Real world space

Real world space
???
Information
Type	Room-scale
Required Devices	None
Description	Users move in real world spaces.
Pros	Very realistic and immersive
Cons	Cannot function with large movements, Spaces are often limiting, spaces have to be designed for physical movement
Creator	???
Examples	Tea For God

Real world space is a Locomotion Method. The user physically walks and turns inside a tracked area, and the headset maps those movements onto the virtual viewpoint. Researchers usually call this "real walking" or "natural walking". In most systems, real-world movement is mapped one-to-one onto the motion of the virtual camera. Because no artificial movement is added, the body's own motion cues match what the user sees. Comparative studies have found real walking easier and more natural than walking-in-place or flying, and better for navigation than button-controlled movement.

The main drawback is that the virtual space a user can walk through is limited to the size of the tracked physical space. Consumer Room-scale VR systems, boundary systems such as Chaperone and the Guardian System, and research techniques such as redirected walking and impossible spaces were all developed to work within, or stretch, that limit.

Reviewed 27 September 2026. New text checked against the six cited research papers (DOIs confirmed) and the cited Road to VR, UploadVR, New Atlas, HTC, Meta and Steam pages. About review dates.

Additional Information

For standing / room scale VR only. Users move in real world spaces. This limits the distance of users' movement because the tracked / available spaces are often quite small. It can be used in conjunction with teleportation to extend the movable space

How it works

The headset, and usually the hands or controllers, are tracked in position and orientation (6 degrees of freedom). Each physical step or turn produces the same change of viewpoint in the virtual scene. Steinicke and colleagues describe this as the usual case: "movements in the real world are mapped one-to-one to virtual camera motions". The real proprioceptive sensations of walking are matched by the synthetic visual and audio feedback, a pairing that Usoh and colleagues linked to a stronger sense of presence.

The available virtual area is bounded by the tracked physical area. In the words of Razzaque, Kohn and Whitton, a user who walks beyond it "loses tracking and/or runs into a wall of the lab". In their 2018 review of natural walking in VR, Nilsson, Serafin, Steinicke and Nordahl named two core challenges. The first is letting users walk without constraint through virtual worlds larger than the tracked space. The second is giving users suitable multisensory feedback as they interact with the virtual environment.

History

Research systems

Usoh and colleagues wrote in 1999 that the most common way to move in immersive virtual environments was "local walking limited by tracker range, combined with button-controlled flying". Moving to a different virtual location required flying. Their study at the University of North Carolina at Chapel Hill used a custom optical tracker. It worked over roughly 10 m by 4 m, with two optical sensors viewing blinking infrared LEDs in the ceiling tiles. The tracked space was divided into a training area and an experimental area, each 5 by 4 meters. The headset was still tethered, so an experimenter walked behind each participant to handle the cables and prevent collisions. The authors wrote that real walking through their visual-cliff scene gave "a strikingly compelling virtual experience". They named wireless links to the user as the improvement their system most needed.

Consumer room-scale VR

Road to VR's April 2016 review of the HTC Vive noted that "Room-scale in SteamVR is defined as a space over 2 x 1.5 meters (6.5 x 5 feet)". SteamVR's Chaperone system shows a transparent grid at the edges of the play space. The grid is "invisible most of the time" and fades in when the user's head or hands get near it. The Vive's front-facing camera can also show a view of the real room. HTC's support documentation gives the same 2 m x 1.5 m minimum for a room-scale setup with either generation of base station. It says there is no minimum for seated and standing use. With Base Station 1.0, movement is supported within a diagonal of up to 5 m. With Base Station 2.0, the maximum area is 7 m x 7 m.

Oculus confirmed its own boundary system, Guardian, in September 2016, for inclusion in PC SDK 1.8 alongside the Oculus Touch controllers. Developers could restyle the boundary to fit their game; in Dead and Buried it appeared as barbed wire. In December 2016 Road to VR compared the published dimensions. The Vive setup guide allowed base stations up to 16.3 ft apart diagonally, for a square play area 11.5 ft on a side. The Oculus Rift three-camera room-scale configuration, which Oculus still called "experimental", gave a play area of 8.2 ft per side.

Inside-out tracking and larger spaces

With inside-out tracking, the tracked area no longer depends on how far external sensors can see. At Oculus Connect 5 in September 2018, Oculus said the Oculus Insight tracking in the Oculus Quest went "beyond roomscale". It showed an "arena-scale" multiplayer version of Dead and Buried and announced a multi-room Guardian that recognizes which room the user is in. UploadVR described the demo as a 4,000 square foot arena with two teams of three players on standalone Quest headsets. It used no backpacks or specialized external tracking hardware. In June 2021 Facebook told UploadVR that from Quest system software v30, users should be able to draw Guardian boundaries up to 15 meters long in either direction, up from a previous maximum of 10 meters. Meta's current help documentation says the Roomscale boundary setting on Meta Quest headsets needs a clear space of at least 6.5 feet by 6.5 feet (2 meters by 2 meters). In smaller spaces the headset can be used in Stationary Mode, sitting or standing in one place.

