• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

New Oculus Prototype "Crescent Bay" - Lighter, Better Display, Audio, & 360 Tracking

1080p is still way too pixelated, 1080p doesn't work.

Still doable for HMZ but problem is FOV, if keep the same wide VR FOV then will pixelated.

That's why I thought eye tracking is better option (ie laser rear projection to your eyes), that go for extreme high resolution at the final end technology.
 
4k+ is totally possible if you go back to basics. The best demos in the new rift prototype today were not the hyper-realistic stuff, but instead the ones that were geometrically simpler and creative. Polyworld and this cool papercraft diorama thing were super compelling and geometrically simple.

I'd rather have high resolution and high frame rate than hyper-realistic worlds. That stuff will come, but I think the most successful stuff in the short-term won't be things like BF4.
 
People saying you need SLI 970/980's perfectly run games with this must mean with a lot of the graphical bells and whistles turned on. I'm sure at low-medium to medium settings, games will run just fine at 1440p with older cards. If they do start going to 4K displays, then I hope it's when Pascal hits shelves.

In concern of Morpheus... Will that be sold for PCs as well. I really don't see games on PS4 performing to well outside of games with simple geometry.

I mean overall I think games are going to have to regress a little bit in terms of aesthetics to really have these headsets perform great without crap frame rate, crap latency, an all the jazz.
 
In concern of Morpheus... Will that be sold for PCs as well. I really don't see games on PS4 performing to well outside of games with simple geometry.

Morpheus won't support PC. I guess it's possible someone could reverse engineer whatever the PS4 sends via HDMI to the breakout box and replicate it on PC, but would it really be worth the bother?
 
Still waiting on my DK2 to ship out but now with the announcements of Crescent Bay‬ makes me wonder if I should cancel my order and wait for the new dev unit.

Yeah... I know technology moves fast and all, but announcing a new, apparently vastly superior version when the old one's not done shipping?

Note that this doesn't mean I want them to not release CB, it means they need to step up their production :p
 
Yeah... I know technology moves fast and all, but announcing a new, apparently vastly superior version when the old one's not done shipping?

Note that this doesn't mean I want them to not release CB, it means they need to step up their production :p

It was absolutely no secret that they had better technology. They said they had better prototypes before DK2 shipped at all.
 
Why are people assuming this is going to be a devkit...they've repeatedly said that there won't be any more devkits before CV1.
 
4k+ is totally possible if you go back to basics. The best demos in the new rift prototype today were not the hyper-realistic stuff, but instead the ones that were geometrically simpler and creative. Polyworld and this cool papercraft diorama thing were super compelling and geometrically simple.

I'd rather have high resolution and high frame rate than hyper-realistic worlds. That stuff will come, but I think the most successful stuff in the short-term won't be things like BF4.

No question.

Super high-end 3D graphics are only a part of their VR equation. When you hear Carmack say panoramic video is Oculus' killer mainstream app, then you know they're not pumping the brakes because your 2D GotY with poorly retrofitted Rift implementation can't hold the required frames. If seen very little evidence that they're at all concerned with what you can't do from a horsepower standpoint, which will eventually self-correct, but rather creating the best VR platform they possible can. The VR gameplay will be diverging anyway, you might as well say goodbye to the current 2D graphics fidelity along with it. You'll get it back later.
 
I missed it probably but... did they announce any release date?

My SDK2 is being sent on October, but now I guess I'd rather have this one! If it's coming in like half a year or longer, I'll have the SDK2, but if this thing comes this year, I'd rather wait.


Edit: Seems they didn't announce any. I wonder when will this be releasing. Either way, I'll just stick to the SDK2 until most released stuff stops supporting it, then I'll just jump into the consumer version.
 
1080p is still way too pixelated, 1080p doesn't work.

I have a DK2, 1080p works just fine. I've yet to have anyone use it and not be impressed.

Sure, it's not as good 4k will be, but you have to be realistic when it comes to the kinds of computers that people actually own. How many people will have a beefy enough pc to run 4k VR games in a year (or whenever the consumer version is slated to come out)?. Not bloody many, I'd say.

Right now you probably can't even buy a graphics card that is capable of running something like Elite at 4k and 90hz, how is the average consumer meant to keep up?
 
Of this particular version? No. This is for developers. Once they finally show off the Consumer version, the discussion on aesthetics would then make sense.

