Stayin’ Alive
a systems game by Lucas W, Leyth T, Krystal L, and Ngoc T
Find our game here: Link to Stayin’ Alive. If prompted for a password, it is “lizardman377g” named after our love for animals and 377g.
Artist’s Statement
Stayin’ Alive explores what it means to live on the edge of survival, where every player choice ripples through an ecosystem and shapes the future of a species. Beginning by inhabiting the life of an endangered kangaroo rat trying to revive its population in the shrublands of Southern California, players experience a world defined by resource availability, predator behavior, and seasonal cycles, as they interact in a living system that continually asks players to weigh risk against reward.
Fig 1. You play as the Stephens’ kangaroo rat, an endangered species and a prey of hawks
At its core, the game is a study of ecological survival. Players must balance immediate, moment-to-moment needs with long-term decisions about reproduction and resource management. Through this experience, they come to understand the precarious reality of conservation, witnessing firsthand how a small prey animal’s fate hinges on shifting seasons and predator pressures that can mean the difference between recovery and collapse.
Our intention is not to lecture about endangered species, but to let players feel the tension that defines their existence. Fear and urgency sit alongside small victories and the satisfaction of planning one step ahead. Together, these moments communicate the true stakes of survival. By giving players control over a creature that rarely receives attention, we invite them to find value in the overlooked and vulnerable.
We make risk tangible through mechanics: tight movement constraints, unpredictable predators, limited shelter, resource depletion, and critical choices between feeding and breeding. Layered systems such as seasonal shifts, den management, and population dynamics, all encourage players to think beyond the present moment and experience the emotional weight of long-term survival.
Ultimately, we hope Stayin’ Alive sparks empathy and curiosity, leaving players with a sense that every organism, no matter how small, carries a story worth protecting.
Overview
It’s not easy being a kangaroo rat in Southern California.
In the dry outskirts of SoCal, sunburnt hills give way to sagebrush flats, and the whole desert holds its breath between day and dusk. Red-tailed hawks drift like shadows on the wind, coyotes slip in and out of their dens, their eyes glinting silver in the light.
And then, beneath it all, the endangered Stephens’ kangaroo rat emerges. Small. Swift. Endlessly outmatched. But stubbornly trying to stay alive.
Every rustle might be danger. Every seed might save your life. Can you gather enough food to revive your population before the desert claims you?
Target Audience
Our primary target audience are individuals who like nature, animals, and learning about natural systems. Stayin’ Alive is designed for players who enjoy strategy experiences that blend atmosphere, survival tension, and ecological thinking. This includes fans of resource-management games, creature-based simulations, and survival loops where thoughtful decisions matter more than fast reflexes.
Final Concept Map
Fig 2. Final concept map, highlighting the ecosystem and food-chain systems within the game
History Versions of Game
Link to Google Drive Folder w/ Playtest Videos: Playtest Videos. Please use your Stanford email to view.
Phase 1: Figma Flow
Phase Description: For the first-ever concept prototype, we designed a Figma page where players could move around images of animals to path them towards food/water every turn. This Figma page was paired with a Python script that tracked the state of every animal on the board.
Our goal with this very early iteration of the game was to analyze how controlling every animal in a larger ecosystem could feel, giving us insights into player behavior and what strategies could be developed.
Ryan, 11/06
We tested this version of the game with Ryan, a coterm student at Stanford studying game design systems.
Fig 3. Playtest with Phase 1 – Figma Prototype and Python script for state tracking
From the playtest with Ryan, we discovered that there was a potential for difficulty and strategy, even in this primitive first prototype. For example, Ryan commented that “Lion A should not have died, that was my negligence” after losing one of his animals to dehydration. When there are many moving parts and needs of various animals, it can become complex pretty quickly.
Ryan’s main strategy in this playtest prototype was to “reproduce rabbits as they migrated, and feed the weakest ones to the lions”. Although this was a valid strategy in our game, this did not accurately represent choices made in real-world ecosystems, so we decided to make our digital port focus more on a single, vulnerable animal. From the point of view of one animal, it would suddenly become a game where your goal is to understand the system and thrive within it. We liked the premise of this idea much more than the previous one, where the player had god powers over the animals’ interactions.
Additionally, Ryan pointed out that he thought the game “could have gone on forever” if he had enough time. This was true, primarily due to the lack of a goal and infinite resources (the water and food, at this point of the game, did not need to replenish). We implemented this feedback later on after spending most of our time working on a digital port of this game, but we agreed that this was a much more realistic view of the system.
