DisplayLink Showing 60GHz Wireless Adapter Reference Design for Oculus Rift

Yes, mobile will never be able to match contemporary PC graphics. That’s just physics. But that doesn’t mean that it won’t reach a point where the graphics provided by mobile are good enough for all but the most hardcore, or at least good enough that the benefits provided by mobile outweigh the value of the improved graphics provided by traditional PCs. And if we reach a point where most people are regularly using VR in their lives and they’re mostly doing it through mobile hardware, then mobile VR won’t so much be a toy as the segment driving the industry.

The problem that I have with comments like your is that it seems to assume current mobile power, when most of the time people talking about mobile being the future of the industry are talking about years in the future. Sure, mobile storage isn’t huge today, but the growth of storage capacities has far outstripped the growth of game install sizes. Right now you can buy a 256gb sd card for just over $50. Sure, that’s not super impressive compared to the multiple terabytes that you can have on a PC, but it’s reasonable to think that in a few years we’ll see mobile storage that’s large enough for some big game installs.

Similarly, yes, the Snapdragon 821 in the Oculus Go isn’t very powerful, but it’s not even the most powerful SoC available today, let alone what we’ll see in the next several years. Yes, mobile will never match what can be done in a desktop machine; that’s just physics. But that doesn’t mean that mobile parts won’t be able to run more graphically demanding games in the future.

As for the last paragraph, no graphics aren’t everything. That’s a ridiculous overstatement that nobody actually believes. A super powerful graphics card that cost $10,000 wouldn’t sell many units to consumers, even hardcore gamers, because the increased graphics power wouldn’t be worth the increased cost. Every feature in a product is balanced against other available features and how they contribute to the overall experience. There is a certain level of graphical realism (and I don’t claim to know what that is) which is good enough for most consumers, or at least good enough that other concerns like cost, ergonomics, ease of use, tracking, ability to interact with friends, etc. become more important. The need to build/buy, setup, and maintain a big gaming PC is a huge drag for most people, and is a big barrier of entry. Some might be convinced to overcome that if the VR experience was sufficiently compelling, but it wouldn’t take much improvement in mobile to make that the more compelling option.

Not at all. I have a Rift that I use fairly regularly on an older but competent gaming PC. There’s a big difference between “graphics are important for immersion” and “graphics are everything”. If the current era of VR has taught us any lessons it’s actually that graphical fidelity is not nearly as important to immersion as many people had assumed. High frame rates, good tracking, good audio, ergonomics, and good inputs are also exceptionally important. So important, in fact, that people can be immersed in relatively cartoony worlds. But, of course, really good graphics are also great. It’s just that it’s not some factor that always must be improved some limitless amount, never to be satisfied. At a certain point, graphics are good enough for most people such that other concerns, like cost, become more pressing. You could build the most amazing VR setup ever and nobody is going to buy it if it costs $100,000. Similarly, you could have the most high resolution displays ever, but if the headset causes neck strain after ten minutes, nobody will care. Or you could have screens covering 250 degrees of horizontal view, but a really clunky controller could ruin it.

The point isn’t that graphics aren’t important at all, or that we’ll reach a point where nobody will ever aspire to better graphics (though I do think that that’s an eventuality, there are diminishing returns are you edge ever closer to photo realism), it’s that while mobile SoCs available today, like the Snapdragon 821 in the Oculus Go, are under powered for a lot of gaming experiences, that will change as newer, more powerful parts are available. And there’s probably a point coming, and I think that it’s at least in the medium term, where the graphics power of mobile chips will be sufficiently good that for most people the convenience of not needing to buy or setup a gaming PC will far outweigh the increase in capabilities that the gaming PC represents. Some enthusiasts will have the interest and disposable income to keep that segment going, but most people simply aren’t singularly obsessed with graphics, and even those who are have limits.

Sure, if you ask people if they wand 4k screens, they’ll say yes, they want 4k screens. And if you ask them if they want a flying car, they’ll say that they do. The question isn’t whether people would like having a given feature, but what they’re willing to trade off for that feature. Sure, people want the tech in the Vive Pro Eye, but few are likely willing to pay to buy it because they don’t care enough about having that tech to justify that cost. The point about mobile vs desktop isn’t that mobile will ever be better in terms of raw power. It won’t. The point is that better graphics isn’t the only thing that people care about. They care about cost a lot. They care about how hard a device is to set up and maintain. The original comment was that mobile is the future of VR, and I think that that’s right.

Desktop-based VR will continue to exist, but it will be a small niche of people willing to pony up the time and cost to have bleeding edge processing power. For most people, a few generations of improvements in mobile graphics processing power will give them a VR product that is powerful enough while also being cheaper and easier to own and use. And it’s the “most people” demographic that drives an industry like VR. Lots of people would love to own a supercar, but most are satisfied with a mid range sedan or SUV because it does what they really want it to do at a price they can afford.

There are a couple problems with that reasoning. First, people posting about how they won’t get into VR until the graphics improve isn’t a reliable measure because it’s a self selection problem. Second, and I’ve said this several times, but you still seem to be missing it, but neither I, nor I think the original poster, are talking about current mobile parts. Yes, the Snapdragon 821 in the Go is under powered for gaming, and even the 835 in the Quest by not be sufficient for lots of games. Even the 845, which is already available in the market, may not cut it. We’re talking about a few years in the future. So mobile VR at that time will likely have better screens than today’s units, and will certainly have more powerful SoCs. Now, desktop PCs will also have whatever follows the RTX/Vega lines. But the point that I have made, many times now, is that given the choice between mobile VR, like the third or fourth generation of the Quest, and desktop based VR, like the third or fourth generation of the Rift, most people will choose the Quest.

