P4: Blank Canvas

Reintroducing Blank Canvas.

Play here via Itch.io: https://knockdev.itch.io/blank-canvas 

NOTE: On first load, it may take your browser a bit. You will get a black screen for some time. Refreshes after will load significantly faster.

Artist Statement

Blank Canvas is a narrative-driven interactive fiction game that explores the experience of being a newly graduated artist navigating financial instability, creative identity, and the growing influence of AI-generated art. Set in the year 2030, the game places the player in the role of an optimistic new-grad artist living in New York City, struggling to balance survival, creativity, and personal values within an increasingly competitive and technologically mediated industry.

The game is designed primarily for adult players, particularly college students and recent graduates who are experiencing post-academic uncertainty. Drawing from empathetic design and player identification, the narrative situates the player within an emotionally fraught period of transition, one marked by debt, unstable income, and the pressure to justify creative work. Through narrative agency and environmental storytelling, players are encouraged to internalize the tension between personal artistic aspirations and external financial and social pressures.

A central emotional pillar of Blank Canvas is the recurring fear of familial disappointment, especially the anxiety surrounding parental financial control and expectations. These moments are intentionally framed to evoke reflective empathy, allowing players to connect with the protagonist’s self-doubt while also examining the broader systems that shape creative precarity. Rather than presenting these struggles as purely individual failures, the game highlights how economic structures and cultural attitudes toward art influence creative decision-making.

Creative choice is a core mechanic of the experience. Player decisions—ranging from financial management to whether to embrace or resist AI tools—are designed to shape not only narrative outcomes, but the player’s sense of artistic identity. The game avoids clear moral binaries, instead emphasizing value-driven choices that reflect different artistic philosophies. These choices are meant to feel uncomfortable at times, mirroring the real compromises artists face when deciding how much of themselves to give up in order to remain financially stable.

Overview

It’s the year 2030, and you’re an optimistic new-grad artist chasing your dreams in New York City. As you navigate financial burdens, the rise of AI-generated art, and the pressures of working in a competitive studio, you’ll struggle to find balance between survival and creativity.

Each choice you make — from paying your bills to deciding whether to embrace or resist using AI to make your life easier — shapes not only your career but also your sense of identity as an artist. Will you stay true to your passion, compromise your ideals to keep the lights on, or explore different perspectives?

In a world where technology blurs the line between inspiration and imitation, your journey explores what it truly means to be an artist in the modern day.

—- Spoilers Ahead —-

Iterations & Changes

Addressing Previous Feedback

First and foremost, I needed to prioritize the feedback from previous playtests, including things I couldn’t get to and new comments from P2 peer reviewers. Many of these were added AFTER Ron’s playtest. Addressing these previous pieces of advice became my main focal point. As such a lot of the iterations and changes come from previous feedback.

Title Page & Art Cohesiveness

My idea was to add a realistic hand “drawing” a picture to really emphasize the idea that you’re the artist sketching / visualizing the cutscenes to address a minor feedback point. To elaborate, one critique from Butch was that the art style felt fragmented at times with 3+ artstyles (sketchy, pixel characters, and rendered / high fidelity). Since some peer reviewers felt like the style was “cohesive” and “aesthetic”, I decided to opt out of redrawing cutscenes, and instead added this to the title scene as an alternative to redrawing things, which would take too long. Plus, it looks cool and a title page just helps to add completeness to the game.

Onboarding

“A simple tutorial message at the start explaining the controls would make the onboarding process smoother and more accessible for new players.” 

Onboarding was one of the main things I missed out on in P2 due to time constraints. A major addition in P4 was adding a button for an onboarding screen in the title page. While the controls are pretty minimal (WASD, E, and space), many users just didn’t know what the buttons were, so this was pretty effective (as demonstrated in playtests 2 and 3).

Removing Accidental First Choice Picks

Something that many peer reviewers and Butch mentioned was that they accidentally selected the first option without reading all the way, as the first option was auto-selected. Now, options are not pre-selected, making users have to confirm their choices. While playtester 2, Kevin, wished that it auto selects the first if you don’t have any other options, Butch mentioned that it is “it’s a bigger deal to avoid misfires / selecting before reading.”

Quality of Life

Another critique from a peer reviewer was that they wished there were “checkpoints” so that they could explore other options better. Adding checkpoints would require a lot of overhead to implement within the timespan, involving data persistence with both world space, but also dialogue choices. To build on replayability, I decided to add an auto skip button (code borrowed kindly from Lucas W.), where users can skip dialogue they already read until the next choice pops up.

