-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
Oculus Rift Kickstarter [Ended, $2.4 million funded]
- Thread starter Lord Error
- Start date
Good interview and great reactions - because he was less technically savvy and knowledgeable, Hiphop gamer comes off as something off an industry outsider...
Like this is the kind of reaction you can expect from the large game playing public that doesn't sit on forums like neogaf all day discussing the shit out of technical details all day - but simply just experience games.
That is to say, yeah, VR is going to blow people away. They'll realize soon after trying it that *THIS* is what's been missing from gaming for years and years.
And it'll be a totally obvious thing in retrospect. Like, why has the industry invested so much time money and effort into so much shit (can you imagine the amount of money Nintendo has invested in that new tablet controller of theirs?) when this was just sitting there, waiting for a small start up to do right? It is the one device that adds more to the graphics and sense of immersion than any other paradigm possibly can.
The one good thing for MS, Sony, Nintendo is that VR synergizes extremely well with their various motion control technologies. For all we mocked them this gen, when you pair it with the tech that allows you to be *in* the game world, where things are represented 1 to 1, motion controls will make immediate and obvious improvements upon the gaming experience.
I think Oculus will have to team up very tightly with one of the prospective motion control companies if they really want to be the next big thing in the computing industry - rather than just the guys that got the VR ball rolling. You really want to capture the whole innovation and design chain to getting things done (controls, interface, design language, OS), not just a single fungible part of it (i.e. other companies can develop alternative display and head tracking solutions and undercut Oculus).
Smartest thing about the rift IMO is the optics. I hope they've patented the model of using post processing to correct for lens distortions. That's a big step towards allowing for large FoV with light weight.
lowering latency across the inputs and outputs is very important but relatively easy to do.
Last big step is position detection that is accurate and doesn't drift. I still think some kind of camera based simple image processing could deliver that
lowering latency across the inputs and outputs is very important but relatively easy to do.
Last big step is position detection that is accurate and doesn't drift. I still think some kind of camera based simple image processing could deliver that
If they have, I hope it's only because they don't want someone patent squatting on it.
This is the thing that really shouldn't be patentable. These type of optics have been around for quite a while if you listen to them talk about the history of such devices. What's old is new again. The difference is today, computing is powerful enough to invert the image on the fly without issue. These are the type of things, that in an ideal world, would not be able to be patented. The evolution of technology should not be patentable IMO.
Agreed 100%. It's sad how nowadays it seems the patent system is destroying far more innovation than it is protecting.
"God's Beard!"
Member
Because it costs 300 bucks and only works for one person.
zephervack
Member
Can you imagine seeing someone sitting on a couch playing Portal 2 with the Oculus plus that Razer motion controller (the one that looks like the Wiimote).
Just like 80's movies showing games from the future.
5 years down the track, you're going to laugh at this comment and see me as the sage I goddamn am.
That's the first time I've seen a hiphopgamer video. Is he always this enthusiastic about stuff?
Because he really seemed genuinely blown away.
5 years down the track, you're going to laugh at this comment and see me as the sage I goddamn am.
I second this!
I like taking it one step further and note that VR makes a strong case for powerful local hardware in general. Streaming will be even less appealing if a compelling low-latency peripheral takes center stage.
Arstechnica posted a review. There really isn't any new information, but there is a link to a review of a $2,000 1080p VR headset that another company showed earlier in the year, which kind of shows what the Oculus could be for the retail version. Hopefully costs are going to come way down in a years time.
The previously shown headset never turned into a real product, and there is no software support for it in any case, but it's still cool to see what a 1080p Rift might look like.
Oculus plans to do a second manufacturing run once all of the pre-orders have been fulfilled. That would likely be late December at the earliest.
The previously shown headset never turned into a real product, and there is no software support for it in any case, but it's still cool to see what a 1080p Rift might look like.
Oculus plans to do a second manufacturing run once all of the pre-orders have been fulfilled. That would likely be late December at the earliest.
Augemented Reality glasses and VR are entirely separate goals. Augmented reality is basically supercharging your brain so you have unlimited info with you as you walk through life. It's not even remotely attempting to transport you to another world.
If people in here have the same reactions as the stories of late, I won't make the mistake of passing on it a second time.
