What Is VR? Definition and How It Works
How a headset actually tracks movement and renders a 3D world, why latency causes motion sickness, and where XR, AR, and MR fit around it.
QUICK ANSWER
Virtual reality is a computer-generated 3D environment experienced through a headset that shows each eye a slightly different image for depth and tracks head and hand movement in real time so the view updates naturally. Modern headsets use 6 degrees of freedom (6DOF) tracking, either inside-out (cameras on the headset itself) or outside-in (external room sensors), and need to re-render the scene fast enough that the delay between moving and seeing the update, motion-to-photon latency, stays low, since delay above roughly 20 milliseconds is what commonly triggers motion sickness. XR (extended reality) is the umbrella term covering VR, AR, and mixed reality together.
VR Is Not New: A Short, Real History
The core idea behind VR predates the modern headset by decades. In the mid-1950s, cinematographer Morton Heilig built the Sensorama, an arcade-style cabinet that combined stereoscopic 3D imagery with stereo sound, wind, vibration, and even smell to immerse a seated viewer, and followed it with the Telesphere Mask, an early head-mounted display, though neither tracked head movement or responded to it. In 1968, Ivan Sutherland and his student Bob Sproull built the first head-mounted display connected to a computer rather than a camera or film, nicknamed the "Sword of Damocles" because it was so heavy it had to be suspended from the ceiling to keep it from breaking the wearer's neck. VR then went through a long quiet period, often called a "VR winter," until 2012, when Palmer Luckey's Kickstarter campaign for the Oculus Rift raised $2.4 million and reignited mainstream interest in the technology that eventually produced the standalone, consumer-priced headsets common today. The underlying idea is 70 years old; what changed recently is that tracking, rendering, and battery technology finally caught up to make it comfortable and affordable.
The Two Things Every VR Headset Actually Does
Strip away the marketing and a VR headset performs exactly two core functions, repeated many times per second: it tracks where the user's head (and usually hands) are in space, and it renders a matching 3D image to each eye based on that tracked position. Everything else, resolution, field of view, controller design, is refinement on top of that basic loop. Get the tracking wrong or the rendering too slow, and the illusion of presence collapses immediately, no matter how good the graphics look in a screenshot.
Stereoscopic Rendering: Why It Looks 3D
Each eye in a VR headset looks at its own separate display area (or its own region of one display), showing a slightly offset version of the same scene, mimicking the small difference between what a person's left and right eyes naturally see in the real world. The brain fuses those two offset images into a single perception of depth, the same mechanism responsible for depth perception in ordinary vision. This is why a flat photo of a VR scene never looks as convincing as actually wearing the headset, the depth cue depends on two simultaneous, correctly offset images reaching two separate eyes, not one image viewed by both.
6DOF: Six Degrees of Freedom
Modern headsets track six degrees of freedom: three for position (moving forward and back, side to side, and up and down) and three for rotation (looking left and right, up and down, and tilting the head side to side). Full 6DOF, as opposed to older 3DOF systems that only tracked rotation and let a user look around from one fixed spot, is what allows someone to physically walk toward an object, lean down to look under something, or step sideways around a virtual obstacle, matching real physical movement rather than just head rotation.
Inside-Out vs. Outside-In Tracking
- Inside-out tracking: cameras mounted directly on the headset capture the surrounding room, and computer-vision algorithms (commonly SLAM or visual-inertial odometry) identify fixed features in that video to calculate the headset's position without any external hardware. The tradeoff is that a controller or the headset itself moving out of camera view, behind the user's back, for example, can cause tracking to degrade or fail.
- Outside-in tracking: external sensors placed around the room track markers on the headset, generally delivering higher precision within the tracked volume, at the cost of requiring that external hardware to be installed and calibrated, and losing tracking if something blocks the line of sight between a sensor and the headset.
Most current consumer and enterprise standalone headsets use inside-out tracking specifically because it requires no room setup, a meaningful advantage for training deployments across multiple sites that cannot each install and calibrate dedicated tracking hardware.