I also added a transparency effect when the player moves beneath overlapping door and wall elements in response to prior playtest feedback. This change improves visual clarity and feedback, making it clear when the player is positioned beneath environmental elements rather than appearing incorrectly layered on top of them.

Lastly, I added a better indicator of time (morning, day, afternoon, night) through a better camera tint transition. While I had a background change in P2, the new version makes the transition more apparent. This works to immerse the player into feeling like they’re staying late at work or that time has actually passed by during the gameplay.

The Cursed Elevator Bug

An additional thing that I fixed was a major bug within my game in P2 where you could skip Bill/Charlotte’s story if you went up the elevator instead of talking to him. You could also repeatedly go up and elevator and get unlimited money by quitting your job over and over. Now, this is fixed!

Adding Remaining Art Pieces

The main adjustment here was to add the rest of the artwork, something I ran out of time to do in P2. This was highly effective towards the empathy and feeling of the game, as demonstrated in playtest 2, where Kevin really enjoyed viewing all the art. He noted that he couldn’t tell which ones were AI or real, questioning himself, rather than appreciating the art. This is a feeling I wanted to emphasize throughout the game, highlighting the larger theme of artists feeling diminished in the modern day.

Story Changes

One player said “I think it’s a bit too easy to get a BE right now, and the player actions that “lead to a BE” don’t seem related to core concepts like “artistic philosophy.” For example, selling the painting to the first buyer will most likely lead to a BE. But selling the painting to him doesn’t mean “the player abandoned their artistic ideals.” Later, I also weirdly got a BE due to lack of stamina, which prevented me from seeing the final ending. If some of the trivial choices could be reduced, and every choice that could lead to a different ending was “staked on the player’s values,” the game experience would be better.

Breaking this feedback up, there are a few grievances from P2 I needed to address.

1. Too easy to get a bad endings when its not your fault in terms of money.
   • How I addressed this: There is now a short scene in the beginning, the night before your popup, where you think about your financial situation. Now, people know that they are going to possibly fail if they sell to Steve, but first-time players won’t know if there are other interested buyers.

Caption: You get a warning about your upcoming bills so you’re no longer blindsided.

1. • If you pick Steve’s route (or Jessica’s route with some spendings here and there that decrease your amount below the threshold for bills), you are forced to have an uncomfortable conversation with your roommate, where you ask them to loan you money. They’ll agree as many times as you ask as long as you were nice initially and don’t try to make them give you $2000. If it’s asking for $1800 (rent cost) or more, then they will only give you $500.
2. 

Caption: Keira is a few bucks off of rent and forced to either quit being an artist or ask for a loan.

1. • Instead of Jessica just throwing a bunch of money at you (which felt unfair before), I decided to make her just give you the flat $2000 and added some income from smaller pieces at the beginning. This way, your money is still super tight, regardless of both choices. You also can’t just ask for your paycheck upfront, which felt too easy and a bit unprofessional.
2. Too easy to get a bad ending based on unhealthy decisions. 
   • I decided to bump up the counter from 4 to 5 (before you get the pass out ending).
   • I also added some feedback if your health was at 1 (on the brinks) so you can be more careful.
3. More thoughts about turning to AI usage. 
   • In Ron’s playtest, he wished there was more incentive to turn to AI. I decided to add some bedtime thoughts where all you can think about is taking the easy way out. This way, some players might want to try that route out.

Caption: Bedtime thoughts about AI usage in the art studio

Playtest 1: Ron

Link: Ron Playtest

Player Demographic: Ron is a recent new-grad from an east coast school, fitting within one of my player demographics of college-aged students who may be worried about the job market. Additionally, Ron is a computer science / AI major, meaning that he comes in with first-hand knowledge about generative AI.

Character Customization

During playtesting, Ron initially expressed confusion about who the player character was meant to be, indicating a gap in player identification early in the experience [4:40]. To address this, I implemented character customization, which was a backburner idea in P2. Adding customization affords players a stronger sense of embodiment and agency by allowing them to visually situate themselves within the narrative rather than inhabiting a predefined character. From a game design perspective, this strengthens identification and emotional investment, helping players more easily project their own experiences onto the protagonist. This adjustment added tons of expression fun to the game (tho took a bit to set up and draw)!

Typos

Ron caught a lot of typos, given that he is an editor. I promptly fixed all of the typos that he caught.

