Oculus' high PC specs are Sony's chance to shine
With PC VR looking like it'll be a high-end niche for years, Morpheus gets first shot at mass-market VR
It is with absolute confidence, and yet a heavy heart, that I say that I won't be getting an Oculus Rift when the consumer model of the headset arrives early next year. This is a reversal of what I'd have told you just last week. I've spent months justifying a headset and the hardware required to run one to myself; each time I've visited a developer and tried out their demo, each time I've been to a trade show and slipped on a Rift prototype to immerse myself in something new and fascinating, my excitement has mounted. I think I'm very much the kind of person to whom this sort of tech appeals; but I won't be getting one, not for quite a long while.
You can probably guess why, if you've been following the news. Oculus announced the minimum hardware specifications required for the Rift a few days ago, and they are... High. Eye-watering high. An NVIDIA GTX 970; an Intel i5-4590 chip. Oh, and forget about laptops, pretty much. Even if they meet those specifications (which even most high-end gaming laptops don't), you need your HDMI port hooked directly up to the GPU, as I understand it, and most laptops don't work like that. Even allowing for six months for component prices to descend as new hardware enters at the top of the market, I'm looking at the best part of two thousand dollars to equip myself with a PC like that - like a great many people, I suspect, I'd be starting from scratch on the whole thing, without any old PCs lying around to scavenge a monitors or PSU from. That's two grand before I even buy an Oculus Rift, or any games to play on it.
"What this might mean to Oculus and to VR on the whole is that a product we all knew would be niche is going to be... restricted to a thin slice at the upper end of what's (mostly) ironically termed the PC Gaming Master Race"
So I won't buy one - not when it launches, not next year at all. By mid to late 2017, perhaps, I'll be in a position where I might think about it again, although it'll likely be conditional upon the issue with laptops being fixed; having to find somewhere in my apartment to set up a hefty desktop PC is honestly an even bigger killer than the financial issue. I'm disappointed, certainly, but the whole thing has also quite honestly dampened my enthusiasm for the tech somewhat - even as I recognise that if Oculus' calculations are correct, then there's honestly no alternative to the specs they've cited. Hard technological limits do not make way for the feelings of disappointed consumers.
I'm cautious about extrapolating from my own experience to the broader market, but at the same time, I know I'm not going to be alone in this situation. Gaming laptops have taken over a lot of ground from desktops in recent years; they're all out of the question, as are all Mac laptops and desktops, which are also far more popular today with gamers and developers than they've ever been. Lots of people are facing building or buying a system from scratch if they want to play VR games; lots of them might not be remotely happy with the form factor (desktop vs laptop) or OS (Windows vs OS X) involved in that. It's a big, big hurdle to jump, even for an early adopter.
What this might mean to Oculus and to VR on the whole is that a product we all knew would be niche is going to be far more niche than we expected, for far longer than we expected. It's going to be restricted to a thin slice at the upper end of what's (mostly) ironically termed the PC Gaming Master Race; the rest of the market, even the core gamers and early adopters, are going to arrive within Oculus' sights only very gradually, as component prices fall and average PC specs rise. It's a conundrum for developers who have been aiming at VR, too; although their initial market is going to be made up of rather wealthier gamers than they'd expected, it's also going to be much, much smaller and significantly more slow-growing than anticipated. It wouldn't be entirely surprising to see several studios reevaluating their budget and resource allocations to VR projects in the coming weeks and months, as they recalculate just how long it's going to be before Oculus can provide a viable market.
On the other hand, Oculus isn't the only player in the VR market any more; they're not even going to be the first to market, with Valve and HTC's Vive headset likely to be the first serious contender to hit the market, towards the end of this year. Valve hasn't announced recommended PC specifications for Vive as yet, which has allowed Oculus to set expectations; whether Vive follows suit with a high recommended spec, or Valve decides to differ significantly with Oculus over this point, will be extremely interesting to watch.