Research

In the 1999 study by Usoh and colleagues, 33 naive participants were split into three groups of 11: real walkers, virtual walkers (walking-in-place) and flyers. Real walking was "significantly better than both virtual walking and flying in ease (simplicity, straightforwardness, naturalness)". Reported presence was higher for real walkers than for virtual walkers, but that difference was statistically significant only in some models. The largest difference in presence was between flyers and both groups of walkers.

Ruddle and Lessels tested a navigational search task in a room-sized virtual space. Participants who physically walked while wearing a head-mounted display performed 90% of trials perfectly, comparable to people doing an equivalent task in the real world. Participants who turned physically but pressed a button to move forward, and those using a desktop display, performed fewer than 50% of trials perfectly. The authors concluded that both translational and rotational body-based information are needed to keep track of one's position. They also observed that walkers tended to avoid virtual obstacles even though there was no collision detection.

The Nilsson review groups walking techniques into three general categories: repositioning systems, locomotion based on proxy gestures (such as walking-in-place), and redirected walking. Walking-in-place was one of the conditions in the Usoh study, where steps were detected from tracked head motion by a neural network.

Problems

The main problem is space. The virtual environment cannot be larger than the tracked area unless another technique is added. Razzaque and colleagues described this as "a serious limitation of real walking". Home users also have to clear an area of furniture. Meta recommends at least 2 m by 2 m for Roomscale, and HTC gives 2 m by 1.5 m as the room-scale minimum for the Vive family. Cables are another hazard with tethered headsets. In the 1999 UNC study, an experimenter followed each participant to keep them from snagging or tripping on cables or colliding with real obstacles in the lab.

Solutions

Boundary systems

Boundary systems do not make the space larger. They warn the user before they reach a wall or obstacle. SteamVR's Chaperone and Oculus Guardian both draw the edge of the play area inside the headset when the user gets close. In mixed reality apps the user can see the room through passthrough. From version 66 of the Meta XR Core SDK, any developer can disable the boundary, but only while passthrough is being rendered. Fully immersive VR still requires it.

Redirected walking

Redirected walking was introduced by Razzaque, Kohn and Whitton at Eurographics 2001. It "interactively and imperceptibly" rotates the virtual scene around the user, so that the user keeps walking toward the farthest wall of the lab without noticing. A pilot study suggested that it let people walk through virtual environments larger than the lab and "does not induce appreciable simulator sickness". In 2010 Steinicke and colleagues measured how far such gains can go before users notice. Users could be turned physically about 49% more or 20% less than the perceived virtual rotation. Distances could be scaled down by 14% or up by 26%. Users could also be steered along a circular arc with a radius greater than 22 m while believing they were walking straight.

Impossible spaces

Suma and colleagues proposed "impossible spaces" in 2012. These are virtual interiors with self-overlapping layouts, so that a comparatively large building fits in a smaller physical area. In their experiments, small virtual rooms could overlap by as much as 56% before users detected the overlap. Larger rooms that filled the 9.14 m x 9.14 m workspace could overlap by up to 31%. The commercial game Tea For God by void room, released on Steam on 19 July 2023, uses the same idea. Its store page says the world "is composed of spaces that overlap each other making it possible to travel through a big world, while never leaving your play-area". Players move only by walking, with no teleporting. The store page gives 1.8 m x 1.2 m as the smallest space the game handles; below that, it scales the world horizontally, which brings the minimum down to 90 cm x 60 cm.

Larger tracked areas

Another approach is to make the physical area much larger. Location-based virtual reality venues do this. Zero Latency, which opened in Melbourne, Australia in 2015, had a tracking space of 375 square meters for up to six players when New Atlas visited in 2016. Each player wore a backpack PC, and the headsets and gun controllers were tracked with PlayStation Move orbs. Arena-scale inside-out tracking, as shown with the Oculus Quest in 2018, allows large shared play areas without external trackers.

Combining with other methods

Many applications combine walking in the tracked area with an artificial method for longer distances. The 1999 UNC study described local walking plus flying as the most common arrangement at the time. Walking in the tracked area can also be combined with teleportation to extend the movable space, as noted above.

See also

• Room-scale VR
• Redirected walking
• Walking-in-place
• Omnidirectional treadmill
• Teleportation
• Artificial locomotion
• Location-based virtual reality
• Chaperone
• Guardian System