(+) Focus on one vulnerable animal (less cognitive load, reflective of choices in a real-world ecosystem; added after Brydie’s playtest)
(+) Make resources finite (added after Brydie’s playtest)
Phase 2: Early Unity Prototype
Phase Description: For the first digital prototype of the game, we ported over the Figma game and made changes with each playtest. The early Unity prototype phased worked to create the main mechanics, establish a win and lose condition, add different game objects into the game, and core loops where those objects could interact.
Alex, 11/09
For this first Unity prototype playtest, we tested with Alex, a college-aged student who loves cats and video games. Alex found that the game didn’t have a clear goal, but noted that he liked the complex pathing system. Alex found that it was impossible to run away from the lion once it was “locked” into you. It was also hard to know what the lion was going to do next. The main changes we noted from this playtest were to (1) add better graphics and maps, (2) add a clear goal, and (3) ensure that it is somehow possible to survive from predators who “lock” onto you.
(+) Goal became to get ready to hibernation (3/3 “readiness”; added after Brydie’s playtest)
Fig 4. Playtest with Phase 2 – First Unity Prototype with pathing for all animals, hunger and thirst, and grass and water as resources
Brydie, 11/10
For this first Unity prototype playtest, we also tested with Brydie, a college-aged woman who is an animal lover and medium familiar with base builder games. She played through the prototype quickly and reacted very emotionally to what was happening on screen, at one point blurting out, “No! He’s coming for the little bird,” [1:52] which showed how immediately she connected with the creatures. When one creature was falling behind she reacted with urgency, saying “He’s hungrier now.” [2:08] and soon after “He needs some food.” [2:19]. These instinctive, caretaking reactions confirmed that players quickly empathize with the individual animals.
She liked the core dynamics of the system and enjoyed the challenge of planning paths around predators and obstacles. She described the puzzle element as fun and appreciated the tension of making “the optimal path” under pressure. When she discovered that movement wasn’t restricted to single steps, she noted “You can also move more than one space…” [2:23], which highlighted both her strategic curiosity and an instructional gap in the prototype. Her comments about creatures nearing danger, like “but now he’s gonna die.” [4:12], showed she was actively reading the board state and emotionally responding to it, which is exactly what we were aiming for.
She also asked whether players could “just make the lion die,” which revealed how she wanted more direct intervention in predator behavior. She noted that certain situations felt impossible, like when the giraffe had no choice but to starve or get eaten. She also wanted grass to deplete and regrow elsewhere (similarly to Ryan’s playtest), seasons to come and go, and elements of randomness to make the world feel less like a fixed puzzle and more like an ecosystem.
Fig 5. Playtest with Brydie
To address Brydie’s request of involving renewable resources (which we felt would increase the potential for strategy), we decided to make certain food items non-renewable if the player did not consider the consequences of their actions. We did this through a two-tiered grass system, where tier one grass could grow if given enough time, but if fully harvested, would be gone permanently. This loop of eating grass to keep the player alive, while also making sure there was still a healthy amount of grass left, became one of the most fundamental core loops of our system. We also added seasons to emphasize the progression of time (since it did not make sense to us that an animal would die in a few turns) and introduce new challenges such as Winter, where food would be especially scarce. We also switched to using Perlin Noise to randomly generate the map rather than work with a fixed environment, to create more replayability.
(+) Perlin Noise map generation for replayability and to feel less of a puzzle game (effective after Luna’s playtest)
(+) Seasons to emphasize scarcity (rabbits don’t breed in winter, grass is slow to grow)
(+) Added Krystal’s artwork for better graphics with style recommendations from Brydie
Alex C. & Alex Y., 11/10
Alex C. and Alex Y. are TAs from Lucas’ CS42SI course. This playtest occurred after CS42SI. Alex Y. noted that he did not like having to click on “Next Turn” every time he wanted them to progress. Alex C. suggested we change the movement to be similar to Crypt of the Necrodancer, a 2D tile-based game. We decided to remove the “Next Turn” button and add a timer for each turn. With the timer system, objects move either when you make your next move or when the timer runs out. Alex C. also recommended added fog of war to really immerse you into the role of a prey with limited knowledge. This fog of war made exploration much more present in our game as a type of fun. The fog of war allows for a new loop of exploring new areas, gaining access to more animals to control, and using those animals to further explore. With this system, players still drew out paths, but instead of clicking the next turn button to move along the path, the movement happened on the aforementioned time steps.
