1. Project Background
In early 2025, we received a special project from Dali Yangbi — to create a VR immersive narrative experience based on Yangbi walnut culture and Yi ethnic customs.
This was a "have it all" challenge: visitors needed to experience Yangbi's history and culture in a virtual world, while the script interaction design turned visitors into protagonists who deeply understood local culture through puzzle-solving and exploration.
The project officially launched in February 2025 and completed delivery in June 2025, eventually achieving single-pass acceptance and earning high recognition from the client.
Today, I want to share the complete development process of this project, hoping to provide reference for industry colleagues while giving ourselves a milestone review.
2. Project Initiation: Understanding Client Expectations
At the project kickoff meeting, the leadership from Yangbi County Bureau of Culture and Tourism raised several core demands:
First, cultural expression must be "authentic." They didn't want to see a "tech show" wearing a VR disguise, but rather an immersive experience that truly conveyed Yangbi walnut culture and Yi ethnic customs. This meant we couldn't work behind closed doors — we had to go deep into the local area and understand it.
Second, the script must be playable. Previous "culture + technology" projects often fell into two extremes — either too academic, making people drowsy, or too entertainment-focused, with cultural elements reduced to decorations. The client wanted a balance, allowing participants to naturally absorb cultural immersion through "playing."
Third, interactions must be smooth, experiences must be complete. As a fixed deployment project within the scenic area, it would face visitors of all ages and technology acceptance levels. Operations couldn't be too complex, yet needed to be rich enough to prevent boredom within 10 minutes.
Fourth, independent operation after delivery. The client's team had no dedicated technical staff, so post-project maintenance had to be simple to learn. This meant we had to deliver not just a product, but a complete training system and after-sales support.
These four demands set the tone for the entire project from the start: Technology is the means, culture is the core, experience is the standard, and delivery is the responsibility.
3. Requirements Alignment: A Journey from Vague to Clear
Initial requirements communication wasn't smooth.
The client's first batch of requirements documents read more like a "wish list" — they knew what kind of experience they wanted, but struggled to describe it in technical language. For example, they mentioned "wanting players to walk into Yi ethnic homes and chat with locals," but hadn't decided whether it should be voice dialogue, text interaction, or preset animations.
To solve this, we adopted a "Requirements Workshop" approach:
Round 1: Current State Mapping. We invited the client's cultural experts and advisors to give us a full day of training on Yangbi's history, Yi ethnic customs, and the symbolic meaning of walnut culture. This day transformed us from "technical contractors" into "students of culture."
Round 2: Prototype Visualization. Before formal development, we built a rough 3D scene creating a 5-minute experience demo, letting client leadership put on a VR headset and personally experience what "VR script experience" meant. Many vague descriptions became clear at this step — for instance, "walking into Yi ethnic homes" was ultimately defined as an NPC-guided exploratory dialogue where players needed to interact with NPCs through choices to obtain clues for the next step.
Round 3: Feature List Confirmation. Based on demo feedback, we produced a detailed functional specification covering scene quantities, NPC dialogue branches, prop systems, achievement systems, and every detail. Both parties signed off before entering the development phase.
In retrospect, while these three rounds of requirements alignment took about three weeks, they saved at least half the rework costs in subsequent development.
4. Field Research: The Road Under Your Feet Matters More Than PPTs
A VR cultural tourism project will fail if developed only from office imagination.
In March 2025, our project team of five went to Yangbi for a week-long field research trip.
Research included:
1. Geographic and Architectural Field Survey
We visited core scene locations including the Yangbi Ancient City ruins, traditional Yi ethnic villages, and the Walnut God Tree, conducting extensive imaging with drones and panoramic cameras. A local Yi elder pointing to remaining rammed-earth walls told us, "This used to be the village gate, that was the place for heaven-worship ceremonies" — this oral history information was detail no document could record.
2. Field Collection of Cultural Symbols
What meanings do Yi ethnic costume patterns hold? What role does Yangbi walnut play in folk stories? What is the ritual process of the Yi Torch Festival? We couldn't find answers to these questions in libraries — we could only ask, note by note, by the firesides of elders in villages and fields.
3. Target User Behavior Observation
We walked the existing tourism routes in Yangbi as "tourists," observing behavior patterns and stay preferences of visitors of different ages and regions. These observations directly influenced the scene transition rhythm and difficulty curve design in our VR script.
On the day research ended, we brought back over 8GB of video footage, a full notebook of research notes, and a profound understanding: Developing cultural VR projects isn't "technology + materials" simple assembly, but "technology to recreate an authentic cultural perception."