One possibility is that Oculus' specs don't just represent a technical necessity but a - somewhat subjective - statement of position on the types of experience VR should provide. Something which a number of VR developers have stated is that the technology at present lends itself not to the kind of photorealistic, "Crysis 3 but in VR" style experiences that some may dream of; rather, that it works best with more simple and highly stylised graphics and artwork. There's a strong logic behind that claim; aside from the fact that a consistent framerate seems to be much more important than anything else in a VR experience if motion sickness is to be avoided, it also makes sense that the brain might be more willing to accept immersion in a stylised world than in a "realistic" one suffering badly from uncanny valley effects. Now, it's possible that Oculus has based its high specs off the requirement even for simple VR titles - we simply don't know - but if those specs reflect instead the requirements of modern high-end 2D games rendering as VR games, that's actually quite a blunt statement of intent and ideology on the part of Oculus.
"No pressure, Sony, but if you mess this one up, the market might not give it another chance for many years to come"
It's possible that Valve may go a different direction, encouraging developers to embrace simplicity and build VR games that still give a perfectly good, frame-locked experience on significantly lower-spec hardware. That would be extremely interesting, and a little worrying, suggesting a rift (no pun intended) between the two emerging VR giants which is as much ideological as technological. Alternatively, Valve may match Oculus' specs, essentially confirming that PC VR is going to be a niche rich man's toy for the next couple of years, at least.
What of Sony, in all of this? Oculus' specs blow the PS4, to which the Morpheus headset is tethered, out of the water; the PS4 can undoubtedly claw back some of that gulf in raw power thanks to its fast RAM, tightly integrated architecture and so on, but it's still pretty clear that games developed with Oculus' spec in mind aren't going to run on PS4 and Morpheus without major changes. This, too, could go two ways. PS4 could turn out simply to be underpowered for VR; that's the fear of many in the wake of Oculus' announcement, although it's clearly not a fear shared by Sony, which has shown off many impressive tech demos of Morpheus in action. Alternatively, PS4 could turn out to be a platform with a different vision for VR - one which may, yes, focus on simpler and more stylised games, but which may be a much more realistic way to ease a broader market into the notion of VR than thousands of dollars worth of hardware pumping the very latest 3D graphics into a headset.
One thing is certain; unless Valve has a very big surprise in store for Vive's recommended specs, Sony is going to have absolutely the only VR solution at a consumer-friendly price. Morpheus will probably launch after its competitors on PC, but this is looking increasingly like an unimportant handicap; if Morpheus is reasonably priced and actually works, even within more strict limitations than its PC rivals; if Sony can support it properly with first- and third-party games, rather than sending it out to die as it did with PS Move and EyeToy; then Morpheus will have an unchallenged run at mass-market VR for at least a year or so, untroubled by the vastly more expensive rivals floating around the top of the PC market. This is, potentially, a huge opportunity for Sony. It also puts far more onus on Sony not to mess up the VR market with a headset that overshoots and under-delivers; it won't take too much motion sickness to put VR back to bed for another generation. No pressure, Sony, but if you mess this one up, the market might not give it another chance for many years to come.
This is vital for a variety of games.
Edited 1 times. Last edit by Keldon Alleyne on 21st May 2015 4:45pm
OR: Has high minimum specs, but those tech specs are future-proofed. By that, I mean the PC you need will last a decent amount of time. High initial outlay, but you're not just getting a VR-and-games machine, you're getting everything that comes with the PC eco-system - games, but also MS Office, media and downloading centre, etc. Contrast with...
Morpheus: Tied to PS4. This means there's a very specific end-date for the consumer, which is the life of their PS4. Sure, that doesn't matter right now, but by the time the Morpheus releases, the PS4 will have been out... say, two-and-a-half years? Reasonably, that's about half-way through the life-time of the console (is anyone expecting this gen to last more than 5 years?). People may be wary of picking up something that is obviously so old, especially if the Morpheus VR system can't be used with a PC in the future (can it?).
Re: PC component prices:
Unless you're going for a laptop (which you discount), you're paying... a little under half that price if you're building yourself (based on Amazon UK prices), and a little over half that if you buy a pre-built from Scan. But, yes, still more than a PS4. Though, as I note above, more possibilities than a PS4.