(-) Removed clicking “Next Turn” after every turn
(+) Added a timer-based turn system
(+) Changed animals to mimic more realistic predator-prey relationships (PREVIOUS: duck, lion, giraffe → AFTER: endangered European hedgehogs, wolves, white rabbits, hawks → LATER: Kangaroo rat, wolves, brown rabbits, hawks)
(+) Fog of war
Luna, 11/12
Fig 6. Playtest with Luna
Luna is a college-aged game designer from the CS377G course.
13:55 : “Is this a successful game, yes. I played it. I wanted to win […] I think you get to know the objective really well.”
15:37: “It’s a lot of distance to cover and I have no protection the entire time”
Luna also noted that they also did not like the mouse dragging movement.
(+) Added different behaviors & relationships in game and ways to survive predators (bushes, dens, predator lost interest, etc.)
Madison, 11/12
Madison is a Stanford student enrolled in 377G. The main takeaways from Madison’s playtests were that she was able to find a strategy of using the rabbits as bait to survive against the predators, but that she didn’t like the time-based turns. She didn’t get far, despite playing several times, but found the game fun through challenge-based fun. Additionally, being able to reproduce in the den or use it as a safe place was unclear to her. During the playtest, Madison asked “why are they so fast?” [3:15] in relation to the speed of the predators vs. how fast she was moving. During the playtest, there were many times she struggled to move due to the dragging movement, which was also a main complaint in other playtests.
This comment, combined with Brydie’s earlier misunderstanding signaled to us that we needed to figure out a more intuitive system of movement. From here, we switched to a WASD (or arrow keys!) control scheme and instead made time only step forwards when the player moved, rather than requiring the player to make careful dragging movements with the mouse. We also decided to make the den system more complex than just providing extra lives when standing on top of it.
(+) Switched to WASD movement
(-) Removed point and drag mouse movement for ease of pathing
(-) Removed time-based turns due to stress and frustration, making time pause between movements, but continue if hiding
(-) Removed animals walking on water
Fig 7. Playtest with Madison, where rabbits are seen walking on water
Butch, 11/14
Butch is a college-aged CS377G course assistant. Butch had suggestions for visual changes, alongside some changes to the game mechanics. Firstly, he suggested we change the appearance of the rabbit dens, noting that they seemed too similar to your own. Secondly, he found that the game lacked a system, noting that getting food to survive is not enough to constitute a complex system. One example he provided was maybe adding some consequences to overeating your current food (i.e. eating all of your berries leads to no berries growing later). Lastly, he also was frustrated when predators were right outside his den, making it extremely difficult to leave without dying.
We decided to make certain food items non-renewable if the player did not consider the consequences of their actions. We did this through a two-tiered grass system, where tier one grass could grow if given enough time, but if fully harvested, would be gone permanently. This loop of eating grass to keep the player alive, while also making sure there was still a healthy amount of grass left, became one of the most fundamental core loops of our system. We also added seasons to emphasize the progression of time (since it did not make sense to us that an animal would die in a few turns) and introduce new challenges such as Winter, where food would be especially scarce. We also switched to using Perlin Noise to randomly generate the map rather than work with a fixed environment, to create more replayability.
(+) Grass foods are non-renewable foods if you overeat them (spreads faster if there is more overgrown grass & faster in certain seasons)
(+) Predator dens can’t spawn by your initial den
Phase 3: Breeding and Seasons!
Phase Description: From here, we realized that we needed more choices that the player could make in order to develop strategy. To double-down on the survival aspect of the game, we made food the primary economy system. Food naturally grows around the area if the player is conservative with their grass harvesting (the tap), and is spent on breeding new workers which, in turn, can help accelerate grass harvesting (a drain and tap). This creates a positive feedback loop that requires the player to consider good placement (so workers do not get eaten by predators, and workers can find grass) and proper resource management, as more workers means more harvesting, and therefore less grass in the surrounding area.
We also realized that seasons were not as punishing as they should have been, so we tweaked the balance of various seasons. For example, Winter now removes 80% of the grass on the board to give the player a sense of scarcity, and a need to stockpile resources in preparation for then. (In hindsight, we would probably make Winter even more punishing than just this, as further playtesters demonstrated it was still somewhat easy to survive through.)