Notes: 

• [01:28–02:40] Onboarding & usability friction: Ron immediately noted UI and onboarding issues. Added after his playtest!

• [03:33–03:57] Character writing provokes reaction: Ron reacted negatively to the dad, saying “He’s kind of getting on my nerves… that is toxic,” indicating successful characterization that elicited an intentional emotional response.

• [06:32–06:46] Recognition of real artists: Upon seeing credited artwork, Ron remarked “These are all by Tiffany Tran… Oh, cool.”

• [12:55–13:24] Dialogue authenticity praised: Ron asked “Did you write this dialogue?” and noted that the AI-enthusiast character felt realistic and “predictable,” suggesting that the dialogue effectively captured real-world rhetoric around generative AI.

• [16:55–17:03] Discovery & deduction fun: While analyzing artwork details (“There’s only three fingers… I should have trusted my instincts”), Ron explicitly stated he liked “the sleuthing aspect,” confirming successful discovery and investigative play inspired by Ace Attorney–style mechanics.

• [18:27–18:35] Environmental authenticity: Ron noted that the studio dialogue “sounds like a legit studio,” reinforcing the credibility of the game’s professional setting.

• [34:29–34:38] Core aesthetic preference: Ron stated “I’m a sucker for top-down exploring things,” affirming that the spatial presentation and exploratory structure aligned with his tastes.

• [48:28–49:00] Strong holistic praise: Ron identified his favorite aspect as the game being “filled with original art” and appreciated the inclusion of AI examples, noting the blend of mechanics and comparing it positively to Ace Attorney.

• [49:20–50:07] Meaningful choice & agency feedback: Ron liked that AI usage decisions were “up to me,” but suggested deeper persistence systems (alignment-style tracking) where repeated AI-positive or AI-negative choices would unlock different dialogue and inner thoughts.

• [52:34–53:11] Systems design validated: Ron praised the decision to keep hunger and bills mostly out of sight, stating it felt realistic and that the end-of-day summary “added to the tone of the game,” validating the game’s systemic pacing.

• [55:55–56:30] Future design opportunities: Ron suggested expanding AI usage into more subtle or stealth-like systems (“a way to get away with it”), identifying a potential avenue for deeper systemic play while acknowledging the added development complexity. Not added due to time constraints and too much state tracking, though this sounds like a cool improvement.

Playtest 2: Kevin

Link: Kevin Playtest

Player Demographic: Kevin is a 30 year old venture capitalist employee. He likes drawing (low-stakes) in his free time, but doesn’t consider himself an artist. He likes gen-AI, but feels “iffy” about it being used for art.

Major Changes

No major changes resulted from Kevin’s playtest, as his feedback primarily confirmed the intended narrative and thematic delivery rather than revealing critical breakdowns. While Kevin noted that the experience felt somewhat “on rails” and suggested increased player autonomy, this linear structure was a deliberate design choice grounded in narrative-first game design. Constraining player agency helped preserve emotional pacing, thematic clarity, and authorial intent, ensuring that ethical questions around AI and artistic labor unfolded in a controlled, cohesive arc rather than being diluted by excessive branching. Additionally, implementing large-scale divergent paths would have significantly increased scope and production complexity, risking narrative inconsistency and mechanical imbalance for a student project with limited development time. Instead, the current design prioritizes narrative cohesion, tonal consistency, and focused player reflection over systemic freedom, with Kevin’s suggestions noted as valuable directions for future expansion rather than immediate implementation.

Notes: 

Playtest 3: Keira

Link: Keira Playtest

Player Demographic: Keira is a college senior who loves drawing (her art is in the game!). She is anti-gen AI and is on a path to be a software engineer. Keira is knowledgeable about AI concepts like neural networks and how gen-AI works.

Major Changes

No major changes resulted from Keira’s playtest, as her feedback largely validated the game’s emotional pacing and narrative focus. While she suggested minor quality-of-life improvements, these were not implemented to preserve intentional friction that supports narrative tension and player stress. Adding further conveniences or systems would have required rebalancing core mechanics and risked diluting the intended emotional experience, so her feedback was documented for future iterations rather than applied in the current scope.

Notes: 

• [01:41–02:16] Character customization & expression: Keira immediately engaged with customization, saying “Wait, this is so cute… I’m about to spend like three hours in the character generation,” indicating strong expression and fantasy fun as well as successful player embodiment.