Why Latency Is a Comfort and Safety Spec, Not Just a Picture-Quality Spec
Motion-to-photon latency is the delay between a physical head movement and the display updating to reflect it. The research consensus generally places the threshold where most people start noticing a lag, and can begin to feel discomfort, at around 20 milliseconds, while delays under roughly 15 milliseconds are usually imperceptible. This is the core mechanism behind VR-induced motion sickness: the vestibular system in the inner ear reports one thing (real head motion) while the delayed visual system briefly reports something slightly different, and that sensory mismatch is what the brain interprets as sickness-inducing. High refresh rates (commonly 90 to 120 Hz) and fast tracking exist specifically to keep that gap small enough to stay below the threshold most users can feel.
XR, AR, and MR: Where VR Sits in the Bigger Category
XR, extended reality, is the umbrella term covering the full spectrum from fully real to fully virtual, and VR is one specific point on that spectrum, not a separate category from it:
- VR (virtual reality): the user's view is replaced entirely by a computer-generated environment, no view of the physical surroundings remains.
- AR (augmented reality): virtual elements are overlaid on top of a live view of the real world, commonly through a phone camera or a transparent display, without replacing the real environment.
- MR (mixed reality): a further blend where virtual and real objects can interact with each other, a virtual object can appear to rest on a real table and respond when a real hand reaches for it, rather than simply floating on top of the camera view.
Understanding VR as one segment of the broader XR spectrum, rather than a wholly separate technology from AR and MR, matters practically: many headsets now support more than one mode, and the choice between them for a given training use case depends on whether trainees need full immersion (VR, best for procedures that require full attention and no real-world distraction) or an overlay on real equipment (AR, better for step-by-step guidance on a physical machine in front of them).
For skills training specifically, VR's full immersion is usually the deciding factor: a trainee practicing a high-stakes procedure, an emergency shutdown, a confined-space rescue, a live-panel diagnostic, benefits from an environment with no real-world distraction competing for attention, the same reason flight simulators fully enclose a pilot rather than overlaying instructions onto a real cockpit. AR's strength is the opposite case, guiding a technician through a task on equipment that is physically present and cannot be replicated in a training room.
Field of View and Resolution: The Other Two Comfort Variables
Beyond tracking and latency, two display specifications shape how convincing and comfortable an experience feels. Field of view (FOV) describes how much of the user's peripheral vision the display actually fills, a narrow FOV feels like looking through a diving mask or a pair of ski goggles, while a wider FOV feels closer to natural human vision, which spans roughly 200 degrees horizontally including peripheral vision. Resolution and pixel density determine how sharp the image looks and whether individual pixels or the gaps between them become visible, an effect commonly called the "screen-door effect" on older or lower-resolution displays, where a faint grid pattern is visible over the entire image. Both specifications have improved substantially generation over generation, which is a large part of why recent headsets feel meaningfully more comfortable and immersive than models from just a few years earlier, independent of any change in tracking technology.
Controllers and Hand Tracking
Most VR systems provide some way to interact with the virtual environment beyond simply looking around, either handheld controllers tracked in the same 6DOF space as the headset, or camera-based hand tracking that reads finger and hand position directly without any handheld device at all. Controllers generally offer more precise, reliable input and physical buttons for discrete actions (grab, trigger, menu), while hand tracking trades some precision for a more natural, unencumbered feel, letting a trainee reach out and manipulate a virtual tool or control the way they would with their own hands on the real equipment. Training scenarios built around fine motor tasks, operating a specific control panel, performing a hands-on procedure, generally benefit from the more reliable input controllers provide, while broader wayfinding or simple interaction scenarios can work well with hand tracking alone.
PC-Tethered vs. Standalone Headsets
A PC-tethered headset offloads rendering to an external computer's graphics hardware, generally delivering the highest visual fidelity at the cost of a cable and a dedicated PC per headset. A standalone headset has its own onboard processor, storage, and battery, needing no external computer at all, trading some graphical headroom for portability, lower cost per unit, and dramatically simpler deployment across multiple locations. Most enterprise and workforce training programs today favor standalone hardware precisely because it scales, shipping ten headsets to ten sites is a very different logistics problem than shipping ten headsets plus ten dedicated gaming PCs, each needing its own setup, maintenance, and IT support.
