Week 11: Iterating on Feedback

Last Saturday, we welcomed guests from different backgrounds to play our game at ETC’s Spring Semester Playtest Day. During Playtest Day, groups of 1-4 come to our project room, learn about the team/project, and play the game for 20 minutes. We asked for playtesters in highschool or older that have some understanding of AI and Rap as we thought they would understand the project better.
We took the feedback from Playtest Day and used it to iterate on ur game design and playtest with the Ojas Sawant (ETC ’15), an AI specialist. Moving into week 12, we will try to implement our new designs before Softs, where we will present our progress to the rest of the faculty.


Playtest Day Feedback:
What’s working:
- Love the art style, UI, and sound
- Excited about AI in the game
- Most players were really engaged with Fly Gull and reacted strongly to her personality
- On average, players were able to find 1-2 weaknesses
- “Very cool”
- Want to play more or with other people- schools
- Almost all had fun, were laughing, and tried to win
General Observations:
- Most people think they’re being scored on the rhymes. We should lean into that
- Most players lost
- Players interacted with Fly Gull more freely when they didn’t think she was AI
- Most players didn’t interact with the audience
- Some players didn’t understand what it means to boast or diss
- When players didn’t know all the weaknesses, they resorted to insults about Fly Gull
- It was common for players to think that Fly Gull’s ego was weakness and that one of the audiences’ preferences was Fly Gull
- Playing in groups worked well
- Want to hear the rap
Design Issues:
- More instructions. Most players must be reminded what the objective is and how to use the rap book.
- Players aren’t always connecting the interactables/rap book suggestions with the larger game goal.
- They don’t know how to use the map.
- Most players don’t fully understand how they are being scored but intuitively understand what makes rap cool.
- Players understood that the red text when they found a weakness was important but not why it is important
Playtesting with Ojas:
Ojas Sawant is an ETC Almun and works at a Stat-up that is revolutionizing software and storytelling with AI. Since our game also combines narrative and GenAI, we wanted to get some feedback from him as well.
He agreed that we need more direct and indirect control in our level design right now. Also, we proposed interacting on the Audience profiles to have Hater, Neutral, and Fan characters that would have different personalities/relationships with Fly Gull.

Solving our Design Problems:
Combining the feedback from playtest day and our playtest with Ojas, we created a list of tasks to complete before Soft in order to address our design issues.
Adding animations, sound effects, and more UI elements are meant to incorporate more instructions/directions to the build. The scoring iterations will help make it easier to understand how to win the game. We’re adding a Creavity score that will be a combination of clarity, consistency and cleverness.
- Rap Book Iterations:
- Add UI Animation
- Add Sound Effect
- Interactabe Iterations:
- Update icons
- Update interactable dialogue
- Battleground Iterations:
- Reconfigure Camera
- Add Character Animations
- Add objective board
- Add “side quest” interactable update
- Redesign map layout
- Shorten AI responses
- Add Fly Gull angry UI
- Fix colliders
- Rap Battle Iterations:
- Add feedback sound effects
- Add feedback ui icons based in highest category
- Intro Iterations:
- Update subway map icon
- Add game objective
- Scoring Iterations:
- Create separate Fly Gull doc
- Add Creativity Category -> clarity, consistency, and cleverness
We’re also finishing up tasks that have been on our backlog for a while now, like walking/idle animations to add style to the environment. In addition, we are going to restructure the level of the platform since most players didn’t interact with the audience when they were in the corner of the map.

Next Steps:
Next week is our Softs presentation to the faculty. We will try to implement our solutions to our design issues and test if they are landing the way we would like them to.