5. Story Script Creation: Bringing Culture to Life
Based on research findings, we began script creation.
What are the core IPs of Yangbi culture? We distilled two keywords: "Walnut" and "Fire." The former represents Yangbi's material civilization — Yangbi is China's Walnut Hometown, walnuts are the life source local people have depended on for generations; the latter represents Yangbi's spiritual beliefs — Yi ethnic fire worship, giving rise to the grand Torch Festival ceremony.
Based on these two core elements, we designed a time-travel themed script:
The player takes on the role of a "Time Traveler," accidentally transported to the Yangbi Ancient City a hundred years ago. To return to reality, the player must complete a series of challenges related to walnut culture and Torch Festival rituals, finally解开谜题 under the guidance of an elderly Yi elder, completing the mission.
This script structure has several considerations:
One is immersion. The "time travel" setting transforms the player from "observer" to "participant" with clear motivation and easier emotional investment.
Two is clear cultural anchors. Every challenge is bound to real Yangbi cultural scenes — completing the game is a deep cultural journey.
Three is adjustable difficulty. We designed three branch endings and different achievement badges, satisfying both hardcore explorers' desire and casual players' sense of accomplishment.
After completing the script, we invited Yangbi local cultural advisors for review, ensuring no cultural misinterpretations or historical errors. After reviewing the full text, an elder advisor said: "This story makes local Yangbi youth feel亲切, and outsiders feel like visiting."
This sentence became the standard for all our subsequent work.
6. Concept Art Design: Every Frame is a Cultural Annotation
With the script confirmed, concept art design began.
The biggest challenge at this stage was: How to recreate historical authenticity in VR scenes while maintaining comfortable visual aesthetics?
The Yangbi Ancient City dates back several centuries. If fully restored according to historical records, the imagery might appear "grayish," lacking the visual impact VR experiences require; but excessive beautification would lose historical weight.
Our solution: "Realistic Foundation + Stylized Treatment."
Specifically, the overall color palette and spatial structure were realistically restored based on photos from field collection, while material rendering and lighting atmosphere received artistic treatment — for instance, in Yi ethnic village scenes, we enhanced the warm tones of "fireside light" during twilight, giving the entire space both historical sense and warmth.
We also carefully handled NPC character design:
- Yi elder appearances referenced elders interviewed during research, with facial textures and costume patterns striving for authenticity
- Young female NPCs incorporated the most representative "Rhododendron (Mapie)" pattern elements from Yangbi Yi ethnic costumes
- Through motion capture and expression collection, we ensured NPC body language was natural and smooth
In the end, we produced over 200 scene concept artworks, 50+ 3D character models, and a complete set of UI and interaction visual specifications. These assets served not only the current project but also provided reusable design assets for Yangbi's other cultural tourism experience products.
7. Interactive Program Development: Technology Serves Experience
Entering the development phase, the real攻坚期 began.
Challenge 1: VR Device Adaptation
Target deployment environments included Pico 4, Meta Quest 2, and other mainstream VR headsets. Different devices have varying resolutions, FOV angles, and interaction controller layouts — how to ensure smooth experiences across devices with the same program?
Our approach: Using Unity engine with XR Interaction Toolkit as the interaction framework, implementing a modular controller mapping scheme for single-code multi-device adaptation. Simultaneously, we conducted specific optimization for each target device, ensuring stable frame rates above 90FPS to avoid motion sickness.
Challenge 2: Story Branch Management
The core VR script experience lies in "choices." Every player choice affects subsequent story direction and ending branches. Using traditional linear development, each branch would require independent development, causing workload to grow exponentially.
We designed a finite state machine (FSM) story engine:
- Main storyline defines core state nodes
- Each node hangs multiple branch conditions
- Branch conditions are triggered by player interactions (dialogue choices, prop usage, scene exploration, etc.)
- Final state weight calculation determines which ending the player reaches
This engine's advantage: Content teams can independently edit story branches without development intervention. This greatly improved late-stage script iteration efficiency and provided possibilities for client content updates.
Challenge 3: Performance Optimization
VR experiences have extremely demanding performance requirements. Overly large scenes, overly refined models, overly complex rendering all cause device overheating, frame rate drops, and experience crashes.
We optimized from several dimensions:
- LOD Models: Dynamically switch model precision based on player-object distance
- Occlusion Culling: Only render content within the player's current field of view
- Dynamic Resolution Adjustment: Appropriately reduce rendering resolution in complex scenes to prioritize frame rate
- Asynchronous Loading: Large scene transitions use fade transitions with background asynchronous loading to avoid stuttering
After three rounds of specialized optimization testing, the final version maintained stable average frame rates above 92FPS on target devices with no significant stuttering.
