It’s an interesting game design question. Shooters in the style of Arizona Sunshine trace their roots back to the arcade galleries, where the target range is at the least limited to one side, frequently further split in narrow lanes. In video games it soon developed into the on rails shooters like Time Crisis.
Many of the VR titles stick to the one side formula so as not to produce much cable tangle or surprises, ranging from the narrow like Holoball or Audioshield to the wider like Space Pirate Trainer or Blarp. Those last two compensate a little by giving access to a shield you can use to cover some directions you’re not currently looking. Blarp further eliminates surprises by having all action originate from player action; the chaotic multidirectional results stem from continuous physics only.
Many video games use an “out of sight, out of mind” strategy, where the simulated game world doesn’t extend far beyond what can be seen. A good example of this is Grand Theft Auto, where cars exist only a small distance out of the screen edges. This traces back to other limitations, such as memory and processing speed. Two well known titles showing this are Pac-Man and Asteroids which employ wrapping at the screen edges, a geometry topologically equivalent to a donut world. This became a tougher problem with 3D graphics and first person perspectives, with workarounds like Doom’s binary space partitioning. Many games have serious issues with pop-in because once you look horizontally there really isn’t a limit to how far, an effect that should be well known in driving simulations.
Technically, in 3D graphics rendering we use a viewing frustum wherein things are visible. Stereoscopic systems use at least one per eye, giving slightly different perspectives. The depths are frequently chosen for numeric precision in the Z buffer, the field of view to match display hardware (often slightly wider to permit late adjustment, known as time warp or reprojection). The resolution of the render buffer is chosen as a compromise between required performance and display capability. VR systems control the placement and orientation entirely by the viewer’s, and in the case of Pimax 5K/8K models, the rendering planes should be angled outwards just like the displays are. More advanced techniques like multiresolution shading are done using more frustums, in this case touching, while Varjo and similar setups (perhaps a projected sim tent with a central monitor) can overlap them. In that case, an occlusion model like SteamVR’s hidden area stencil mesh can be used to reduce the effective overlap. Multi-screen systems, like CAVE and early Doom’s three screen mode, also use these techniques. Things like nvidia’s surround really should, but I think for compatibility reasons just stick to adding more views into one plane.
Many other games compensate for a wider field of action by placing the viewer elsewhere, ranging from over the shoulder views to full on god view. The Lab contains a game, Xortex, which applies that technique nicely in VR; the player controls a ship at arm’s length. This makes it easier to survey the surroundings of that ship, and threats can occur from many directions. Another well known example is Lucky’s Tale. Some games even shrink the action range further to a theatrical view, for example Witchblood. This technique is automated in Theatre Mode.
It’s no accident that the last two I mentioned in the reduced field of action example are Oculus sponsored titles. Essentially, the bigger the budget, the wider they want the audience, and that will include stationary people who don’t turn around (perhaps sitting in a sofa). This is the main reason why wider field of action will be rare in “triple A” games.
With my very limited knowledge of available titles, if I were to pick a game to try to demonstrate wide FoV gameplay, I would first consider Space Pirate Trainer. It’s well known, mostly sticks to a 180 degree field, but does successively widen the region you want to pay attention to while permitting the player to remain largely stationary. The second game I’d consider is Racket: Nx, third Blarp. These aren’t your big budget titles, but they’re ones where the game can grow in width without demanding the player move along excessively (though SPT certainly encourages it a bit more). It wouldn’t surprise me if Blarp is also very low on GPU demands. They’re all focused heavily on the motion controls too.
Pardon my unstructured rambling. Did I get any closer to addressing the question?