Amelia, 11/14
Fig 8. Playtest with Amelia
Amelia is a college-aged player who does not have much experience with video gaming, and considers herself a cozy gamer. Once the core loop clicked, Amelia quickly understood hunger and the urgency around finding food. She consistently recognized when her character was starving and immediately shifted her attention toward locating edible resources. This was great to know that the hunger bar was working well in cuing people to eat.
The largest point of early confusion came from the rabbits since she wasn’t sure whether they were predators, neutral creatures, or food. Because they didn’t interact with the player, their presence felt ambiguous and distracting. She also struggled with item usage and wasn’t sure how to activate or consume items in her inventory, suggesting the need for either a clearer tutorial, contextual hints, or a help menu. The fog of war also confused her at the beginning since she couldn’t tell what parts of the map were reachable, but this resolved itself naturally once she spent more time exploring.
(+) Added tooltips below text when an item is selected to indicate that they can be used for special effects
Abram, 11/15
Fig 9. Playtest with Abram, with the finalized animals, inventory (3 which was changed to 5 later), and rocks that block movement
Abram is a college student (at another University) who deeply enjoys complex systems games whom we met at a board game night at Stanford. Abram took his time harvesting resources and showed interest in building up a sustainable loop before expanding outward. He quickly understood the fog of war concept for revealing land and navigational context, and he liked exploring the environment piece by piece. He also responded well to the idea of map geography shaping strategy, particularly when thinking about how water segments the terrain and how that affects movement.
However, he felt that winter didn’t pose enough of a threat and suggested that seasonal events like colder temperatures and grass dying back would add needed tension. We found that the fog of war helped him understand the layout of the land but didn’t help him anticipate creature movement, which he noted sometimes felt unfair. He also pointed out that animals could overpower the main character more often, implying that predator pressure needs to be turned up for players seeking a challenge. He encouraged leaning more into environment-driven difficulty, such as weather events, harsher winters, and water-based obstacles dividing the map into meaningful regions.
(+) Drastically increased the punishments of Winter, forcing grass in the world to die in addition to having grass grow much slower
(+) Tweaked water tiles to be more interconnected rather than splayed out
(+) Changed specific animal behaviors: coyotes now take breaks when chasing for too long, but predators gain infinite sight range to hunt down prey when critically hungry
Phase 4: Workers, Den Inventory, and Tutorial
After implementing a functional den inventory system for storing food as a currency, we expanded on these ideas by allowing the player to build additional dens, and a primitive mini map which shows the relative positions of a player’s dens, and allows them to teleport between them.
Amaru, 11/17
Fig 10. Playtest with Amaru on the older tutorial scene
Amaru is a college-aged student and game designer in CS377G. Amaru enjoyed several aspects of the playtest, including the fun animations, the presence of bushes and environmental elements, and the overall concept of a resource-management game with workers. He also liked discovering efficient strategies on his own, such as collecting all food before breeding, and found the worker system logically fitting within the genre. However, he was often confused by unclear UI elements, missing labels, inconsistent layering, and mechanics that weren’t well explained—especially the purpose of rabbit dens, how assigning workers affects MVP, and how predators interact with the player. He didn’t initially notice key systems like the hunger bar, and rarely encountered predators (partly got an uneventful map generation), making the world feel low-stakes and lacking consequence or scarcity.
He suggested a wide range of improvements, mainly focused on better onboarding, clearer communication of systems, and stronger gameplay feedback. He wanted clearer labels on maps and menus, more emphasis on the breeding screen, and a better explanation of how resources, workers, and animal relationships function. He recommended stronger predator threats, more environmental cues (especially in winter), and additional den management options. He also expressed interest in more interactions with animals, multiple paths to victory, more meaningful tradeoffs, the ability to refresh workers, expanded winter effects, and improved visual feedback such as death animations.
While we were able to make most of the changes Amaru suggested, we were unable to add more interactivity to rabbits. We did bump up the amount of rabbits and added heart animations (they also breed) to emphasize that they are your competitors for food. Additionally, players can still use them as bait, a strategy that Madison noticed early on.
(+) Onboarding much more fleshed out now, with information about how to manage a den, highlights on what each UI element in your screen means (like hunger and season), and no need to walk backwards (after Butch’s 11/19 playtest)
(+) SFXs at the start of each winter to provide more seasonal cues (sorry, no changes to in-game sprites due to art & time constraints!)
