For Klei Entertainment's Nels Anderson, making a great stealth action video game meant throwing out a lot of a lot of the genre's existing tropes. Anderson was lead designer on Mark of the Ninja, a stealth game that brought an unconventional twist to the genre through its 2D side-scrolling vantage point. While the game casts aside certain stealth elements laid out by games such as Splinter Cell, Thief and Tenchu, it does still rely heavily on cornerstones of stealth game design. "With Mark of the Ninja," Anderson explained, "we wanted to make sure that that the fundamental experience of being a ninja was brought across and everything else supported that." After breaking the genre down into its fundamentals, Anderson established Mark of the Ninja's five stealth guidelines he called the "Heresies of the ninja." These "heresies" broke the rules laid out by stealth design's predecessors.

Heresy One: Transparent Stealth System

"In contrast to most stealth games where you have a visibility meter or something like that, in Mark of the Ninja light and darkness are totally binary," Anderson said. "The way your character looks immediately reveals if you are visible or not - if you are concealed you are in black with red highlights, and in light you are fully colored." This on/off approach had one caveat, however. "If you are at the absolute edge of a guard's field of vision, they will catch a glimpse of you. They will start to move towards you quickly but not on full alert." The binary light system - and the fact that all other information is available on screen, such as the limits of the guard's field of vision, or the "ring" of how far a sound such as a smashing light will travel before you destroy it, are intended to "make the systems all very, very understandable." This intention to visualize exactly what will happen before an action is intended to allow the player to know "full out what will happen before they commit." "It allows the player to factor this into their intentional play."

Heresy Two: Transparent AI

Also in aid of intentional play, the guards have three clear levels of awareness, with cues to how they are set into those modes and what they are doing during them (such as racing to the spot the player was last seen, represented by a ghost, or moving towards and then looking at an area where a sound was heard.) "The point is to ensure that if a guard's behavior ever changes, the player will immediately understand what set it off."

Heresy Three: Narrow gulf of execution

The Mark of the Ninja team felt the game's design shouldn't fight against players' intentions. Anderson used an example from the Thief series. "The primary was you affect the world in Thief is through magic arrows. You have a limited supply of them and you can't pick them up again if you waste them. The thing is that the only way you have to aim those arrows is a little tiny reticule, and they're modelled with physics. So actually becoming skilled to affect the world becomes very satisfying, and was very appropriate to what they were trying to provide in that game, but was absolutely not what we were trying to achieve." Placing aiming skill over stealth tactics wasn't the direction Klei wanted to adopt. One problem was the fact that "2D aiming is actually really god damn hard," said Anderson. "In most 2D shooters your previous bullets act as tracer bullets, but in Mark of the Ninja you only want to fire once... the player is a ninja, and if you are constantly bumbling through it kind of undermines the thing we're going for," he said. As a result, the game's solution was "focus aiming," which stops time completely and is not limited by the player's resources. Focus aiming wasn't always an unlimited-use design feature. "[Previously], we added a meter [to the focus aiming] and it was just horrible," Anderson said. "It just completely undercut that whole design decision; sometimes you have to double check your assumptions. Power-balancing is not really objective ... you must let players do what they want to do."

Heresy Four: Limited Consequences for Failure

Anderson said he was wary of using widely-spaced checkpoints "as a means of providing challenge." "This kind of difficulty is very, very dangerous," he said. "In a stealth game people come in with a very patient style of play. There's a lot of waiting. If something goes wrong and they have to do all that waiting again, it quickly descends into tedium," he said. If people get through a section of a challenge before failing, they will generally do that section again exactly the same until they reach the point they failed at. "With Ninja, it basically had a checkpoint between every meaningful encounter," he said. "This allowed us and the player more experimentation in the game. You've all heard of degenerative strategies, where people have worked out one thing that works and it's boring as hell, but they keep doing it because they don't want to lose work. They don't want to lose the last six, seven minutes of play so they just keep doing that boring thing." That's not to say failure states were completely unacceptable -- within reason. Although he agreed that it would have a "significant impact on your engineering," Anderson argued that there is no single feature in a game that features death or failure states that will gain "more positive feeling" than instant load times. He was in fact prepared to argue that "neither Super Meat Boy nor Trials would be as successful as they were if you had to look at a loading screen each time you died."

Heresy Five: Less-Open World Design

"We made [open world] levels," Anderson said, "and they were just god damn terrible." "Mentally mapping 2D space is really not something our brains are meant to do," he said, while admitting he had "done no hard science" on this theory. "In 2D you are simulating a space that's not the way it would be in reality, so your brain has to perform a translation. There's a difficulty there so you can't build a contiguous space nearly as large as you can in 3D." Anderson claimed the sweet spot for a 2D encounter space was the game screen size plus "0.5 to 0.75 of a screen in any direction. ... Any more than that people start to feel lost or not really competent in that space. A little more if it's just horizontal or vertical. If you look at a Metroid or Castievania that's roughly how all of their encounter spaces work. I think that's because they stumbled upon a weird cognitive way the brain works."