They have said it was a full Windows 10 computer, so I'm not entirely shocked by the $3000 price, but I do think that the consumer version will likely have to be cheaper to sell it to the average consumer.
They can define what "full Windows 10 computer" means. If Microsoft ported it, that $9 ARM based computer that showed up on kickstarter last year could be a "full Windows 10 computer".
They're trying to weed out early adopters to avoid dooming a product with known shortfalls that can be solved with product iteration and time (low FOV, tracking lag, etc).
Like what Google Glass is doing, except they seem to be making it less "glamorous" and possibly tethering it to a PC of some sort.
Honestly, I think they've defined pretty well what "full Windows 10 computer" means in this case: You are strapping a full Xbox One with Kinect onto your head. Last I checked, you can't run a Kinect very well on a $9 ARM chip. Based on graphics we've seen in demos I doubt the system hardware is anything less than we see in an Xbox One, but it's definitely not less than what we see on a Surface Pro, form factor for form factor.
It's possible that they are trying to weed out early adopters with a high cost, but with entire new hardware that's not been mass produced to date (the holographic display, the proprietary holographic processing unit) it certainly seems like this is probably more a reflection of actual production costs today. It's certainly long been a tradition of console game systems that the early development machines are an order of magnitude (as much as even 100x, historically) more expensive than the final retail hardware. (Which also historically has been a profit center for some console generations when they were charging the same development box costs long after the retail cost margin drops.) It's very much a case that I think the only real comparison is to the Xbox One and we don't have a lot of data on that as Xbox One never sold dev kits to the public, that's certainly something new with the Developer Edition of the HoloLens.
Also, for what it's worth I've not heard any reports of "tracking lag" (in fact just the opposite from articles I've followed to date, that the HoloLens seems to have "tracking magic") and I'm willing to believe that the adjustments of the device's FOV are as much for safety reasons as anything else (we humans use our peripheral vision for a lot of things when navigating the real world). Certainly the same journalists complaining about the FOV have also stated that they've experienced different FOV in different demos.
They've said that the first version isn't targeted as a consumer device. They initially see it as a professional piece of kit for use by, for example, product designers. (At least that's how it was presented at the New Zealand MS conference.)
Of course, then they show it off using Minecraft demos, recipe books, and a parent helping their kid in another city fix their sink. So sure, marketing it to consumers is the long game. But I'd guess that's some way off.
I guess they haven't ramped up manufacturing for full production yet? Small batch runs will cost more.
It may also be artifically high to stop so many people buying one for the fun of it. As they will (almost certainly) have limited supply they want everyone buying one to be serious about building an interesting application with it.
I'm not sure the narrow field of vision is that important for a dev. kit. The goal is to let people work out how best to produce apps for this kind of interface.
If they can't sort it for consumer release, that's a much bigger issue of course. My guess (and it's just that) would be that they'll be trying to improve field of vision for the consumer release, as from the reviews I read it was the one major negative people cited.
well i would be more inclined to agree if they didn't announce their product with full view (at least 120º) capability.
i mean, we all saw the videos from the announcement, and the "live" demos. and that was just pure lies no matter how you want to paint them.
also even if they didn't lie, don't we need the dev kit to have the same full capabilities as the final product? would you acquire an early dev kit for some new console, only to find out that it has less ram, gpu and/or cpu than the final product? how can we actually test it fully?
oh sure, in an ideal world the dev kit would have all the capabilities of the live consumer device.
But when you're developing a new category of device, that's not always possible.
For a very related example, look at the Oculus Rift approach to VR. they've been iterating on developer kits slowly improving the hardware while letting people feel out how to use that class of device.
Ultimately it'll be down to individual dev. houses whether it makes sense for them to invest at this stage, or wait till it's closer to consumer launch
> would you acquire an early dev kit for some new console, only to find out that it has less ram, gpu and/or cpu than the final product? how can we actually test it fully?
I guess for MS is important to have a good portfolio of apps before the product is launched to the public. This class of unpolished dev kits are not unheard in MS world, Do you remember when the dev kits for the XBox were Apple Power Macs?
The reason the narrow FOV is important is because right now, the FOV is so small that you can't see a person's hands and face simultaneously at a comfortable conversation distance. There's a huge number of applications where you won't know the suitability, or even be able to visualize/test if it'll work well w/o knowing a target FOV for a consumer revision.
"nice one hackernews, downvoting the most important question in this thread... "
We're less than an hour in and the OP question is the top comment... I'm not sure what's so wrong to you about the HN community here?