No, the "mass market" is not part of the target audience for developer prototypes. Why is this question even coming up?

Pretty much. That's not what dev kits are for.

But you don't make incremental prototypes that are supposed to come closer to a final solution and then completely change it again for the final product. I know this is not produced for mass market, but if they completely redesign the consumer version this prototyping makes no sense.
 
But you don't make incremental prototypes that are supposed to come closer to a final solution and then completely change it again for the final product. I know this is not produced for mass market, but if they completely redesign the consumer version this prototyping makes no sense.

They're prototyping the most important aspects of the product...the virtual reality itself. Are you saying if they don't iterate on the plastic shell that houses all the technology, the prototyping doesn't make sense? I'm not sure what you're getting at here.
 
But you don't make incremental prototypes that are supposed to come closer to a final solution and then completely change it again for the final product. I know this is not produced for mass market, but if they completely redesign the consumer version this prototyping makes no sense.

They are prototyping the technology, the aesthetics are an afterthought. Why would they spend time and money on making the development kits look good, when huge portions of the specifications are subject to change. For instance, the lenses have changed shape in the new version.

These kits are primarily made for developers, there it absolutely no reason for them to worry about how they look. Not to mention it would make them cost more!
 
Still doable for HMZ but problem is FOV, if keep the same wide VR FOV then will pixelated.

That's why I thought eye tracking is better option (ie laser rear projection to your eyes), that go for extreme high resolution at the final end technology.

VRDs that project lasers on your retina is not an existing technology
 
They're prototyping the most important aspects of the product...the virtual reality itself. Are you saying if they don't iterate on the plastic shell that houses all the technology, the prototyping doesn't make sense? I'm not sure what you're getting at here.

They are prototyping the technology, the aesthetics are an afterthought. Why would they spend time and money on making the development kits look good, when huge portions of the specifications are subject to change. For instance, the lenses have changed shape in the new version.

These kits are primarily made for developers, there it absolutely no reason for them to worry about how they look. Not to mention it would make them cost more!

So if they are not prototyping for ergonomics why do the prototypes change in that area? Like getting smaller and more lightweight? To change the shell with every prototype obviously costs them more than just replacing screens and adding LEDs.
 
But you don't make incremental prototypes that are supposed to come closer to a final solution and then completely change it again for the final product.

What is a final solution if not the best thing they're able to come up with at the time?
 
So if they are not prototyping for ergonomics why do the prototypes change in that area? Like getting smaller and more lightweight? To change the shell with every prototype obviously costs them more than just replacing screens and adding LEDs.
They are prototyping for ergonomics, and weight for that matter, as well as all technical aspects of the kit. Because ergonomics and weight are functional parameters, just like resolution or refresh rate.
They aren't prototyping the aesthetic design, a non-functional parameter.
I still don't understand why this is hard to get. The picture of the devkit posted earlier is a perfect comparison. Sony didn't bother to make that look flashy either.
 
Obviously Oculus acquired the guys who designed the 360 controller so they can make a headset that looks like a pencil case with some straps and severed walkman earbuds

Aesthetics seem like a weird place to differentiate between the two products, if anything you'd point out that Sony has an established input solution if you want to find a favorable comparison.
 
But they still have a lot of time till CV1 release. I don't think that they will stay at 1440p for the next 8-12 months.
If that's the best they can do for the time being, they will. There are no 5-6" 4k OLED panels available anytime soon that we know of, so their hands are simply tied here.

The rest of the development time will probably be used to flesh out some of the more important useability, platform and content concerns about VR, which are lacking right now.
 
I may be wrong, but i think you can still run 1080p (not native) on 4k. You get great performance and it should look better than native 1080p display due the less apparent screen door effect. Screen door effect is the main issue right?
 
Screen door effect is an issue. I wouldn't call t "the" issue, real apparent resolution is also very important. So is aliasing, just listen to Carmack's recent talk at the panel about it.

Still, for me the huge thing is asynchronous timewarp. If you get that working at the driver level, you are decoupled from the need to maintain a hard 90 (or whatever) lower FPS limit. That's a huge relief. Really, its importance cannot be overstated.
 
Still so far behind in design in relation to Morpheus. Is good that a company like Sony, with so much design experience for electronic products is entering this market segment.

And wish to clarify that im not saying that Occulus will not offer a better VR experience, just speaking strictly in terms of the physical aspects of the device.