(Edit: Since I'm in the UK, I assumed you were talking in pounds. If you're in dollars, then you're still off, but less so. :) )
The Vive is an unknown, not least because of the possibilities for it to be promoted with the Steam Machines. The cost of the bundle would be a lot (£1500-ish?), but I could easily see a Steam Machine+Vive bundle being a thing (assuming Valve's inept marketing doesn't kill it first). I think the Vive has the best shot out of all three for consumer-purchasing, but that's because it's got the most compatibility and the most interesting experience (that we currently know of).
Edited 5 times. Last edit by Morville O'Driscoll on 21st May 2015 5:19pm
Edited 1 times. Last edit by Jehferson Wohllerz Curupana da Rocha e Mello on 21st May 2015 5:01pm
The GTX 970 (and similar) is already the go-to GPU for PC gamers who want to comfortably run multiplaform games, and prices will drop further (possibly quicker if VR sparks a wave of upgrading) by the time the Rift hits store shelves.
If you're really determined to buy into what is still clearly a nascent technology without a decent PC, there are always the mobile VR solutions.
This page might interest you, re: Valve's openness and standards:
https://steamdb.info/blog/steamvr-what-we-know/
:)
Also since the Vive uses exactly the same displays as the Oculus CV1 (only in landscape position rather then Oculus's portrait), don't count on the Vive needing less specs with the same games.. And let's not forget, if it runs on the Vive (which uses SteamVR) it will run on the CV1 (which also has SteamVR support)..
Ofcourse it is possible to do with lesser specs, but don't count on the graphics to look like battlefield 4 on ultramode..
Edited 1 times. Last edit by Andrew Jakobs on 21st May 2015 5:34pm
Valve must have a vast amount of marketing intelligence from their 125 million registered users including their PC specs and how often they upgrade so I fully agree that their choice will be very interesting. My personal hunch is that mobile VR will also be important and that could obviously be good for Oculus. I took a punt on the Wearality Kickstarter last week. I couldn’t tell if it will be as good as they claim and there was only one way to find out: https://www.kickstarter.com/projects/wearality/wearality-sky-limitless-vr
After all the hype I had assumed we would get a Hololens type form factor. I can’t help feeling that a tethered box on your face won’t be that popular. I'm just glad that the phoney-war is nearly over.
I'm really confused about laptops not working. I went to an event a while back where folks had DK1s on laptops all over, and everything seemed to be alright. Has the tech just evolved past that point, to where that kind of power level isn't viable?
The limitation to desktops seems like the biggest blow to the early potential. It means I can't grab my Rift and go to a buddy's place to show them the games, and means it's locked to a desktop somewhere, unless it does a really, really good job of being plug-and-play, but with new hardware and something this technically demanding, I just can't see that being the case.
"Oh, and forget about laptops, pretty much. Even if they meet those specifications (which even most high-end gaming laptops don't), you need your HDMI port hooked directly up to the GPU, as I understand it, and most laptops don't work like that."
Edited 1 times. Last edit by Morville O'Driscoll on 21st May 2015 7:00pm
That said I'm perfectly fine with the CV1 specs (if I'm not going to play analyst) because I'd like to have as much variety as possible. I'm just not that interested in playing party games and tech demos for another couple of years until everyone can afford an upgrade. If VR can be an incentive to build more powerful hardware and push prices, maybe it won't have to take 5 or 10 years until we get an actual 'made for VR game'. Also, there is still much work to do in terms of input devices, and the recommended specs will hopefully make it easier to implement new peripherals along the way. It's not just about being able to run a 'stylised experience'. Maybe raising the bar is actually the only way to avoid VR turning into another PS Move or Kinect.
Right now what makes me most skeptical is the resolution of the device. DK2 works fine most of the time, but the screendoor effect can be pretty annoying at times. Call me crazy, but if the resolution was good enough I could even see myself using a Rift for Photoshop work (I've already tested PS CS5 with Virtual Desktop and it works fine for what I'm doing but again, the screendoor effect makes it too much of a hassle). Oculus have said many times over that the Rift isn't just for gaming, and who knows how much horsepower it takes to render those VR movie experiences they are working on.
Edited 4 times. Last edit by Petter Solberg on 21st May 2015 8:23pm
I'm really interested in what the future holds for VR but I'm going to wait for a little bit to see what game support it actually starts to receive before I decide whether to go PC or PS4..
