Week 4 Blog

We made a few key design decisions this week:

1. Maximum of 3 players
2. Collaborative experience
3. Experience divided into stages
4. No stereoscopic glasses

The first two decisions go hand-in-hand. We feel strongly about making the experience collaborative. The Music Experience I wrote about in my Week 2 Blog definitely influenced my thinking. People enjoy being part of a group, and that feeling can become even more meaningful when you feel connected and in sync with the people around you. “Connection” is a keyword that has come up repeatedly in our team discussions, and we believe that social connection is an important part of wellness.

As for the decision to support three players, we arrived at that number based on several considerations. First, we want to keep throughput as high as possible, especially since this experience will likely be part of Fall Festival this year. Second, we want to keep the technical scope manageable for our solo developer. Finally, the Cavern has three projectors and three Orbbec cameras, so three players feels like a natural and practical upper limit for us to work within.

The third decision — to divide the experience into stages — may not be final yet. We are waiting for our two designers to present their detailed concept next Monday, so this is something we may revisit once we see how their ideas take shape.

It was helpful to make these decisions because they give us concrete constraints to design around. Rather than continuing to explore an entirely open-ended set of possibilities, we now have a clearer framework for what the experience can be. I think this week marks an important transition for the project: we are moving out of the blue-sky phase and into more specific, intentional design.

For this week, we tested some new tech developments with the Orbecc trackers and were able to configure the Orbecc to track more than 1 person. Hopefully I will be able to push this to the toolkit once it is documented and tested properly to also support other ETC students!

We got in touch with Terri Lim (from the prior Spelunx toolkit team) regarding tying in 3 cameras at once, but did not get it working just yet. Seems the problem is that it gives you back a point cloud but does not merge the data so we would have to merge things on our own.

We also messed around with adding ‘hands’ that are  to be able to turn the skeleton off and just have the hands show in the world so players know where they are particle systems (Although the world position of them is slightly misconfigured and broken).

Our team hosted a Sound Bath for the larger ETC community. The table below is a summary of our key findings.

Here is the complete report:

Conversation with Sam Polglase (Subject Matter Expert in CAVE Systems)

To gain additional expert perspectives on the CAVE system, Christina reached out to Stef Harbuck, VP of Production at Schell Games and leader of the ETC’s Producer’s Roundtable sessions. Stef kindly connected us with Sam Polglase, Principal Engineering Manager at Schell Games.

Most recently, Sam worked on Superman Experience: Defenders Unite, an immersive experience that opened to the public on April 18, 2026 (Superman Day). Located at Stage 5 on the Warner Bros. studio lot in Burbank, California, the experience uses highly sophisticated, commercial-grade computer vision technology, including Captury Live. While the scale and technology of the Superman Experience differ significantly from our project, our conversation with Sam provided valuable insights into the practical challenges of tracking, player interaction, and technology selection.

Here are the key takeaways from our conversation:

• Stereoscopic Vision Limitations: Stereoscopic vision works well when players remain relatively still, but movement can introduce visual distortion. Synchronizing 3D glasses can also be challenging, since stereoscopic glasses are relatively old technologies.
• Tracking Challenges: Orbbec cameras have proven unstable in their experience, and there is no native solution for fusing body tracking data from multiple cameras. For reliable tracking of players and physical props, Vive trackers paired with base stations (Lighthouses) have been more successful.
• Lighting for Stable Tracking: Consistent ambient lighting is important for reliable tracking. Fully dark environments can negatively affect visible-light tracking and introduce jitter.
• Gameplay Guidance: Clear, simple instructions and intuitive player paths help improve cooperation and reduce confusion, particularly in multiplayer experiences.
• Aim Assist: Body tracking is inherently imprecise, making aim assist valuable for compensating for tracking inaccuracies. The goal is to subtly correct player actions without making the assistance feel noticeable or excessive.
• Technology Strategy: Proven, reliable technology should generally be prioritized over experimental tools when stability and delivery are critical. Rapid prototyping can help validate ideas early and reduce technical risk before committing to a particular solution.

One important consideration is that their exploration of markerless tracking was driven by the client’s specific requirements. Sam noted that when other sensors and tracking systems are available, their team has had success with Vive trackers. This means that we are not necessarily limited to markerless tracking for our own project.

Moving forward, we will first focus on experience design and use those design requirements to determine which technologies are actually necessary. Rather than selecting a tracking solution first and designing the experience around its limitations, we can let the desired player experience guide our technology decisions. This will be our focus next week.