(+) Fixed some layering issues (worm under puddle and UI being covered by things)
(+) Dens aren’t made by pressing “B” anymore, but rather by using sticks. This way you need to go to the trees (hawk’s den) to collect sticks, encountering more danger. A few sticks still spawn in safe, random places to help you get started. (added after Butch’s playtest on 11/19)
(+) Players can withdraw food (added after Butch’s playtest on 11/19)
(+) Death animation where animals turns red and flip over (added after Butch’s playtest on 11/19)
(+) Heart breeding particles on rabbit dens (added after Butch’s playtest on 11/19)
Evelyn, 11/17
Fig 11. Playtest with Evelyn, where you still need to use B to make dens (and it doesn’t use sticks)
Evelyn is also a college-aged student and game designer in CS377G. Evelyn immediately felt the tension of hunger and resource scarcity. She naturally gravitated toward a food-first strategy and reacted strongly once the systems finally clicked, especially after discovering the last patch of grass in the world and building a den by it. Once she understood grass regrowth and the broader loop, she entered a flow state and described the run as “chill and fun.” She also liked that worms reliably spawned near water which meant that she could strategize around the environment
However, Evelyn thought the tutorial was confusing because many things weren’t explained clearly. She also wiped during the tutorial when food ran out, and a predator died of hunger during onboarding, so both moments signaled that the tutorial needed more guardrails. She repeatedly asked about the goal and why the predators were gone, so without other animals surviving most of her session became resource stalling rather than actually interacting with the ecosystem. She also said she couldn’t tell when seasons were changing and also she realized she could simply stall indefinitely if the world’s resources were exhausted.
(+) Made it so there is a minimum population of predators + rabbits at all times to prevent them from all dying
(+) Added the changes that overlapped with Amaru’s feedback
Winnie, 11/19
Fig 12. Playtest with Winnie with a whopping 33 Kangaroo rats on him
Winnie (Tianze) is a college-aged student and game designer in CS377G. He was able to beat the game within ten minutes of playing (including tutorial time), accumulating food and spamming the “breed” button at the very end to match the MVP. Because this was an unintended way to win, we had to start thinking about how to discourage this behavior. He also expressed some confusion about game mechanics that weren’t explained well in the tutorial. We made these changes later on in our revisions, as we wanted to playtest first with others to get further validation on strategies that players would take while playing our game. Additionally, he suggested the food bar scaling with the Kangaroo rats following you or one of them dying when the hunger bar reached the bottom. We tried both of these ideas out, but decided to opt out of them as the scaling threw players off and it was nearly impossible to die from hunger.
Butch, 11/19
Fig 13. Playtest with Butch, where he was unpleased with our old version of the tutorial
In this round of playtesting, we also playtested again with Butch, so this is his second time seeing the game. Once he understood how the game’s systems connected, Butch responded positively to the idea of progress between dens and liked the notion that dens could act as a network. He found the concept of followers helping with tasks promising and noted that being able to see what they were doing would be “very helpful.” He also liked the idea of teleporting between dens in theory, and appreciated the direction the game was taking in terms of strategy and pacing.
However, there were a few points of confusion. For example, onboarding was his biggest pain point. Specifically, he felt the tutorial should be mandatory and found the existing tutorial text outdated and occasionally unclear. (“I’m just going to create babies with myself”). He also felt the tutorial environment wasn’t kind enough for the player to learn, and also repeatedly asked questions like “Who am I breeding with?” that emphasized that onboarding should either explain all future mechanics or clearly state that it does not cover everything, so players aren’t misled.
Several mechanics were also unclear to the player like teleporting between dens wasn’t intuitive, grass spawning wasn’t explained, and the mini-map “made no sense” because it didn’t show predator dens or provide actionable information. He also struggled when children physically blocked his movement and felt the eat key should be closer to WASD to make food consumption easier during frantic moments. He asked about colorblind accessibility, suggesting the UI might need more contrast or labeling. Finally, he reached the extinction screen and expected a restart button that wasn’t there.
Given that other testers did not mention the struggle to press the eat key as a pain point, we decided not to move forward with this feedback on the controls. And also decided to focus on the mechanic changes and think more more focused accessibility features as a possible P4 exploration.
(+) Added a small note that says the mother Kangaroo rat is in the den waiting for you…
(+) Added minimap overlay in the den management UI
(+) Adjusted tutorial to address Amaru, Evelyn, Winnie, and Butch’s feedback
Phase 5: Final Polishes
In order to reinforce the core loop of creating dens with good access to food we added seeds that allow the player to plant more grass which have a chance to drop from grass harvesting. We also reworked the inventory system in order to actually keep track of individual pieces of food and other items the player deposited. This meant that the player could actually withdraw food from their den in order to eat, or store things like sticks and seeds for later use. It also meant that the food brought in by workers/followers, could actually be eaten by the player and other workers. These two changes really enhanced the den building loop and the seasonal loop by adding more agency over the resources players can save up for things like winter.