Dude, prototype/devkit.

Remember, they bought a multi-award winning design firm that made the 360 controller and the Arc Mouse to do the final design on this thing. It's gonna look sexy as hell when this thing comes out, and you know it!
 
CV1 wont have a 4k screen.

Apparently Samsung will be producing 4k display by the end of 2014 beginning of 2015 for the note pro and devices after the note 4, so I do believe the retail version will have a 4k screen. What everyone seems to forget is just because the CV will have a 4k screen doesn't mean you HAVE to run it at 4k......it's just future proofing.

Carmack has said the the CV won't be as demand as a traditional setup running a 4k display to maintain a solid 90hz refresh rate in the VRGear press conference.....god knows what wizardly they have going on under the hood

Why are people assuming this is going to be a devkit...they've repeatedly said that there won't be any more devkits before CV1.

They also said the same thing after they released the DK1
 
I may be wrong, but i think you can still run 1080p (not native) on 4k. You get great performance and it should look better than native 1080p display due the less apparent screen door effect. Screen door effect is the main issue right?
That's what I think. Now I don't know how much of an issue the screen effect still is at 1400p, but with DK2, IMHO it's what makes things like watching movies, a bad experience.

It might not be the main issue for everyone, even I agree that sometime you even forget it's it's there, but I would still call it one of the 2 biggest issue I have with DK2, FOV being the second one ( but that one might be something they won't be able to tacle until VR become big enough for them to have screen made just for them ? )

And again for some use, rendering at 4k @90 shouldn't be too much for our CG, a very basic cinema application, or some evolved version of Virtual desktop ( multiple virtual monitor etc ) don't take much power compared to a game, but would benefit immensely from a better resolution / almost no screendoor effect. And for me that kind of application is almost as important as VR gaming.

edit : not that I am saying that 4k will be there for CV1, I have no idea about it, every thing depend of what Samsung will be releasing next year, and when CV1 will be released ( my guess is Q3 or even Q4, and with a very limited stock at first ).
 
Outside of the 1440p(?) OLED, morpheus was revealed with the 360 positional tracking, dedicated stereo audio and the asymetrical lenses that are headliners for this update.

The flag waving seems pretty petty and immature. Sony/Oculus are well on their way to providing the best VR they can for their target platforms. That includes looking at the competitiors ideas and re-thinking some things that they previously may not have thought about.
 
Screen door effect is an issue. I wouldn't call t "the" issue, real apparent resolution is also very important. So is aliasing, just listen to Carmack's recent talk at the panel about it.

Still, for me the huge thing is asynchronous timewarp. If you get that working at the driver level, you are decoupled from the need to maintain a hard 90 (or whatever) lower FPS limit. That's a huge relief. Really, its importance cannot be overstated.

Yeah, I've now watched the talk and Carmack's description of asynchronous time warp has to be one of the most exciting developments for VR in quite a while.

Interesting that he specifically talked about team green and with their recent Maxwell VR announcements it looks like they're fully board with getting Carmack what he wants.
 
Some new info on the consumer model(not necessarily the Crescent Bay prototype) and some more things to strike off the worry list:

wZtpWgv.png
 
They also said the same thing after they released the DK1
No, DK2 was always planned. It was required because of positional tracking.

Yeah, I've now watched the talk and Carmack's description of asynchronous time warp has to be one of the most exciting developments for VR in quite a while.

Interesting that he specifically talked about team green and with their recent Maxwell VR announcements it looks like they're fully board with getting Carmack what he wants.
Yeah, there were even slides in the NV presentations a few days ago about asynchronous timewarp.
 
Outside of the 1440p(?) OLED, morpheus was revealed with the 360 positional tracking, dedicated stereo audio and the asymetrical lenses that are headliners for this update.

The flag waving seems pretty petty and immature. Sony/Oculus are well on their way to providing the best VR they can for their target platforms. That includes looking at the competitiors ideas and re-thinking some things that they previously may not have thought about.

The positional tracking in Morpheus is nowhere near as good as what has been reported for Oculus. Also, they have latency issues and a slightly lower FOV. So I'd say Oculus pretty much caught up and sort of stepped ahead.
 
The positional tracking in Morpheus is nowhere near as good as what has been reported for Oculus.
The tracking systems are basically the same, except Move/Morpheus is more robust, because it uses visible light instead of IR. That's why Sony don't need to cover their headset with dozens of markers.

