Comfort in VR has been the unsolved problem for a decade. People who tell you they wear a headset for hours have either gotten used to it, are exaggerating, or have a head shape that the hardware happens to fit. For everyone else, the standard headsets of 2018 through 2023 produced a hard limit at about 45 minutes before the discomfort outweighed the experience.

In May 2026 the picture has shifted, but unevenly.

What’s genuinely better

Weight has come down. The flagship consumer headsets are now landing in the 380 to 450 gram range with the strap, down from the 550 to 650 gram range that was typical three years ago. The reduction comes from better optical stacks (pancake lenses replacing Fresnel in nearly all serious products), more efficient SoCs that don’t need as much thermal management, and battery placement that distributes the load more sensibly.

The weight reduction sounds modest on paper. In actual use it’s the difference between a headset that becomes uncomfortable around 30 to 45 minutes and one that’s wearable for genuinely longer sessions. For people whose neck strain was the binding limit, this is a meaningful change.

Strap design has stopped being the afterthought it used to be. Halo straps with rear support and forehead distribution have become standard at the higher end. Counterweight options — battery packs that mount at the rear of the strap to balance the front weight of the optics — are no longer DIY accessories; they’re shipped in the box on most premium products. The cumulative effect is that pressure points that used to drive 90 per cent of comfort complaints are largely engineered out of well-designed current hardware.

Light leakage and facial interface design has improved. The foam and silicone face gaskets in 2026 products contour better, breathe better, and accommodate a wider range of face shapes than the one-size-fits-all approach of earlier generations. Several manufacturers now ship multiple gasket sizes in the box; some require purchase of optional sizes, which is a worse story for consumers but better than the no-options alternative.

Optics have advanced enough that the IPD adjustment range covers a wider population. Mechanical IPD adjustment with continuous (rather than detented) movement is the dominant design now, and the eye relief adjustment ranges accommodate glasses-wearers more reliably than the 2022-vintage products did. The number of users who simply can’t physically use a given headset because of optical incompatibility has dropped meaningfully.

What still doesn’t work

For users with broader-than-average heads, the comfort story is significantly less good. The strap mechanisms in most current products top out at strap circumferences that exclude perhaps 15 per cent of adult male users in a typical Australian population sample. Asking those users to comfortably wear a headset for an hour-long session is asking them to adapt to hardware that doesn’t fit them.

For users with narrower-than-average IPDs — particularly women and younger adults — the optical sweet spot of many flagship headsets sits outside the IPD range that would suit them. The result is either a soft visual experience or persistent eye strain after extended use. The hardware that addresses this exists; it’s not always the hardware that gets bought because the buying decision is often made by users in the IPD range that’s well-served.

For users who wear glasses, the story is improved but not solved. Prescription lens inserts are now widely available aftermarket and from manufacturers, and they substantially improve the experience. The friction of ordering and installing them is non-trivial, and a meaningful portion of glasses-wearing users never get around to it. The default out-of-the-box experience for these users remains poor.

For users prone to motion sickness, the underlying problem hasn’t been solved. Better refresh rates, lower latency, and better tracking have softened the symptoms but haven’t eliminated them for the susceptible population. The percentage of users who experience meaningful nausea after 30 minutes of locomotive VR content has come down but is still noteworthy.

What enterprise deployments tell us

The enterprise side of the market — training, simulation, design review — has stopped pretending that current consumer hardware is comfortable enough for full-day use. The realistic deployment pattern is sessions of 45 minutes to 90 minutes followed by mandatory breaks. That’s a meaningful constraint for some training applications and a hard limit for others.

Where enterprise users have moved toward longer-session workflows, they’ve done it with hardware that prioritises comfort over the visual fidelity of the consumer flagships. Lower resolution, lower frame rates, but lighter weight and better thermal management. The trade-off is rational for the use case. It’s also a different category of product than what gets reviewed.

What I’d watch

Three things over the next 12 months.

The mass-market push toward genuinely lightweight (under 250 gram) headsets driven by improved optical stacks and reduced compute load. Several manufacturers are signalling this for 2026 launches; whether the products actually land at that weight while delivering useful experiences remains to be seen.

Strap and gasket customisation as a standard product feature, not an aftermarket. The manufacturers who ship genuinely fitted hardware will see better engagement metrics and lower return rates. The ones who ship one-size-fits-most will continue to lose users in extended-use scenarios.

Glasses-free correction inside the headset. Several research efforts are working on adjusting the optical stack to accommodate the user’s prescription without requiring a separate lens insert. If this lands as a shipping product, it changes the comfort story for a meaningful portion of the population.

The honest assessment for May 2026: VR comfort is genuinely better than it was three years ago. It’s not yet at the point where extended wear is comfortable for everyone, and the hardware progress is uneven across user populations. The headlines that suggest comfort is solved are wrong. The complaints that suggest nothing has improved are also wrong.