i think they rethinking slightly due to valve coming out with such a pin point accurate (if first impressions are to be believed) 1-1 motion tracking vr system.
Will this likely descend into VHS Betamax where people wait to see what product wins or will the market be similar to say racing wheel peripheral market where many products compete with different feature sets such as cross compatibility, resolution etc? If correct competition has made the oculus better but could we see it slowing sales till there is a clear winner?
I wouldn't be that surprised if Crescent Bay and the Vive Devkit were using the same panel pair (or integrated dual panel if they're on the same PCB/FCB). It seems unlikely there are two manufacturers both producing the same very specialised part with very similar specs (at least in terms of dimensions).
The HTC Vive really does look one impressive piece of kit. In pure spec terms its dominating. Oculus have just spent too long in what feels like perm development. Other company's have now passed them in specs and have given the consumers actual release dates.
In comparison to the DK2*. Crescent Bay, according to everyone who has tried both it and the Vive devkit, is directly comparable to Vive in terms of the HMD itself (display and head tracking), with only very minor differences in FoV and subpixel smoothing. Main differences being that Vive is marginally heavier and lacks integrated headphones. Lighthouse on it's own is very interesting, but I'm less enthused around the hand controllers. Hand-tracked physical (no finger-waving) input devices are pretty much the best we can do in a consumer-usable at the moment, but it sucks that if they come bundles with HMDs they'll pretty much be locked in as the 'standard VR input device' for a good few years, regardless of what else might turn up. At least both Sixense (STEM SDK) and Valve (SteamVR API) both want to make their interfaces device-agnostics, so we won;t be stuck with one implementation of hand-tracked physical controllers. *Which is, at a minimum, 14 months old (Crystal Cove public unveil). Wow, time sure does fly.
Comparing its specs to what the expected oculus rift will have its also miles ahead. Oculus needs to get there finger out and give the customers a firm release date not sometime soon.
Comparing specs with Wild Speculation of future headsets isn't going to give useful; results. Both headsets as demonstrated are at comparable stages of development, and very similar in terms of technical specs for what specs have been released (both 90Hz, both using dual OLED low-persistance panels, both with positional head tracking of similar working volume, both of similar resolution from a purely visual comparison). Both companies do not intend to release consumer models for quite some time (HTC Vive consumer release: scheduled for November, Oculus as yet unscheduled). Declaring one or the other to be 'miles ahead' when neither are going to arrive for consumers for over half a year is rather premature.
I like my DK2, but there is only one thing I would change - resolution. I still don't know if Crescent Bay 2560x is enough - if it has two screens - is it 1280x1080 per screen?
Crescent bay almost certainly has one screen with a resolution of 2560*1440. Therefore its res is 1280*720 per eye (vs 960*540 on the DK2). However, do remember image quality isn’t just about resolution. Things like Sub Pixel layout, and lens type are equally important for VR. Same as a cameras quality not just being about megapixels.
DK2 is 1920x1080 so 960x1080. Now Crescent bay IS 2 screens headset, Luckey confirmed it. There were speculations that it has one sceen 2560x1440, so 1280x1440, but now we do not know, cause they confirmed it has 2 screens.
1280x1440 per eye, you mean. True, but it's not the subpixel arrangement, but the subpixel count and the Fill Factor that matter. The DK2, Vive, and presumably Crescent Bay, use the 'Diamond Pentile' subpixel layout. This uses a sequence of RGBG subpixels (rather than the normal RGB), and defines two subpixels per pixel, rather than three, as the resolution. The latest Morpheus devkit has a 1920x1080 panel, but with RGB subpixels. Total subpixels: DK2": 1920x1080x2 = 4,147,200 Vive: 2160x1600x2 = 5,184,000 Morpheus dev.2: 1920x1080x3 = 6,220,800 Crescent bay: probably the same as Vive, but we don't really know the resolution or subpixel arrangement. Fill Factor refers to the amount of 'black space' between pixels, giving rise to the 'Screen Door Effect'. The higher the Fill Factor, the less pronounced the SDE. SDE can be partially countered by blurring the panel using a diffusion filter, but this has some unpleasant visual drawbacks depending on the degree of blurring and filter used (e.g. loss of contrast, dimming, 'grainyness', etc. Anecdotally (because nobody has taken shots of the bare panels with a reference scale for a truly objective comparison) people who have tried the Morpheus, Vive and Crescent Bay have reported that the Morpheus has a really excellent Fill Factor with little/no blurring, Crescent Bay has some sort of diffusion filter but a pretty good fill factor, and Vive has little/no filtering and a very obvious subpixel structure. Finally, those images of CRTs are not of pixel layouts, but of phosphor triad layouts. This is important because phosphor triads are significantly smaller than most LCD or OLED pixels for desktop displays (that image is scaled wildly incorrectly), and a single pixel on a CRT does not correspond to a single triad, but covers many triads. 4K 28" and 24" displays are now approaching a pixel pitch around the same as the phosphor dot pitch of high quality CRTs.
CB's been confirmed as having 2 seperate displays. Screw hole between eye peices. Dual display ribbons at the tops And the reddit thread They've had dual screens for quite a while, and were pleasantly suprised when no one caught on. Quite how they are synchronizing output - unsure, but I bet it's a vestigial from the early relationship with Valve. It's pretty good to know they've had this in their headsets for over a year, to know Oculus are ahead with the Rift, considering everything else has been silence. Thoroughly looking forward to what they have in store for E3
There's rumours going around that Facebook will take oculus more to the social side VR and focus on it and leave vive and Morpheus to gaming. ( was on tr news post so guess its a little more than rumours) would not be a big surprise though.
Would be a massive surprise, and completely opposite from everything Oculus (and everyone employed there) have said in the past. It would also mean pretty much everyone employed there jumping ship in disgust (which hasn't happened), so I really don't think it's very likely.
edzieba what do you think about the Wearality SKY? http://www.roadtovr.com/wearality-s...e-fov-vr-smartphone-adapter-kickstarter-live/
In the form of a 'skeleton' smartphone-holder? Pointless, not worth bothering with over cardboard. Buy a Gear VR if you want portable VR that doesn't suck, or save a pile of money and get Cardboard if you're happy with something that sucks for short-period demos. At the moment, anything that isn't GearVR is nothing more than Cardboard in an overpriced case due to the lack of a proper VR API for Android (massive latency due to compositing, no low-latency access to USB for an external IMU, internal phone IMUs suck ass). tl;dr The product they're kickstarting is a waste of money. The lenses themselves would be interesting a few months ago, but now it's become clear that both Oculus (with Crescent Bay) and HTC/Valve (with Vive) are both using composite lenses composed of two solid optical elements fused with a Fresnel mating layer (i.e. the interface between the two elements forms a 'third' Fresnel lens) of very fine pitch, the course-pitched facing air-gapped Fresnels are going to offer inferior optical characteristics (distortion and aberration, as well as 'stair-stepping' from the Fresnel pitch, and a 'lens flare'-like effect on high-contrast edges from diffraction in the Fresnel). Using only Fresnel lenses allows for an increased FoV though, as demonstrated with the InfinitEye and VRUnion's Claire 12M, both of which have a much wider FoV than Wearality's lenses. tl;dr The lenses are a middle-point between those in the Vive and (probably) CV1 and those in wider FoV Fresnel-based HMDs. Whether they are a 'best of both worlds' or 'worst of both worlds' solution would need objective testing from a third party with access to all these headsets, which is currently not possible outside of short guided demos. My gut feeling from using Fresnel lenses is that it will probably have more drawbacks than benefits.