Also, they have latency issues and a slightly lower FOV.
Morpheus has 30-40ms of latency, just like Oculus.
Last I heard, anyway. Does Crescent Bay significantly improve latency over DK2/CC?

Morpheus has a 90º horizontal FoV, but I'm pretty sure Oculus give diagonal FoV, which makes it seem larger than it really is.

So I'd say Oculus pretty much caught up and sort of stepped ahead.
I'd say they're more or less neck and neck. Oculus have adopted Sony's tracking and audio solutions, and Sony have yet to implement a low persistence display, but they're been using that tech in their TVs for over a year, so it seems like a fairly safe bet.
 
I seem to remember them saying that there wouldn't be a 2nd development kit and the CV would be the next release from them
That was back when a 1080p display was going to be their consumer goal. I think their initial investor funding and dealings with Samsung allowed them to set their sights a bit higher. I dont think they can keep waiting and waiting for something better than 1440p. And more to the point, I dont think there's any critical new feature implementations that developers need to get their heads around.
 
What purpose is Crescent Bay supposed to serve? If it's not slated to become DK3 or CV1, then what is it, exactly? Just a peek at the latest prototype to build hype?
Not that there's anything wrong with that.

I know Crystal Cove wasn't technically released either, but isn't DK2 basically the same as CC?
 
I think the (optional) headphone are a good move. I got a wireless set just shortly before getting my DK2, but they aren't impervious to some sliding around with all the head movement.
I'm not sure how useful these technologies are for VR - they seem to be about making framerate fluctuations more acceptable on monitors, whereas in VR what you really want is no fluctuation at all to avoid nausea. It seems like G-Sync/FreeSync's main benefit is to allow users to push graphics settings above what they'd normally use to get nice v-synced motion, because the frame drops don't appear so horrible. But they're still frame drops, and you don't want that in VR - you want high framerate all the time.
That's the ideal, sure, but real world there's no way every user of CV1 is going to be maintaining 90 fps all the time.
mugwhump said:
What, another prototype? I thought the next one was gonna be the consumer version?
They've gone through more prototypes than have been released as development kits. Here's an image they released showing many of them up to the DK2.
Majukun said:
what kind of requirements would oculus' software have?for 4k play with >30fps at the moment you need a gtx 980 or some high end amd card..aren't they pointing a little too high?I
For one, of course frame rate would very much depend on the particular software. For another, this is a case where rendering at a lower resolution and scaling up could beat showing that lower resolution on a natively lower res screen, simply because the less obvious the screen's individual pixels are, the better.
Daniel Jackson said:
I just hope the final version will be comfortable for people with glasses - me being one of them.
Yeah, that is a pretty big problem too. Contact lenses as gaming accessory isn't a great solution.
Quasar said:
And personally it makes me wonder if it will be out of reach for me. I can't afford to build some monstrous i7 SLI rig for gaming (I'm still happy with my 2500k/7870 based machine and i tend to move from one sub 200$ GPU to the next every 2-3 years).
I'm generally in the same boat. I'm using a base machine I got 4 years ago with a $200 GTX 650 Ti from nearly 2 years ago. DK2 is still pretty cool but imperfect, though there are still a lot of demos that hit 75 fps--it helps that these are from individuals or small groups who can't afford to model insanely detailed assets. I'm sure if someone were to spend the same money I've spent on this system today or a year from now when the CV1 hits, they'd be able to get a much improved result.
What purpose is Crescent Bay supposed to serve? If it's not slated to become DK3 or CV1, then what is it, exactly? Just a peek at the latest prototype to build hype?
Not that there's anything wrong with that.
Pretty much. If you're having a big VR development conference, you want to show off your latest thing.
 
What purpose is Crescent Bay supposed to serve? If it's not slated to become DK3 or CV1, then what is it, exactly? Just a peek at the latest prototype to build hype?
Not that there's anything wrong with that.

I know Crystal Cove wasn't technically released either, but isn't DK2 basically the same as CC?

That. The purpose is pretty much show where they are now and let "public" try it out and give some feedback/praise it in the Internet.
 
The tracking systems are basically the same, except Move/Morpheus is more robust, because it uses visible light instead of IR. That's why Sony don't need to cover their headset with dozens of markers.