Naked Snake
Member
These are my thoughts too. I'm surprised/skeptical how nobody who has used it complained about the low resolution, or the quality of the LCDs, or other issues for that matter--But I haven't really been following this thread closely, so I might be have missed some negative impressions if there were any.
And I'm curious about comfort issues... anyone know how it compares to the HMZ-T1 in that regard (comfort/weight)?
The rift weighs far less than the HMZ-T1/T2. I've read/watched maybe twenty hands on impressions, and no one has ever commented that it felt uncomfortable or heavy at all. Oculus has stated that a lightweight design is one part of making the immersion work, so that is unlikely to change for the retail model.
As for people giving negative impressions, there have definitely been some. The screen is low res, and there's nothing you can really do in software to make up for it. Even though the lenses cause more pixels to be in the center of your vision, things are still going to be kind of blocky, and you may be able to see individual pixels/the pixel grid. There is also an issue with your brain not liking the fact that it is receiving input from your eyes about movements that your body isn't actually making (for instance, if you use the thumb stick to turn, your brain doesn't accept that you've actually turned, which can cause nausea).
I don't think Oculus did enough to stress the fact that this is a developer model that is feature incomplete compared to the planned retail version. That's probably why a lot of reviewers are glossing over the fact that the screen is low res; it's going to change for the product actually marketed to consumers, so complaining about it now doesn't really help anyone. As far as Oculus goes, they needed the publicity from the kickstarter, but they really didn't need people to buy the developer version and then complain about shortcomings that they introduced into the design for the sake of making it as cheap as possible (so as many devs as possible could use it/build up game support for the retail version).
Occulus have been very careful to try and manage expectations. But the press are not putting the headset on and going 'oh, yeah you're right, its pretty low res'. They're going 'oh shit wtf bbq this is teh awesomes!'
So I think it should be pretty good even with the caveats mentioned.
Hopefully the brain has a way of just accepting the visual input it's being given, and kind of not focusing on any imperfections. Once the retail version is out, that shouldn't even be an issue. At that point, we'll be complaining that our rigs can't keep up 60fps in SBS 3D mode.
Naked Snake
Member
I have a feeling/hoping that could be the case... Kind of similar to how one can still get very immersed playing very old games today, despite the graphics and mechanics being very dated. After the initial adjustment, if you're "into" the game, any game, the technology tends to fade into the background.
And yes I'm fully aware that this is the developer model, and I'd much rather have the consumer model, but if impressions continue to be this good after December, it would make the wait for the consumer version pretty difficult for folks like me.
Feels bad man.How would it feel if my PC or console could deliver only 30fps gameplay stream to Oculus?
But seriously, with the focus they are putting on latency it seems like 60 Hz is really the bare minimum you need. Carmack always sounds like he'd much rather have 120 for VR in his interviews.
another reason I'm relatively happy with a low res screen. I only have a laptop right now which should play 1280x800 at 60fps with some tweaking of details (depending how intensive the distortion filter is). For the consumer version I'll probably need to build a beefy PC
I would be happy if consumer version of Oculus received 1680x1050 or even 1400x900 screen. Anything larger than that will be really bad for the gamers with midrange PC configurations.
It turns out that Hawken may not launch with Oculus support. The dev kits may not even be shipped out by the twelfth of December anyway, but it sounds like we all may be getting a lot more Doom 3 time in than some of us may have planned.
I think the retail screen will end up being 1080p, just based on the availability of 1080p screens. People have talked about Oculus maybe using that 2560xWhatever 6" screen from Sharp, but I don't know who is going to have a rig capable of properly driving that in the near term. 1080p would be 960x1080 per eye, which is roughly twice the pixels of the current screen; even driving that (at 60fps) is going to take a pretty hefty rig.
I think the retail screen will end up being 1080p, just based on the availability of 1080p screens. People have talked about Oculus maybe using that 2560xWhatever 6" screen from Sharp, but I don't know who is going to have a rig capable of properly driving that in the near term. 1080p would be 960x1080 per eye, which is roughly twice the pixels of the current screen; even driving that (at 60fps) is going to take a pretty hefty rig.
Unless I'm thinking about this wrong, I wouldn't want to look at a full FOV image that has been upscaled. Scaling would also introduce some latency, wouldn't it?
thats ok, I've never played Doom 3. I did wonder about Hawken - even getting early kits to help patch in support, they'll need some time to test on final dev units. Probably shouldn't be a big delay though? Bit silly/dodgy of them to pronounce loudly on the kickstarter page that they'd have day one support. And for Occulus to put that up on their site you as a viewer assume they are getting pre-dev hardware to assist.