Daniel, 11/21
Fig 14. Playtest with Daniel, where he really liked the cutscenes (yay!)
In our phase of final polishes, we playtested with Daniel, who is a college-aged man who is not that familiar with base-builder games specifically but plays a lot of games and likes nature. From the beginning he reacted with enthusiasm, immediately responding to the visual immersion with “Wow, I love that.” [1:10]. Early on, when he discovered the worker system, he exclaimed “Oh, it’s automation!” [2:01], showing how quickly he understood the underlying loop.
As he continued playing, he explored the mechanics with curiosity and excitement. He reacted to discovering items with “that’s cool” [2:35] and began developing a strategy around breeding and distributing tasks: “But if I go here… I can assign more.” [2:59. He naturally started optimizing his base layout, noting “I’ll plant the grass for next day. I feel like that makes sense.” [4:41], and later commented “I love the music and the sound effects… adds a lot to the ambiance.” [11:05]. He also loved the random map generation and saw its value immediately for replayability, and noted that the teleportation was important strategically to evade the coyotes, “I used it a lot, felt like a little mouse.”
However, we noted a few balance issues from this playtest, for example teleportation’s lack of cost confused him since once automation kicked in, food became abundant enough that teleporting felt “free”. He also flagged that once baby automation ramps up, the game can feel too easy.
From this, we decided to change how much food babies eat, and scale how much it costs to breed, that way the end game is still challenging.
[Image of playtest with Daniel.]
(+) Added coding implementation for cost to teleportation BUT decided to not to have any due to other players already having a hard time (aka Krystal and Ngoc)
(+) Food to breed scales with your population (added after Ari’s playtest)
Ari, 11/22
Ari is a college-aged student who loves to make games. Ari’s playtest revealed some usability and balance issues. He struggled to identify which den he was in because the UI indicators were too small, making navigation confusing. He also discovered a tutorial bug that let him return to a previous den, which ended up softlocking him. In terms of gameplay balance, Ari bypassed the intended worker-based resource loop by gathering all fifty grass from a single patch on his own, noting that the current worker cost makes them unnecessary. He even mentioned planning to plant seeds and have his workers collect the grass, but ultimately found it more efficient to ignore workers entirely, suggesting that adjustments to worker costs or resource distribution may be needed. We decided to bump up the cost of breeding as you progress in the game, requiring you to either be intentional with your grass planting and strategic with den / worker placement.
(+) Adjusted the size of some UI elements
(+) Fixed softblocking bug in tutorial
(+) Food to breed scales with your population
(+) Fixed surface bugs like grass spawning off grid and in water.
FINAL: Angela, 11/22
In one of our final playtests, we playtested with Angela, who is a college-aged woman in our target audience, whose favorite games are resource management like Minecraft and Stardew Valley. She also LOVES animals and nature.
In this playtest, she loved the art style and enjoyed the core survival loop, and also found the limited inventory + health bar pressure engaging. She liked assigning children and called it “interesting” as a resource-management twist. Her reactions captured her curiosity like “Go! go! go!” [15:38] when running away from predators, and asking questions out loud as she strategized where to place dens like “do worms spawn near lakes?” [22:41]. She appreciated the increased cost of dens and breeding, noting that it created meaningful decisions. At the end she said, “I think it was fun… I wanna play it again!” [20:35] which is a great sign.
A few times she was surprised by how quickly hunger hit: “How am I already starving?” [8:09] “gosh, stop starving” [8:53]. This was great to see that she was putting herself in the shoes of the animal and felt how urgent the need to find food was.
One point she noted was that predators felt a bit too sparse and too passive, making the survival loop feel gentler than expected. She did feel like the hunger depletion was well balanced though. The moment grass mysteriously disappeared in winter made her question whether the mechanics were acting on her: “Where did my grass go?” [14:28] and it was just the changing seasons. This was a good outcome as before other players didn’t feel winter was that punishing, but this indicated the difference in season was definitely felt by the player.
