jQuery PageSlide Demo
The White Abyss. Part 4

by: Richard Terrell  @KirbyKid

March 3, 2014


The three-key door challenges are unlike the fleeting world challenge. Each three-key challenge features a set (non-random) obstacle course challenging players to move from the starting place to the floating heart goal. As static levels, each three-key challenge was designed and tuned to accentuate a specific gameplay idea. For example, one level challenges players to move down and around their starting location. For this three-key challenge, tan, red, and green seeds are the only viable seeds for the initial downward movement.

Like traditional puzzle challenges featuring very limited solutions, with the three-key challenges, there are often "good" or "right" seeds for the job. To access these "right" seeds, players are forced to PLANT useless seeds to the side, effectively wasting them. Droughts are still possible due to the randomized seed order. However, droughts are worse in three-key challenges because pink seeds are unavailable. Whatever seed count players have entering the three-key challenge becomes a dwindling resource that, once depleted, essentially mean it's game over. Having a surplus of seeds helps reduce the chance of losing due to droughts, but players cannot eliminate the possibility altogether.

The open challenge of the fleeting world allows for a wide range of viable options and strategies. Coupled with the various random elements, Starseed Pilgrim's fleeting world gameplay cannot create consistent challenges that reflect the player's precision and skill in a very clear, measurable context. Conversely, the stricter puzzle-like challenges of the three-key rooms generally give players few opportunities to use starseeds in a creative or skillful way as there are often "right" seeds to clear the rooms. While this is typically how puzzle challenges are designed, the degree of uncertainty and variation from the random seed order weaken the puzzle gameplay by reducing the player's ability to obtain usable pieces and to predict with 100% accuracy how those pieces will play out. Ultimately, the negative effects of randomness in Starseed Pilgrim's design is largely the result of random design elements applied in an even way to an uneven (wrinkly) design space. Each starseed is too unique to be useful when arranged in a random, unchangeable seed order. To rely on pink seeds as the only method to increase one's seed count perhaps puts too much of an importance on 1/8 of the seeds. The result is a gameplay experience that plays more like a roguelike--where unfortunate, game-ending scenarios are common as players fight against the odds--than a puzzle or platformer game.

Starseed Pilgrim has a lot of potential to create unique, varied gameplay of interesting choices through the fusion of puzzle and platforming challenges. The challenges and level design are the structure through which all the individual rules and elements of the design space are realized. Sometimes the smallest rule can have a significant effect on the whole experience. My biggest criticism of Starseed Pilgrim's gameplay design is that it features too many random elements that significantly work against player control, which prevents the gameplay from fulfilling much of the potential of its excellent seed design space.


Randomness is a powerful and dangerous tool in game design. It can be implemented into the rules of a gameplay system at every level from mechanics design, enemy design, level design, and beyond. It can add a significant degree of unpredictable variation into a challenge to stress adaptation skills.



The concepts of agency and control sit at the core of what games are. By understanding the game state, making decisions, and executing actions, players can achieve specific results in games. What makes games different from other interactive systems is games have specific outcomes and goals that are meaningful to achieve according to the system. These goals and outcomes require the player to exert mental and physical effort to achieve. This is why many love video games. We use our skills to engage with the challenge and own the results.


The biggest potential drawback to random game elements is damaging player agency. With too much randomness, outcomes can become so unpredictable that player agency (using gameplay mechanics in a specific way to produce an expected outcome) is compromised.


With too much randomness in the system, it's possible that no matter how much effort we put into our interactivity, the results will reflect a random outcome more than our agency-influenced outcome. While the exact ratio of skill to chance, order to chaos, or game to gamble people prefer varies from player to player, to understand the effect random design elements have in Starseed Pilgrim, we must analyze how these elements affect the ratio of skill versus chance, and how such a design affects the ways players can embrace the seed design space.


In Starseed Pilgrim, there is a significant amount of randomness in the mechanics and level design that works against the player's ability to control outcomes in their favor. Starting with the gameplay mechanics, JUMP, DIG, MOVE, and PLANT are all deterministic, i.e. nonrandom. Building a game around non-random mechanics, the only tools players have to interact with the game system, is a good way to support agency. However, connected to the PLANT mechanic are seed order and seed growth, both of which feature random elements.


First, seed order. The order of seeds players can PLANT is completely random. There's a 12.5% chance to get each seed, with two exceptions: 1) The player cannot get two of the same seed in a row. 2) A pink seed is automatically given if one hasn't appeared in the previous six seeds. These two rules reflect a smart level of gameplay balance. The first keeps the seed order varied by preventing the same seed from appearing back to back. Considering that players cannot PLANT a seed into a block of the same color, this rule works well to limit scenarios where players are unable to use their seeds one after another. The second rule is smartly balanced because the seeds are not equally valuable due to the wrinkled design space. Regardless of how you feel about the dark blue JUMP seeds that many consider to be useless or the tan seeds that seem to hold players back more than it helps, pink seeds are essential for building seed count and exploring. Leaving the seed order completely random (with no rule tweaks or weighted probabilities) would have made it possible for the player to experience pink seed droughts, or long periods of time where no pink seeds appear in the seed order. Such a possibility would be game-ending no matter how much skill you have.