It is, of course, possible that I’ve read this wrong, but the vast majority of people in my life are simply never going to buy or build a gaming PC. They don’t have the money, interest, or technical knowledge. Yes, the graphics will always be better on desktop, but mobile VR has a lot of features that are really appealing that desktop-based VR will never be able to do. Maybe the future of VR is to remain a niche hobby enjoyed by enthusiasts who already own gaming PCs. But if the future of VR is mainstream, and everyone interested in VR should want it to be mainstream, then the future of VR will be in mobile products.

I don’t seem to be talking about hardware that isn’t created yet, I am explicitly talking about hardware that isn’t created yet. And I’m not talking about literal smartphones, but mobile hardware. Mobile VR has used a lot of off the shelf phone parts previously, but going forward they’re going to need to have hardware designed for VR (Carmack has said as much). I don’t know where you got the idea that I’m talking about smart phones. I haven’t mentioned them at all, and the examples that I have referred to, the Go and the Quest, don’t use smartphones.

Console VR is certainly going to be more appealing than PC VR (and indeed the PSVR has already sold a ton more units than technically superior products like the Rift), but a console + HMD setup is still going to be more expensive and complicated than an all-in-one system like the future versions of the Quest.

And seriously, nobody is talking about smartphones.

I didn’t say anything because I was clear that I was talking about mobile hardware, not phones, and your comment about the Go, which doesn’t use a phone, made it seem like you understood that. Your comment about being limited by the fasted phone of the moment could have been a reference to mobile hardware targeting high end phones, but I don’t know, man, that’s on you. The Go, and soon the Quest, use a lot of the parts that are in higher end phones, particularly the SoC and the screen, but it’s not literally a phone, which changes lots of things, like, for example, thermals as compared to phones. I have consistently talked about self contained, mobile VR, like the Quest. I literally never referenced phone-based VR, and it was only when you started to really focus on that that I brought it up because it made no sense as a response to the points that I had been making. For example, “Similarly, yes, the Snapdragon 821 in the Oculus Go isn’t very powerful, but it’s not even the most powerful SoC available today, let alone what we’ll see in the next several years.” If you thought that I was talking about mobile phone-based VR, man I don’t know what to tell you. I was talking about mobile parts, like the SoC.

Yes, the Quest needs to be set up. That doesn’t really address whether it is more or less complicated than desktop-based VR. That also needs to be set up. And you have to set up base stations for tracking (unless you’re using Window HMDs). And you have to buy, own, and maintain a desktop PC, or, in the case of consoles, a console. For both consoles and PCs, that’s an extra layer of cost and complexity that mobile doesn’t have.

I don’t really have an opinion on what other people think like. Indeed, what I try to get across to people in these discussions is that the kinds of people that hang around on enthusiast VR websites are not typical consumers that VR companies are going to want to appeal to to make VR go mainstream. I’m someone with a gaming PC and a willingness to upgrade it because I also want better graphics, etc. But appealing to my interests is not going to produce a product that my non-gamer sister is going to consider buying, or my aunt and uncle. The self-selection problem is that people who post on VR websites have self-selected themselves because this is a topic on which they are at least interested in enough to go to a VR website to discuss current and upcoming technology. That’s the definition of self-selection.

We’re gamers and enthusiasts, which means that we’re not typical consumers. Some company will likely see enough of a market to continue making enthusiast-targeted VR systems, but companies like Oculus and HTC are making mobile-based VR specifically to appeal to the mass market, and if they are correct and the mass market adopts those products then that means that mobile is the future of VR. Yes, “people” want expensive, cutting edge tech. It’s just that those people are a niche audience of gamers, not the broader population of consumers that are the future of VR.

Of course it does, but this conversation specifically focused on graphics. Do you have something of value to add?

Mate, we’ll NEVER have enough power, surely you know that by now? People have said what you’ve said time and time again since the 60s… ‘ok 128k ram is more than enough’.
It’s not.

This is VIRTUAL REALITY we are taalking about, don’t judge power needs based on the crap they are serving up today cos of limits, yeah when mobile 10-20 years from now is faster than 10 of today’s fastest PCs… PCS will be 100x faster and they will be driving VR and what VR means, vast resolution, FOV, physics, haptics, and photorealistic worlds that deliver on the term virtual reality. Mobile will never be enough.. for a long long time, and it will always be second to a larger computer system.

When we can do wireless properly from a PC, anyone with a brain who has a PC will want to do it that way, not limit themselves to cartoon games.

And btw, VR has a long long way to go yet to get it right, more than just the visuals, we have to solve proper movement, nothing cuts it right now, certainly not roomscale (real world doesn’t map to VR worlds at home – neither texture, shape or walls). I’ve had dk1,2,vive (in a 15×10 VR room just for it), PSVR and Rift and it’s clear VR has a lot of work to do that will not be solved by relying on all-in-one mobile rubbish.

I don’t see any HMD driven VR becoming mainstream. Ever. Visions of massive amounts of people spending their time inside an HMD is a sci-fi pipe dream. VR is no match for drugs when it comes to escaping reality. That’s the people’s choice. I would expect something like a more advanced google glass AR device to hit mainstream.