Still, this is very reassuring as I consider having an open standard a necessity for a healthy VR environment going in the future. Knowing all this has considerably increased my own hype for this. Who knows, maybe I'll get into VR earlier than I imagined
The VR effect is so powerful you don't need stereo vision, photo levels of rendering or even high resolution. All things that are nice to have, all things that come behind effective positional sound in importance. Going straight to the high end for everything is insanity unless you aren't interested in building a market. As I feared, Rift is turning into a rich man's toy.
The winner will be the one that ships an affordable solution that's 'good enough' at launch, that doesn't look like Occulus right now.
How exactly could Sony shine in comparison, when they've got a console that in most major releases struggles to hit 60fps, and has to be limited to 30?
The ONLY way they can get VR working well at 120fps on the PS4 is quite simple - developers will need to make massively parred-down or simple VR experiences for it. It's not an option this time to fob everyone off with "But 30fps feels more cinematic!" They ideally need to hit 120 minimum, and even dropping as low as 60 will render it unplayable. There's no magic button Sony are going to be able to press that enables their console to suddenly quadruple in power just because you plug a VR headset in. The only way it will work is by limiting the complexity of the games that run on it.
I suspect the difference will be so near to an order of magnitude from a GTX 970-equipped PC, that the class of games being played on each VR solution will essentially be completely different. You won't be playing Project Cars VR on the PS4... not unless they rack the graphics down to lego-block levels. I currently am playing Project Cars on the DK2, at a consistent native 75fps, and it's already stunning! That is going to be the fundamental difference between Morpheus and Oculus.
The other thing to note is - anything of the complexity that Sony will be able to get running at 120hz on the PS4, will run absolutely perfectly on the Rift on any middle-to-low-end cheapo gaming PC too. If all you want is to run simple VR experiences and games with all the settings set to Low, the Rift will work just as well as any competitor product. The recommended Rift specs are only there to give an idea of what you'll need to run major games at high graphical settings - a usage scenario that Sony won't even cover, because how could they?
I've said it for a long time, people are going to be shocked at the sheer power required to run a decent VR experience. It's not a failing of Oculus, it's just plain and simple fact. You either spend a LOT of money on equipment, or you massively compromise on graphical complexity. No VR solution is going to work miracles.
This backlash seems to just be a manifestation of exactly the shock I figured would happen, but as usual on the internet, if someone tells people something they don't want to hear, they're sure as hell going to stuff their fingers in their ears and start searching for someone else who will tell them something they do want to hear.
Would you switch to a different monitor if your games ran slowly on your current one? Of course not, why would a different monitor make your games run better? You'd do what you've always done... start saving for that next GPU upgrade!
Edited 2 times. Last edit by Dan Wood on 22nd May 2015 3:38pm
http://www.roadtovr.com/sony-120hz-2015-morpheus-demo-vr-bots-magic-controller-bedroom-robots/
I agree that the PC spec is sensible. You never know, if it proves popular enough and devs adhere to it (its only a recommendation) it may narrow the difference between PC and consoles. i.e. everyone knows when to upgrade and what to.
The best result for VR fans is if all these products become successful.
The thing is, if someone can prove that a core game can actually run quite well in VR due to the recommended specs, then maybe we'll see bigger games sooner than people predict. Though I hope to see made for VR experiences, I don't see why we can't see more traditional genres on the Rift because 'it just works'. I love all kinds of games, and I think for VR to actually work it has to be able to deliver variety. What's certain is that the novelty value will never be enough long term.
Edited 3 times. Last edit by Petter Solberg on 22nd May 2015 8:34pm
I do think async timewarp may very well end up being the thing that saves VR, in all cases, but even with it, there's still going to be a gulf between what a PS4 can manage and what a decent PC rig will be able to do with the Rift.
http://www.roadtovr.com/visualizing-latest-features-found-oculus-sdk-v0-6-0-0/
As for the thesis of this article, I find it a little flawed (the headline certainly is). Sony is absolutely focussed on making Morpheus PS4's ultimate differentiator. The VR community as it stands right now treats Morpheus and Oculus Rift as two very separate offerings for this reason. Fusing the success of one to the other isn't really a reflection of the demographics involved. Overall opinion of virtual reality and consumer desire for it will be the ultimate decider of whether the technology succeeds and this affects all the players.