Morpheus has 30-40ms of latency, just like Oculus.
Last I heard, anyway. Does Crescent Bay significantly improve latency over DK2/CC?

Morpheus has a 90º horizontal FoV, but I'm pretty sure Oculus give diagonal FoV, which makes it seem larger than it really is.


I'd say they're more or less neck and neck. Oculus have adopted Sony's tracking and audio solutions, and Sony have yet to implement a low persistence display, but they're been using that tech in their TVs for over a year, so it seems like a fairly safe bet.

Good points. Morpheus doesn't have integrated audio either it seems as all recent videos show headphones being taken off etc.
 
Just red the PCgamer piece on the new prototype, damn it sounds awesome. Wonder how many DK2 orders will be canceled... I'm thinking about canceling mine but keep thinking that order might start to move along much faster now.
 
Good points. Morpheus doesn't have integrated audio either it seems as all recent videos show headphones being taken off etc.
It's integrated in that there's a headphone jack built in to the headset, allowing 3D audio with any stereo headphones the user chooses to use. That's been there since GDC.
 
Just red the PCgamer piece on the new prototype, damn it sounds awesome. Wonder how many DK2 orders will be canceled... I'm thinking about canceling mine but keep thinking that order might start to move along much faster now.

I don't get it, what do you mean? Why would you cancel a developer kit because Oculus VR shows off a prototype?
 
The tracking systems are basically the same, except Move/Morpheus is more robust, because it uses visible light instead of IR.

I'm not into this "my father can beat up your father stuff" between OVR and Sony, but I noticed you writing that visible light is more "robust" than IR. Why do you think that? I mean, IR tracking has been a tried and tested way of positional tracking for years, it's pretty damn robust and there shouldn't be a difference for the camera as long as it can see the spectrum frequency (duh). ..And one big advantage is: the lights aren't visible :-)
 
I don't get it, what do you mean? Why would you cancel a developer kit because Oculus VR shows off a prototype?
Because everyone's assuming that this will be DK3 and up for pre-order next week. Completely ignoring that there is zero evidence to support such a claim, and quite a bit to prove it false. The most obvious of which is that there's absolutely nothing in Crescent Bay that requires developers have one in hand in order to develop for it, quite unlike Crystal Cove that became the DK2 (that added positional tracking).

It's my opinion, and the opinion of those that actually think these things through, that this is probably a prototype CV1, and is giving us an early glimpse of the kinds of improvements we'll see when CV1 launches.

I just got my DK2 a couple weeks ago, and I've joked about this new one, but I don't seriously consider the DK2 obsolete, nor do I think it will be replaced any time soon as the primary development tool for Rift content.
 
I'm not into this "my father can beat up your father stuff" between OVR and Sony, but I noticed you writing that visible light is more "robust" than IR. Why do you think that? I mean, IR tracking has been a tried and tested way of positional tracking for years, it's pretty damn robust and there shouldn't be a difference for the camera as long as it can see the spectrum frequency (duh).
Marks explains it to 4Gamer here. (Shitty English courtesy of Google.)

In a nutshell, "spectrum" is actually the key issue. IR doesn't give you a spectrum of colors to key off of. Since the IR camera's view is black and white, the only thing you can key off of is the luminance information. Since you can effectively only search the image for bright spots, that leads to a lot of false positives in your marker detection. Oculus compensate for this by using a large array of markers. This allows them to eliminate some of the false positives if they don't match a known position in the array of LEDs.

Sony avoid this by using RGB LEDs. Now they can key off of chroma as well as luma, and they can eliminate false positives that way. Since the LEDs are RGB, they can choose a color which contrasts nicely against the background, and can alter that color on the fly if the room conditions change.

..And one big advantage is: the lights aren't visible :-)
I thought we weren't supposed to worry about what a dork we looked like while wearing this stuff? :p

Besides, I think the lights make it look futuristic. :)
 
FOV appears to just be back to DK1.

After poking around, it seems to be that while the resolution is around 1440p,the *distance between the pixels* is significantly reduced, leading to a vastly superior experience. Seriously. Barely any screen door at all. It was amazing.

Great
I haven't gotten a chance to try the rift myself but saw videos taken from a camera looking through one of the lenses and the screendoor effect on dk2 was wayyyyyyyy worse than I had imagined.

This makes me a lot more excited for the cv1 again
 
Top Bottom