8. Delivery and Acceptance: The Confidence to Pass on First Try
In June 2025, the project entered the delivery and acceptance phase.
In the final two weeks before delivery, we conducted three rounds of internal testing and one client pre-acceptance review.
Round 1 Internal Testing focused on functional completeness: Can all story branches trigger normally? Does the prop system function properly? Does the achievement system record correctly?
Round 2 Internal Testing focused on experience smoothness: Are scene transitions smooth? Do NPC dialogues feel natural? Are VR interactions natural enough?
Round 3 Internal Testing focused on edge cases: Does it run normally offline? Can data from abnormal exits be saved correctly? Are multi-language versions without omissions?
After three rounds of testing, we fixed 47 bugs, including 3 high-priority issues — if these had reached the client acceptance stage, they could have caused acceptance delays.
On pre-acceptance day, the client team brought 8 people, including Bureau of Culture and Tourism leadership, scenic area operations managers, and two specially invited cultural advisors. They put on headsets and experienced the complete VR script.
During acceptance feedback, a leader said: "Better than I imagined. The scenes are very realistic, the script is interesting, and I can tell you genuinely researched Yangbi culture."
Final acceptance was passed in one attempt. On the acceptance report, the client wrote: "Project meets design requirements, agrees to delivery."
9. On-Site Training: Passing the Baton
Delivery isn't the end — operations are.
To enable the client team to independently operate and maintain this system, we designed a 3-day on-site training program:
Day 1: System Operations Training. Including equipment on/off, headset wearing and cleaning, system startup and exit, daily inspection procedures. Every operational step had a screenshot-based operations manual, ensuring "even zero-basis personnel can understand."
Day 2: Content Management Training. Including script dialogue modification, NPC dialogue configuration, and achievement badge management backend usage. This content had slightly higher technical barriers, so we recorded complete operation videos for client review later.
Day 3: Emergency Response Drill. Including common fault troubleshooting and handling, standard procedures for contacting after-sales support, and system restoration operations in emergencies.
After training, we left complete "Operations and Maintenance Manual" and "Emergency Fault Guide," establishing direct communication channels with the client operations manager to ensure prompt response to any subsequent issues.
10. Post-Delivery Maintenance: Earning Trust Through Service
Nearly a year has passed since delivery, and our cooperation with Yangbi hasn't ended.
During this year, we handled several after-sales requests:
- During the scenic area's summer peak electricity usage period, equipment power supply anomalies occurred once — we remotely assisted in troubleshooting line issues
- Before National Day Golden Week, the client wanted to temporarily add a "festival-limited script" — we completed custom development within one week
- During equipment firmware upgrades, we proactively sent compatibility test reports and upgrade guides
These "extra" services weren't written in the contract. But it's these details that make the client feel: We're not just doing a project — we're establishing a long-term cooperative relationship.
In early 2026, the client proactively discussed a second cooperative project with us — construction of the Yangbi Smart Scenic Area Digital Twin Platform. This project has entered the requirements research phase.
11. Closing Thoughts
Reviewing the entire Yangbi VR script project, I want to share several insights:
First, the core barrier for cultural VR projects isn't technology, but understanding of culture. Technology can be learned and outsourced, but cultural perception and expression require immersive investment. Time spent on research and scripting ultimately transforms into "sincerity" felt by the client and users.
Second, sufficient early communication saves most late-stage troubles. The three weeks of requirements alignment seemed "slow," but it meant subsequent development barely went astray. Rework costs are always much higher than prevention costs.
Third, passing acceptance on first try isn't luck — it's accumulation of countless self-checks. Three rounds of internal testing, 47 bug fixes, serious pre-acceptance handling — these "backstage works" determined the "onstage" performance.
Fourth, delivery is only the beginning of cooperation. Good service creates new opportunities. Reputation isn't spread through promotion — it's built piece by piece.
We thank Yangbi County Bureau of Culture and Tourism for their trust, and every team member for their dedication.
We believe good technology lets culture be seen, and good content gives technology warmth.
We'll continue walking this path.
If you have ideas for cultural + technology integration projects, welcome to exchange with us. Perhaps the next story is one we write together.
Project Information
- Project Name: Dali Yangbi Cultural VR Script Program
- Delivery Time: June 2025
- Delivered By: Yunnan Yunguan Digital Technology Co., Ltd.
- Technology Platform: Unity 3D + XR Interaction Toolkit
- Deployment Devices: Pico 4 / Meta Quest 2