Much more worrying up until recently, was that first-timer VR users and their first impressions, specifically ensuring that they didn't feel (too) physically ill when stepping out of the experience and put of the idea of VR for life. This was the last subject the mainstream press pronounced to be the death of VR and hey, we got past it (well, mostly).
The gaming PC market is booming, inversely to the traditional PC market. Gaming PCs meeting this specification are already considered mid-level and can be had for circa 5-600 quid right now. By the time the Rift is released, it'll be much less. There's absolutely no denying that the specs are probably more than some expected, but Oculus have deliberately been open and pessimistic releasing these specs. Palmer Luckey has said that in doing this, it gives developers a benchmark specification to target (in lieu of a fixed hardware platform like PS4). And lets not forget Valve, who will have Steam VR running on Steam Machine hardware by the end of the year, ready to roll out to under-TV cabinets, another blurring of traditional markets.
Ultimately, I expect VR to move from an initially niche, enthusiast product, to aspirational then to mainstream - in much the same way as flat panel HDTVs have progressed. Just how long all this will take if anyone's guess. But even a lacklustre launch for the Rift AND the Morpheus AND the HTC Vive wouldn't spell the death of VR. Sony, Facebook and Valve are all in this for the long haul.
Edited 1 times. Last edit by Paul James on 23rd May 2015 3:16pm
I've a couple of questions for anyone who can answer them. I believe the Morpheus does the re-projection in the headset. Does that mean they effectively half the bandwidth that goes down the cable? Secondly, do they introduce a frame of delay in order that they can read the frame buffer out twice at double the frame rate? (it might not be a whole frame but hopefully you get what I mean).
Edited 1 times. Last edit by Julian Williams on 24th May 2015 12:29am
1. They don't really care about graphics.
2. They don't care about what things cost.
There's just no escape from the facts that everything in a laptop costs more than it would in a desktop, and it's never as powerful as the desktop version because laptops don't like heat or battery draining parts.
Looking up the parts on TigerDirect, I get an i5-4590 for $240, a GTX 970 for $405, 8GB ram for $116, Windows 7 fpr $109, a 256GB SSD for $188, then say another $250 for motherboard and power supply and case. Those will all be cheaper in 2016, or 2017, or 2018, or whenever Oculus Rift finally goes on sale.
You won't need that spec to play Surgeon Simulator in VR or the majority of the Indie output. I'm keen for an Unfinished Swan port (I know that's PlayStation but still).
Edited 2 times. Last edit by Shane Sweeney on 26th May 2015 12:53am
"Recommended" specs are not "Minimum Specs". The framerates are going to vary wildly based on the speed of the machine and game-engine rendering the images.
Unless Morpheus is going to be released for PC and Mac and Vive/Occulus are going to be available for PS4, the idea that they are in competition is moot.
Even if you ignore the fact that the headsets run on different platforms and go for pure fanboy chest-thumping with price and features (not everyone is going to have or even considers having both a PS4 and a gaming-PC) it maybe that the Morpheus is cheaper but has both a lower per-eye resolution and has a fixed, upper limit on frame-rate & graphical fidelity - limited by the fixed specs of the PS4. This isn't PS4 bashing its just a fact.
PC hardware can always improve, PC GPUs get cheaper and more capable. The PC headsets may end up more expensive but they are higher-resolution and considerably more future-proof than the Morpheus.
Again this is moot because the Morpheus is a PS4 only peripheral and the choice is therefore down to whether you want your VR experiences to be based on the PS4 software library or the PC's.
Rather than get bogged down in 'who's is better' we should be supporting both markets 'spread the word' of VR.
Oculus is shooting for the stars but we all know that power does not equal sales. Content is king but what good is it if it is unaccessible to the masses.
Vive... is a wild card in my eyes. Maybe the middle ground but may be lost in translation just as you said with Valves way of doing things. Marketing will be its savior or executioner.
You can't afford for VR content to 'vary wildly' in framerate, doing so incurs artefacts which induce sickness. One reason why neither the Rift, the Vive or the Morpheus should be treated as just 'peripherals' - you have to develop from the ground up for the device and the best content will be likely be developed only for VR.