The next major random element in Starseed Pilgrim is the seed growth. Three out of the eight seeds – dark blue, light blue, and red seeds-- grow into the same shape given the same initial conditions (based on default pilgrim). The shapes the rest of the seeds grow into vary randomly.



• Orange blocks grow in a straight line 6-10 blocks long.

• Depending on the space restrictions, tan blocks tend to grow anywhere from 10-24 blocks large.

• Purple seeds can grow anywhere from 6-17 blocks large.

• Green seeds can grow from 6-19 blocks.



The shape and size most seeds grow vary randomly. On top of this, the rate at which the seed blocks grow into these shapes and are corrupted by the void varies randomly as well. This design adds one more element of randomness to most of the interactivity in Starseed Pilgrim. While the random design of the seed order creates possible droughts of important seeds, the negative effects of the random seed growth and corruption are more subtle. When coordinating the timing of multiple events, like the incoming void and continually growing pink seeds, it becomes difficult to execute strategies with precise timing. If you find yourself surprised by how close the void is or missing a platforming opportunity by a split second, then it is likely that your timing was thrown off by the randomness.


When growing seeds, maximum seed growth is usually preferable as more growth gives players more options to shape the platforms with DIG, more room to move away from the void, and more blocks to PLANT on. The random, variable size that most seeds can grow adds a degree of unpredictability and thus stresses adaptation skills. For the most part, I find few drawbacks with the random growth. By building strategies around the guaranteed minimal growth, hoping for the maximum growth, and being prepared to fine tune via DIG to stop the growth early, the negative effects of the random growth can be mitigated. The variable range the random design gives the seed growth isn't as bad as the random seed order droughts and bad combinations. The worst the seed growth can do is not grow as much as you want, barring the unwieldy green seeds that have a mind of their own. The worst the seed order can do is create droughts of useless seeds for the player's current objective.


Finally, the goals (star and triple star blocks) of each level are randomly placed throughout the fleeting world. Players never know exactly what lies off screen. We have to explore to find out; a part of the experience I enjoy. However, the randomized placement of these key elements allows for the possibility of droughts where players go long periods of time without finding a triple star blocks while exploring the fleeting world. And because moving down is harder than moving up or horizontally, even if players come across a triple star block below their position, they may not get the seed order to effectively maneuver to it.


Perhaps the biggest effect the random key placement has on the gameplay is a lack of context for skillful play when star blocks are not visible on screen. When in search for off screen star blocks, every move you make can bring you closer to your goal, or not. It's hard to say either way until the star blocks come into view. And because the star blocks are random and hidden off screen, players can only play efficiently by conserving seeds. The frustration of a triple star block drought isn't an "our princess is in another castle" situation. A better analogy would be, sometimes the flag pole at the end of a Mario level is missing, forcing you to keep searching in hope of completing the level as the timer ticks down.


Consider the stack of issues discussed so far. Because the order of seeds players get is random, the way seeds grow is random, and the goals are randomly placed, there are random elements at every layer of the design (mechanics, level elements, level design/goals). The random elements in Starseed Pilgrim's design have the pros of unpredictability and variation. The cons are droughts and a lack of precision in player control when it comes to seeking goals. It's hard to know what skillful play is because the lack of control and precision for the player, coupled with lacking feedback from the level design challenge, make it hard for players to measure their skill. I don't consider amassing a high seed count playing well because my ability to harvest pink seeds depends on the random seeds I'm given to stall the void and to PLANT pink seeds. Furthermore, because I can take a surplus of seeds back and forth between the overworld and the fleeting world, my seed count doesn't necessarily reflect my current play session. I don't consider moving efficiently in any one direction as playing well because efficiency is determined by how well one maneuvers toward star blocks. When you don't know where the star blocks are randomly placed, you cannot determine how well you are pursuing the goal. I don't consider executing unique seed combination strategies as playing well for similar reasons; my most unique combo strategies neither conserve seeds nor move in a direction effectively. Sure, I can pull off some crazy split-second maneuvers to get me out of trouble, but at the same time, I could have played it safe and steered clear of that troubling situation in the first place. After all, there are no points, no score, no bonus object that would reflect my mastery of the system. Starseed Pilgrim's systems and the challenges of its level design don't give me a sufficient ruler with which to measure my skills.


There are random elements in every layer of Starseed Pilgrim's design: mechanics, level elements, and level design.
The negative effects of randomness in Starseed Pilgrim stem from randomness applied in an even way to an uneven (wrinkly) design space.
#Design Space
Read the next article in this series:
The White Abyss Part 5: A Touch of Randomness
comments powered by Disqus

©2014 Game.Play.Critique. All rights reserved. Designed using Adobe Muse

v 1.2, April 2014

status: live