Fig 15. Screenshot from Angela’s playtest
FINAL: Leyth’s Friends: Violet and Miguel, 11/22
Fig 16. Playtest with Violet and Miguel with our super awesome logging system in view
Violet and Miguel are both “hardcore” video game players, and enjoy a wide variety of game genres, but have very high standards for games. Violet and Miguel are both gaming friends of Leyth’s. Leyth and them have participated in high level Destiny 2 races, and are good at developing game strategies together. They both enjoy “breaking” games and finding exploits, which is why we wanted them to playtest the close to final version, that way we could find issues with our system representation and balancing rather than programming errors from earlier iterations. We decided it would be good to have them test the game at the same time in a call with each other, so they could bounce ideas off of each other and “break” the game. They did end up finding a few exploits which existed as unintended consequences of intentional design decisions, as opposed to bugs.
One of these exploitative strategies is one another playtester found, where unassigned workers (intentionally) don’t drain additional food from the player’s storage. During V’s playthrough, she never actually consciously pointed this out, but used a similar strategy to Ari in the end, albeit with some modifications (like having one den which harvested the majority of the grass, while Violet replanted it). Violet also thought she found an infinite food glitch at one point, where grass can be eaten infinitely and have seeds be dropped, planting more and repeating the cycle. When after a few cycles of this, the seed didn’t drop (since it has a random chance to) she said “damn you got me!” She seemed to be having a lot of fun exploring the bounds of the system.
Miguel played the game much more in the way we intended as designers, slowly building up bases and grass around them. He started making projects and goals for himself, treating the game almost like a sandbox. Namely he created something he called a “garden” in the top right corner of the map. Miguel kept bringing many seeds to this corner to plant grass, and each winter when the garden would mostly die he would exclaim “NOOO MY GARDEN” to the point that every time he did this, V would start laughing because of how often he did this. In Miguel’s run we also saw cool emergent behaviors that arose from the constraints of the system. One cool feature that took shape in Miguel’s world was a field of grass that grew quite a bit, because it was “defended” by a hawk from the rabbits. Miguel eventually got locked into a sort of stalemate with the game, and his run ended up going almost an entire hour. For almost 30 minutes of this, he hit a point where his taps and drains were perfectly balanced, to the point where he was gaining followers exactly as fast as he was losing them to starvation. This tells us that we will need to rebalance our drains as we continue work on this game for P4.
Probably the coolest thing about this playtest though, was seeing the buy in from two relatively picky gamers. While watching Miguel’s playtest, Violet said “Okay you’ve got me hyperfixated on this game, I’m writing up some notes on things I think you could improve or add.” Violet tends to be very picky with games, and seeing her take such an interest in the mechanics and base system we created was really cool! Both of them had a lot of fun, and after losing after an hour of gameplay, Miguel was laughing talking about how much fun he had playing the game, and the two of them proceeded to talk about different possible strategies for a few minutes afterwards.
Fig 17. Violet’s notes on Miguel’s playtest
FINAL: Ngoc’s Younger Brother: Ryan, 11/22
Ryan is a younger player who is curious about biological concepts in school, nature, and animals. Unfortunately, Ngoc did not crop her OBS screen input right so the bottom of the game can’t be seen in the recording. At 4:55, Ryan says “I’m going on an expedition and my workers will go on and find food.” At 5:56-6:30, Ryan says “Let me hide. I’m not here. I’m not here,” as he hides in a bush. During this time, he runs into a lot of dangers, like wolves, hawks, and hunger. At 7:40, he makes a den and puts his sole follower in the den, noting that he “trusts [it] to be the sole survivor.” He plays risky, not having a single unassigned baby until 8:40. At 8:55, he calls his workers “loyal servants.” At 9:05-9:30, he didn’t notice what grass seeds do, but once learns, he says “Oh it’s like a farm. So if I let it grow, it’ll become more valuable.” At 11:05, he begins to plant the seeds near his den. At 12:36, he dies due to not having any followers on him while a hawk dashes at him twice. While playing the lost cutscene, he noted at 13:05 that he “feels guilty” due to his followers “all being dead.” At 13:35, he mentioned “this time, I need to keep a bit with me, like two. They can go on an expedition, while I do my own things.” At 14:06, he starts to plant crops near his den this time. At 16:47, he goes towards a tree to collect sticks. He panics and says “holy. No. No. No. Get away from me I don’t like you,” as predators were nearby. At 18:09, Ryan says “he needs to work on the right side, cuz we’re running out of food.” At 18:38, he continually loses all of his animals to starvation, because you can see they’re running around and can’t find any food. In fact, he can’t find any food for himself.
