One of the slow and quiet phase changes in the industry is IMHO the broader recognition of how to create VR experiences that don't induce motion sickness, such as your example of "you can 'legally' move forward at a constant speed and no-one gets upset", which for instance Pistol Whip puts to good use. That game has never given me trouble.
I got into VR about six months ago with the Quest 3, and for a moment I wondered if all the talk of motion sickness was overblown, or maybe it didn't bother me. So I boldly downloaded a roller coaster simulator. Literally couldn't do more than about 10 seconds of it, even after setting up the blinders (which also effectively ruin the nominal experience anyhow).
So I do think the experiences are getting better about that over time.
However, the problem is, the resulting limitations are a pretty big deal in terms of game design. The VR industry really, REALLY wants to do first-person shooters. I mean, first person, it's almost right there in the name. But the technology that on paper is the very embodiment of "first person" also can't really do that. I've played a bit of Batman on the headset, with it set to 'normal' motion and navigation, and that rides the line of nausea for me, and Batman is really generally a slowly-moving character. I can't imagine trying to do a high-mobility FPS in the style of Quake 1. None of the "I really want to be a first-person shooter" games I've tried really quite work for me. (Have not tried Alyx, admittedly. Next sale. But even if that works, nobody else is copying it very well.)
So if you think of all the possible games, and then remove from them all the games that induce motion sickness, you've cut out rather a lot. Then you face the problem of, of the remaining games, how many of them are actually improved by VR? For instance, you can make a VR chess game, but beyond the initial "oh wow" of the 3D environment, chess is chess, and if anything the VR controls become an impediment versus the precision of the several paradigms for playing chess with mouse and touchscreen that already exist that are able to just fade away until you are just playing chess without thinking about the interface. The VR interface is always there.
The answer is absolutely not zero. Beat Saber, for instance, sure, nominally it could be done in conventional 3d on a 2d monitor and some motion controls, but the millisecond-by-millisecond kinesthetic experience is certainly a qualitative change in VR. It is The Game for a reason. But for VR to ever be more than just a niche, it is going to have to get to the point where it is nearly transparent. We're talking a "VR headset" that is basically the weight and encumbrance of conventional glasses. Maybe a thin wire that goes to a dedicated battery and external compute. If I had something like a dedicated shoulder pocket for that or something it might not be too bad. And we're a ways away from that still. And even then gaming is going to remain a niche unless someone finds a genre that works in VR, doesn't work well without VR, and basically doesn't involve motion through space, and I've got no more idea what that is than most game developers. I just can't entirely exclude its existence.
I got into VR about six months ago with the Quest 3, and for a moment I wondered if all the talk of motion sickness was overblown, or maybe it didn't bother me. So I boldly downloaded a roller coaster simulator. Literally couldn't do more than about 10 seconds of it, even after setting up the blinders (which also effectively ruin the nominal experience anyhow).
So I do think the experiences are getting better about that over time.
However, the problem is, the resulting limitations are a pretty big deal in terms of game design. The VR industry really, REALLY wants to do first-person shooters. I mean, first person, it's almost right there in the name. But the technology that on paper is the very embodiment of "first person" also can't really do that. I've played a bit of Batman on the headset, with it set to 'normal' motion and navigation, and that rides the line of nausea for me, and Batman is really generally a slowly-moving character. I can't imagine trying to do a high-mobility FPS in the style of Quake 1. None of the "I really want to be a first-person shooter" games I've tried really quite work for me. (Have not tried Alyx, admittedly. Next sale. But even if that works, nobody else is copying it very well.)
So if you think of all the possible games, and then remove from them all the games that induce motion sickness, you've cut out rather a lot. Then you face the problem of, of the remaining games, how many of them are actually improved by VR? For instance, you can make a VR chess game, but beyond the initial "oh wow" of the 3D environment, chess is chess, and if anything the VR controls become an impediment versus the precision of the several paradigms for playing chess with mouse and touchscreen that already exist that are able to just fade away until you are just playing chess without thinking about the interface. The VR interface is always there.
The answer is absolutely not zero. Beat Saber, for instance, sure, nominally it could be done in conventional 3d on a 2d monitor and some motion controls, but the millisecond-by-millisecond kinesthetic experience is certainly a qualitative change in VR. It is The Game for a reason. But for VR to ever be more than just a niche, it is going to have to get to the point where it is nearly transparent. We're talking a "VR headset" that is basically the weight and encumbrance of conventional glasses. Maybe a thin wire that goes to a dedicated battery and external compute. If I had something like a dedicated shoulder pocket for that or something it might not be too bad. And we're a ways away from that still. And even then gaming is going to remain a niche unless someone finds a genre that works in VR, doesn't work well without VR, and basically doesn't involve motion through space, and I've got no more idea what that is than most game developers. I just can't entirely exclude its existence.