Uprising suffers from the same fate with its design of stacked weapon abilities and stacked powers. The weapons allow players to stack up to six of the games 40 different properties to enhance a weapon. These properties can have positive or negative effects, but in general, the pros of the fusion optimization far outweigh the cons. If your weapon gives you good defense, you can stack on more defensive and health bonuses to stretch the design space even further. A little variation is a good thing. It can give a game enough flexibility to break out of design space roles. But too much flexibility just smoothes over the design space while making overpowered abilities and dominant strategies more likely.
One of the biggest design drawbacks of Kid Icarus Uprising is that its feedback in gameplay cannot convey all of the complexities and variations that go into the weapons. When you're battling, you have too little information to determine if the weapon you're fighting is vastly different from the default specs. And even if a weapon inherits new cons due to the fusion process, there's no feedback to clue you in making it highly unlikely that players will be able to make informed decisions about any new weaknesses.
And it's even worse for the powers system. Yes, which powers you can take into battle are balanced using a clever grid based packing system. However, once in battle, the graphics don't have enough fidelity to convey which powers players have activated, and he touch screen doesn't have enough room to display all of the icons to indicate as such. So instead of making informed decisions about who to fight based on their weapon and power combinations, you have to guess and play largely out of your long-term memory, which is still somewhat ineffective considering the wide range of variation.
Stacking abilities is simply not a very good idea if you want to design games of interesting choices because it's harder to implement into a clean and balanced gameplay system. For this reason, many RPG systems and sub-systems do not support gameplay of interesting choices. The abilities and powers players unlock over time are often just stacked on top of each other with no new drawbacks to consider to play around. This is also why Advance Wars: Dual Strike is the worst game in the series in terms of gameplay. The leveling and perk system for the double COs allows for players to find really cheap combinations, expand CO strengths, and cover up CO weaknesses. The delicate balance of the Advance Wars design space was ruined, yet restored (ironically) in the 2nd DS Advance Wars game Days of Ruin.
At the end of the day, to better enjoy and appreciate games for what they do best (gameplay) players have to learn the rules. This means that players have grasp and memorize the design spaces of games. As a designer, understand when your design choices make this harder for players or help them along.