Ryan’s playtests highlighted several emergent consequences driven directly by player decision-making. For example, in the first playtest, he wanted to send all of his followers out for an “expedition,” leading to him getting one-shotted by a hawk. The poor management of workers led to his difficulty surviving. In the second playtest, he built a majority of his dens in the same bottom left corner, overpopulating the area, and causing a shortage of food until all of his kind starved to death. All in all, the playtest demonstrated that the game was difficult and the choices Ryan made in terms of den location, breeding, Kangaroo rat assignment, and grass management ultimately contributed to his failures in the game.
Fig 18. Ryan, pleased with the bush hiding mechanic
Playtest Appendix
| Date | Player | Phase | Link |
| 11/06 | Ryan, a Stanford student studying game design systems | 1: Figma | https://mechanicsofmagic.com/2025/11/06/natural-ecosystem-playtest-1/ |
| 11/09 | Alex, a college-aged student who loves cats! | 2: First Digital Prototype | https://mechanicsofmagic.com/2025/11/10/p3-playable-first-draft-ecosystem/ |
| 11/10 | Brydie, a college-aged woman who is an animal lover | 2: First Digital Prototype | https://drive.google.com/file/d/14k-s-WYq3rZuPFqKsSQ7AMv5gHx9i7sV/view?usp=sharing |
| 11/12 | Luna, a Stanford student studying game design systems | 2: First Digital Prototype | https://drive.google.com/drive/u/1/folders/1eshyBv-wfk00uH1Hz0o1DeLUufpm5df1 |
| 11/12 | Madison, a Stanford student studying game design systems | 2: First Digital Prototype | https://drive.google.com/file/d/1q0PFfHwczRwxq_ENR_NDIUSnQGRLHO_e/view?usp=sharing |
| 11/14 | Butch, a Stanford student studying game design systems | 2: First Digital Prototype | N/A |
| 11/14 | Amelia | 3: Breeding and Seasons! | N/A |
| 11/15 | Abram | 3: Breeding and Seasons! | N/A |
| 11/18 | Amaru, a Stanford student studying game design systems | 4: Workers, Den Inventory, and Tutorial | https://drive.google.com/file/d/1HickqXCiheFpn47IkTffqPF67UMcSHSp/view?usp=sharing |
| 11/18 | Evelyn, a 377G student | 4: Workers, Den Inventory, and Tutorial | https://drive.google.com/file/d/1h_6yGYLY5001d0AmS_VHEa81DrkdG07r/view?usp=drive_link |
| 11/19 | Winnie, a 377G student | 4: Workers, Den Inventory, and Tutorial | https://drive.google.com/file/d/1unRCo45ishkE7CKY4INba_DhbbZr1P3H/view?usp=sharing |
| 11/19 | Butch, a Stanford student studying game design systems | 4: Workers, Den Inventory, and Tutorial | https://drive.google.com/file/d/1m8t0NpTNAY-f7wKlsn_7BjmOFloI6g-M/view?usp=sharing |
| 11/21 | Daniel, a college-aged student good at picking up games, love nature | 5: Final Polishes | https://drive.google.com/file/d/1gsVELBFFbSDlfV4wRsF0G3kEbz4f1MhI/view?usp=sharing |
| 11/22 | Ari | 5: Final Polishes | N/A |
| 11/22 | Angela | 5: Final Polishes | FINAL: https://drive.google.com/file/d/1FOALUsRiOiUTDH8FJkHc7V7498WxLfZw/view?usp=sharing |
| 11/22 | Leyth’s Gamer Friends: Violet and Miguel, two college-aged gamers | 5: Final Polishes | FINAL: https://docs.google.com/document/d/1ZllidibUp_XBsaGFxPuzELBHW2P6av482t1EBqeronQ/edit?usp=sharing |
| 11/22 | Ryan (not 377G Ryan), Ngoc’s younger brother who loves games | 5: Final Polishes | FINAL: https://drive.google.com/file/d/1DL-8zL_vUtV2L9egF_uL3ISwaa8DzbBI/view?usp=sharing |
Extra Credit / Game Credits
All programming and art was done by our team of four. The product is a high-fidelity game with hand-drawn assets and music and SFXs incorporated.
Art – Main Game Assets by Krystal Li
Art – Title Page, Cutscenes / Cutscene Animations, and some Game Assets by Ngoc Tran
Code by Lucas Wang (main developer), Leyth Toubassy (main developer), and Ngoc Tran
Music and SFXs found on Pixabay and freepd.com (copyright free)
