CubeVi is building a desktop-based spatial computing ecosystem designed for real-world applications.

We are looking to collaborate with developers to explore how spatial displays—without the need for headsets—can transform gaming, creation, and interaction experiences, and to bring these ideas into reality through applications and live demonstrations.

Enabling two complementary directions:

• Bringing spatial interaction into PC experiences
• Bringing desktop precision input (Keypad and Mouse) into XR applications

The program is application-based. Developers are invited to submit their previous work, proposed adaptation ideas, and development timeline. Selected participants will gain access to PLAY at a developer price ($699, MSRP $1399) and benefit from a structured incentive system designed to support and accelerate development. 

Once an application is successfully completed and meets our platform’s publishing standards, the full device cost will be reimbursed.
Projects that demonstrate strong experience quality or exhibition value will receive additional rewards starting from $500+.
Exceptional projects may receive $1000–2000 in funding, along with promotional support.
For developers with mature projects or strong demonstration capabilities, we may also provide devices free of charge.

(All reimbursements and rewards are contingent upon meeting CubeVi’s quality and showcase standards.)

Apply Now!

Build for CubeVi PLAY

Spatial Desktop Gaming: Design Constraints and Adaptation Strategy

1. For PC Games & Applications

Extend existing PC experiences into spatial environments with minimal changes.

Level 1 — OpenXR-based Spatial Rendering

Developers can build using standard OpenXR workflows. In engines like Unity or Unreal Engine, this typically means simply enabling OpenXR support with minimal additional work.

CubeVi integrates with a Monado-based runtime, and Quest can be used as a tracking source, enabling controller-based spatial interaction.

This pathway is well-suited for bringing existing PC games or applications into a spatial display context without requiring major interaction redesign.

Level 2 — Hybrid Interaction Extension

PC applications can be further enhanced with spatial interaction while retaining familiar keyboard and mouse controls.

This hybrid model combines traditional keyboard/mouse input with spatial hand interaction. Users continue to rely on keyboard and mouse for movement and core control, while using one or both hands for spatial actions such as pointing, triggering, or manipulating objects.

Rather than replacing traditional input, this approach extends it with a spatial dimension. By mapping a small set of essential gestures to key and mouse actions, it maintains simplicity, stability, and ease of integration.

A common pattern in this hybrid model is asymmetric interaction:

Users rely on keyboard (e.g., WASD) for movement and navigation, while using one hand for spatial input such as ray casting, pointing, or object manipulation.

This separation keeps precision where it matters, while enabling intuitive spatial interaction without overwhelming the user.

This pattern works well both for Level 2 PC adaptations and for XR applications undergoing input rebalancing.

2. For XR Games & Applications

Adapt XR applications to a desktop-based spatial display environment.

Field of View & Interaction Model Adaptation

Compared to head-mounted displays, spatial displays provide a more focused field of view. This requires not only adjustments to camera framing and scene composition, but also a broader rethinking of interaction and gameplay logic to fit a desktop-based spatial context.

The experience is closer to observing a world through a window rather than being fully immersed inside it. As a result, applications should be designed around this “windowed spatial” paradigm—where scale, depth, and focal areas are more controlled and centered within a stable viewing frame.

Interaction also shifts accordingly. Unlike in VR, where users can directly reach into the environment, interaction on spatial displays is indirect. Users operate virtual hands or controllers mapped into the scene, rather than physically entering it. This “remote mapping” model enables precise and comfortable interaction without requiring large physical movement, and is better suited for longer, everyday use.

Input Rebalancing (Hybrid Model)

In practice, this enables a complementary shift: XR applications can incorporate keyboard and mouse input to reduce reliance on controller joysticks and improve precision and usability.

Core movement and precise control should be handled by keyboard and mouse, while spatial interaction (e.g., pointing, triggering, grabbing) remains mapped to hands.

Hardware Setup

Current Stage Hardware Setup

As an initial step, developers can use CubeVi PLAY together with XR headsets (e.g., Quest or Pico) and their controllers as a proxy for hand interaction.

In this setup:

• The headset is used primarily for positional tracking
• Controllers are used for basic spatial input such as trigger and grab
• Joystick-based movement is intentionally avoided and replaced by keyboard and mouse

This allows developers to begin building and testing hybrid interaction workflows using existing, widely available hardware.

Next Stage (Hand Interaction Hardware)

In the next stage (expected around June), CubeVi will provide a dedicated hand-tracking USB peripheral that attaches to the display.

This device enables direct hand-based interaction and is designed to replace controller-based input in the hybrid interaction model.

With this setup:

• Spatial gestures replace controller inputs
• Interaction becomes more natural and lightweight
• The full hybrid interaction model (keyboard/mouse + hand gestures) can be realized as intended

Apply Now!

Our Vision

We believe spatial computing will not be confined to headsets alone.
Desktop-based spatial experiences will become an essential part of this paradigm.

CubeVi is not just building a display.
We are building a platform for the next generation of interaction and content.