In terms of resolution - 'just enough' would be enough. I don't want them going crazy retina or anything, or silly resolutions that need hefty SLI setups to drive. Just enough resolution for most people to go 'hey that looks ok' and then gets sucked into the world. I don't know if 1080p is enough or not, we'll need to wait for more impressions
Yes. And not that I know exactly, but it doesn't strike me as something that would add a whole lot of latency.
Hawken having day one support never really made any sense to me either. I think the publisher saw an opportunity for some quick publicity, and decided to go for broke with the whole "Hawken will launch with Rift support" line. I have no doubt that the game will launch with support for the visual warping that the Rift requires, but head tracking and the "custom cockpit" they touted will probably come a bit later. I wouldn't be surprised if the Rift compatible cockpit cost money either (the game is F2P with microtransactions after all).
I think 1080p is a good spec to hit for this gen. The hardware exists to push that at 60fps in SBS mode, and it's not too overly ambitious. If that's not enough resolution, I guess they'd have to wait for the next gen of GPU's before switching to a better screen.
I'm trying to imagine how it would work in my head. The image is stretched to the display, and then warped...or warped and then stretched to the display. Either way sounds like it would introduce visual oddities. If it's a possibility, I'm sure they'll figure it out though.
Regarding latency, for a modern GPU scaling an image is really something that takes maybe a few dozen microseconds. It's negligible.
I think they should go as high in terms of resolution as they can, and just rely on scaling for low-end systems.
I think they should go as high in terms of resolution as they can, and just rely on scaling for low-end systems.
Not really. You could even do the distortion and scaling in one step. GPUs are terrifyingly good at sampling images.
I'm kinda surprised. I thought that they might've already had a lot of the code included in the game - i.e. they have a head look button, as well as torso look - like in Mechwarrior Online.
That said, maybe they've had to go in and flesh out the cockpits (maybe built it only for facing forward views). If they're going all the way, they'd include HUD elements into the dials and monitors of the cockpit as well.
HUD elements shouldn't require that much more work... but maybe they'll have to redraw them to get them looking right (i.e. it's not good enough to just reposition and resize them).
With Sharp circling the drain (according to an Engadget story I read), I doubt we'll have to worry about very high resolutions in the short term. That is, unless they switch to micro displays anyway (since no one else seems to be making a screen higher than 1080p anywhere near the necessary size of the Rift's screen).
It would be nice if one SKU could accomodate people with mid-range systems, and people with high end systems at the same time.
I'm not sure if they're doing the coding for the head tracking themselves, or if Unreal Engine 3 is going to have support that they'll then use. In any case, I hope they can get it sorted out fairly soon after launch.
4-5" 1280/800. Carmack mentioned that iphone4 panel has a little better performance [faster pixel switching] than the current one in OR.
IIRC, the problem with phone screens and such is that they aren't driven in the same way as the type of screen inside the Rift (which is sourced from an ultra mobile PC screen, the likes of which no one really makes anymore). I guess that means higher latency, or higher cost? I'm not really sure.
Palmer luckey repeatedly mentions the advantage of phone screens and phone tech in helping making the necessary parts of this tech cheaply available for VR.
I can't imagine what practical differences a mobile display vs a phone display there'd be - in terms of driving either panel.
I'm really just parroting what I've read on MTBS3D in the past. Maybe it has something to do with the fact that they have integrated touch screens now?
I really shouldn't speculate, since I don't know for sure. It just keeps the misinformation train going.
Most screens still have a separate digitizer (the touch sensor) but I haven't looked in to the retina displays and how they implement it. If they can get a deal going with the manufacturers it should be fairly straightforward to get a display without the extra stuff like the digitizer.
Yeah, I'm getting tired so I'm probably mistaking one issue with phone displays for another. I know that the race for higher res small displays is part of what has made the Rift possible, and I know that there is some reason they're using a UMPC screen instead of a phone/tablet screen, but I can't remember exactly what it was right now. Maybe this was the best display in the price range that they could use without having to custom order screens?
Thinking about it, I'm probably remembering a discussion of why phone and tablet screens weren't good for the developer version specifically.