WE BUILD THIS IN VR — THE PRIME VR
We build enterprise training on standalone VR hardware specifically because it deploys across every site without a dedicated PC per headset, and we design every scenario around the tracking and latency fundamentals above, so trainees stay comfortable through long practice sessions instead of fighting motion sickness caused by a poorly optimized build.
Book a discovery callFrequently Asked Questions
What is VR? +
Virtual reality is a computer-generated 3D environment that a user experiences through a headset showing each eye a slightly different image (for depth) while tracking head and often hand movement in real time, so the view updates naturally as the user looks and moves, creating a sense of presence inside the simulated space rather than watching it on a flat screen.
What does VR stand for? +
Virtual reality. The word "virtual" describes an environment that functions like the real one for the purposes of the experience even though it is computer-generated, not physically real.
How does a VR headset actually work? +
It shows each eye a separate, slightly offset image on its own display area (stereoscopic rendering, for depth perception), tracks the position and rotation of the headset in real time (6 degrees of freedom: moving forward/back, left/right, up/down, and rotating around each of those axes), and re-renders the scene fast enough that the delay between a head movement and the updated image is imperceptible.
What is 6DOF in VR? +
Six degrees of freedom: three for position (moving forward/backward, side to side, up/down) and three for rotation (looking up/down, left/right, and tilting the head). Full 6DOF tracking, as opposed to older 3DOF systems that only tracked rotation, is what lets a user physically walk around and lean into a virtual space rather than only look around from one fixed point.
What is the difference between inside-out and outside-in tracking? +
Inside-out tracking uses cameras mounted on the headset itself, using computer vision to map the room and track the headset's position with no external hardware required, at the cost of losing track if a controller or the headset moves out of the cameras' view. Outside-in tracking uses external sensors placed around the room to track the headset, which is generally more precise but requires installing and calibrating that external hardware, and loses tracking if something blocks the line of sight between the sensor and the headset.
Why does VR sometimes cause motion sickness? +
Largely from motion-to-photon latency, the delay between a head movement and the display updating to match it. Research generally finds that delays above roughly 20 milliseconds become noticeable and can trigger discomfort, while delays under about 15 milliseconds are usually undetectable, which is why headset refresh rate and tracking speed are treated as comfort and safety specifications, not just picture-quality specifications.
What is XR and how is it different from VR? +
XR (extended reality) is the umbrella term covering virtual reality, augmented reality, and mixed reality, along with anything in between on the spectrum from fully virtual to fully real. VR replaces the user's view entirely with a virtual environment; AR overlays virtual elements onto a view of the real world; MR blends the two further, allowing virtual and real objects to interact. XR is the category name, not a fourth separate technology.
Do you need a powerful computer to run VR? +
It depends on the headset category. PC-tethered VR headsets rely on an external computer's graphics processing and generally deliver the highest visual fidelity. Standalone headsets have their own onboard processor and battery and need no external computer, trading some graphical power for portability and simplicity, which is why most enterprise and training deployments today favor standalone hardware.
What is field of view in VR? +
How much of a user's vision the headset display actually fills. A narrow field of view feels like looking through a diving mask, while a wider one approaches natural human peripheral vision. Wider field of view generally increases immersion but is also more demanding on the headset's rendering hardware.
Do VR headsets use controllers or hand tracking? +
Both exist and serve different purposes. Handheld controllers, tracked in the same 6DOF space as the headset, offer more precise and reliable input with physical buttons. Camera-based hand tracking reads finger and hand position directly with no device in hand, trading some precision for a more natural feel. Many current headsets support switching between the two depending on the task.
See what VR training actually looks like
We build standalone VR training programs designed for comfort and real skill transfer.