• [04:34–05:15] Financial anxiety & systems feedback: Keira reacted strongly to the loss of money and recurring bills (“Not my $5,000… $900 a month for credit card debt. Damn.”), demonstrating effective ludonarrative alignment between financial mechanics and emotional tension.

• [07:44–08:02] AI ambiguity & thematic discovery: While viewing gallery artwork, Keira questioned whether pieces were AI-generated (“Wait, that’s super cute. Is this AI? I don’t know.”), confirming that the blurred authorship successfully prompted reflection and discovery.

• [10:37–10:58] Emotional engagement with characters: Keira expressed frustration with Landon (“Is there an option to deck Landon?”), showing strong emotional investment and successful character writing that provoked an intentional negative response.

• [14:46–15:00] Empathy & narrative resonance: Keira explicitly connected to the parental pressure storyline, stating “I totally get the whole parents thing,” reinforcing the effectiveness of empathetic design and player identification.

• [34:23–34:49] Overall positive feedback: In her post-playtest reflection, Keira praised the game’s art, storytelling, and flow (“The art is super cute… the storytelling went very well… overall very smooth gameplay”), validating improvements to narrative pacing and presentation.

• [35:43–36:11] Constructive feedback for future work: Keira expressed a desire for expanded exploration and additional self-sabotaging behaviors (“More world exploration… more ways to ruin my life… procrastination”), suggesting future opportunities to deepen systemic agency and discovery fun.

• [36:28–36:55] Thematic success: Keira noted that the game sparked meaningful conversations about AI and art, stating that expression is “something uniquely human,” confirming the game’s broader thematic impact beyond moment-to-moment gameplay.

Reflecting on Game Improvements

During P4, I set out specific categories to improve on. With these goals in mind, below is a reflection about the work accomplished and the affordances of new features.

Ludonarrative Consonance

A major focus of iteration was aligning player actions with the narrative themes of artistic compromise, financial precarity, and value-driven decision-making. Earlier versions of the game allowed players to reach bad endings through outcomes that felt disconnected from their artistic philosophy, such as being financially punished without sufficient narrative foreshadowing. By adding an early reflective scene about upcoming bills and restructuring how income and expenses are distributed, player failure now feels more clearly tied to their values and choices rather than arbitrary systems. Similarly, adjustments to AI-related bedtime thoughts were designed to narratively motivate temptation toward automation, ensuring that turning to AI feels like a meaningful thematic decision rather than a purely mechanical shortcut.

Technical Implementation

Several improvements addressed technical flaws that previously disrupted the player experience. Critical bugs—such as the elevator exploit that allowed players to skip story content or accumulate infinite money—were resolved to preserve narrative progression and system integrity. Additionally, quality-of-life features such as dialogue auto-skip and visual transparency under overlapping environment elements were implemented to reduce friction during play. These changes improve spatial feedback and interaction flow, allowing players to focus on narrative and decision-making without being distracted by technical inconsistencies or visual ambiguity.

Narrative & Emotional Engagement

Narrative revisions were made to better support emotional pacing and player empathy. Expanding the amount of artwork and intentionally blending AI-generated and human-made pieces heightened thematic tension around authorship and creative worth. Moments such as the uncomfortable conversation with the roommate over rent were designed to externalize internal anxiety, reinforcing the emotional stakes of financial instability. Character customization was added to strengthen player identification and embodiment, helping players more easily project themselves into the protagonist’s situation and increasing emotional investment in narrative outcomes.

A future improvement would be to incorporate Kevin’s feedback about feeling “on rails” by expanding narrative branching and systemic reactivity. While this feedback is valuable, addressing it meaningfully would require restructuring core dialogue logic and state-tracking systems to support more divergent paths rather than surface-level choice variation. Given the project timeline and the risk of introducing narrative inconsistencies or bugs late in development, I prioritized depth and clarity within the existing structure over adding breadth through additional branches. This decision allowed the narrative to remain emotionally cohesive while leaving room for future expansion.

Fun & Player Experience

Based on playtesting and user feedback, Blank Canvas most strongly supports Narrative, Fantasy, Expression, Discovery, and Submission fun as defined by the MDA framework. Narrative fun drives the experience through emotionally focused storytelling and value-based decision-making. Fantasy and Expression fun emerge through role-play and character customization, allowing players to project themselves into the protagonist and define their artistic identity. Discovery fun is supported through gradual narrative reveals and consequence-driven progression, while Submission fun is present through the game’s low-friction, contemplative structure, encouraging reflective engagement over mechanical challenge.