They are using I think 240hz sensors, which build up errors only after prolonged shaky use. Its not good enough for Carmack, but its good enough for us.
Out of all the first hand accounts I've read/watched/etc, I've never heard anyone mention that there was any noticeable drift in Doom 3 during demos. Whether John Carmack has done that through software, or if it's due to the hardware, I have no idea.
It's called GPU scaling for the low end cards. And thankfully they're not taking the approach of some of you guys with wanting a low resolution screen in the consumer version (which is why I doubt we'll actually see the retail version till 2014).
1080P is extreme now? A lowly GTX 460 could handle 90+ percent of games at 1080P/60fps. That's hardly what I'd call extreme. Now you've got the new cards. And by the time this thing is released, you'll have at least another new series of mid-range cards (and possibly another set of new architecture cards).
If you absolutely need to play Xbox3/PS4 ports at 60fps, either turn a couple settings down, or instead of budgeting 800 - 1000 for the headset, add an additional 300 for the latest and greatest mid-range card. By then, it should have no trouble running everything at 2500/60fps.
EDIT: Not that I think that Sharp will ever go into production. If they could get a 6 inch 1080P panel that could be driven at 120hz through firmware updates, even that would seem dream scenario. I just don't see anyone releasing a 6inch panel with such a high resolution. Although maybe Apple will eventually release a 6inch tablet and get into a pissing match with Amazon. One can only hope.
60 FPS is tough to hit at 1080p in current games. I have a GTX 580, and even that isn't enough to secure 60fps when all the bells and whistles are turned on in games like BF3, GTAIV (with mods) etc. That's without SBS as well (though I can't claim to know how much of a performance hit SBS 3D is at any given resolution).
I think they'll have to make some tough choices between going all out, and going for a product that will actually have wide appeal. If the device doesn't work in the next gen consoles, it's basically a niche PC device that relatively few people will ever use. If the device is incredibly expensive, no one but the most hardcore PC enthusiasts will buy it, no matter how good it is.
Edit: Thinking about it... GTA IV with IceEnhancer mod and first person mod + Oculus Rift... homigod.
You picking out one or two games. I'm saying generally speaking, most games have zero issue. Games like Dead Space 1/2 run at a locked 1080P/60fps using about 30 percent power of the GTX 460.
And although next gen systems are coming, something like a 660ti is twice as powerful as a 460. And by the time Rift comes out, they'll likely me mid ranges on even newer architecture.
SBS really doesn't matter in this case. It's not like it's using shutter glasses. It may be side by side 3D, but it's still only a 1080P screen. So it's not like it needs to output 3840/1080 to create a 1080P/60fps image and compensate for the shutter glasses. It's simply gonna result in cutting the 3D resolution in half.
plagiarize
Banned
it WILL drift, but you aren't going to really notice it, because you'll also be using the analogue stick to turn. so if it drifts off a little bit, it's as simple as tapping the analog stick to 'recenter'. using the HMZ my head drifts a lot before i notice that i've drifted and have to 'reset' by turning my head. i don't imagine this will be much different.
Palmer keeps this thing on his head for 10 hours at a time. i'm sure it took some getting used to to get there, but I could never get used to my HMZ for that long.
SBS means that the image has to be rendered twice, doesn't it? Or is it a matter of rendering once, and then generating the offset frames from that one...
I'm going to stop now. My forays into 3D have mostly been sticking a disc into a blu-ray player, or else enabling 3D vision support. I've never had to screw around with SBS
except for some uh...well, you know, the kind of SBS material one finds on the internet
With the new console generation, we should stop getting these console port games that are underwhelming compared to games made for the PC. That may mean that more games are difficult to maintain 60fps in at 1080p (like you mentioned). I guess we'll just have to hope that hardware has improved enough to the point that it doesn't matter. Ideally, you wouldn't have to sacrifice too much visual fidelity for immersion.
I guess it doesn't matter that it's drifting if you're basically recalibrating it with the sticks constantly. The important thing is that no one has complained about it at this point, so hopefully it's not an issue.
for the rift, wouldn't you really want 60fps locked? Dropped frames or tearing would be really bad and ruin the immersion.
perhaps games with 'rift' modes could use dynamic details - a bit like wipeout on PS3 can change resolution on the fly to keep the framerate up
No kidding I'm just waiting for Apple to get a patent for patents...
Similar threads
- Locked