'narrow as hell' is your assessment based on what?
Is there some industry standard minimum field of view for AR glasses that this fails to meet? Have the people who tested HoloLens out who complained about the field of view being narrower than they would like said that it made the technology useless, or just that it limited it? Seems like you think it's a showstopper, I'd be interested to know why.
There's a lot of certainty here about the capabilities of a just-announced device that was only seen under strict NDAs before today. Can you give more information?
You don't need to be "read in" to know that they are using a version of the waveguide-based display [1] though it helps to be an AR developer ;). The hard limit on these is the shape and resolution of the SBG display system which maxes out around 40 degrees.
ODG uses a similar system and has the same FOV.
You need a wholesale new way of doing displays to get to a wider FOV.
I don't believe the device demo experiences themselves were under NDA, however recording devices had been restricted. Oliver Kreylos did a great writeup a while back on measuring the FOV: http://doc-ok.org/?p=1274
I would hazard a guess that the $3,000 price tag for this dev kit would easily pass $10,000 with a significantly larger FOV. I don't have a detailed understanding of the optical magic Microsoft is using for the display, but given how small it was in the previously demonstrated prototype, where cost was no issue, an economical way to scale it up is probably still a few years off.
Pretty sure they don't use any holographic technology and it's "just standard AR".
US only...meh. There also doesn't seem to be any university licensing program. Could be done quite a bit better.
They use a holographic waveguide, but it isnt like looking at real volumetric hologram, the actual image formed is still flat at fixed focus. But you do get stereo and parallax from head movement, the only thing lacking is variable focus/eye accommodation.
It's as much holographic as the Oculus (or other head-tracked HMDs), and the missing depth cue (accommodation) doesn't seem very important in practice. See here [0] for a good discussion on what is and is not holographic, in the context of the HoloLens, from an academic VR researcher.
Does anyone have any thoughts about the pros and cons of this over the Oculus rift for scientific visualization? It seems that it might be easier to deal with user interaction for the Hololens, but without playing with one, it would be hard for me to say.
The hololens's first big advantage here is that you don't have to loose sight of the other people around you while using it. A couple of people can sit around and see both each other and the visualization they are working on.
Secondly, you can see the environment around you. For science purposes this may not matter, but for industrial visualization, this makes all the difference in the world.
Yep, multiple viewers, including remote viewers, seems to be a big deal for Microsoft. Certainly that has been a big explicit focus of the JPL, Architecture, and Skype demos they've been showing. (It's also an implicit part of all their on-stage demos as well; supposedly the fancy camera rig is acting just like any other HoloLens and sharing the same scenes directly as the actual devices will in real time.)
43 comments
[ 3.1 ms ] story [ 118 ms ] threadI really wanted to get my hands on one of these, but it looks like I'll have to wait a little longer.
They're trying to weed out early adopters to avoid dooming a product with known shortfalls that can be solved with product iteration and time (low FOV, tracking lag, etc).
Like what Google Glass is doing, except they seem to be making it less "glamorous" and possibly tethering it to a PC of some sort.
It's possible that they are trying to weed out early adopters with a high cost, but with entire new hardware that's not been mass produced to date (the holographic display, the proprietary holographic processing unit) it certainly seems like this is probably more a reflection of actual production costs today. It's certainly long been a tradition of console game systems that the early development machines are an order of magnitude (as much as even 100x, historically) more expensive than the final retail hardware. (Which also historically has been a profit center for some console generations when they were charging the same development box costs long after the retail cost margin drops.) It's very much a case that I think the only real comparison is to the Xbox One and we don't have a lot of data on that as Xbox One never sold dev kits to the public, that's certainly something new with the Developer Edition of the HoloLens.
Also, for what it's worth I've not heard any reports of "tracking lag" (in fact just the opposite from articles I've followed to date, that the HoloLens seems to have "tracking magic") and I'm willing to believe that the adjustments of the device's FOV are as much for safety reasons as anything else (we humans use our peripheral vision for a lot of things when navigating the real world). Certainly the same journalists complaining about the FOV have also stated that they've experienced different FOV in different demos.
Of course, then they show it off using Minecraft demos, recipe books, and a parent helping their kid in another city fix their sink. So sure, marketing it to consumers is the long game. But I'd guess that's some way off.
It may also be artifically high to stop so many people buying one for the fun of it. As they will (almost certainly) have limited supply they want everyone buying one to be serious about building an interesting application with it.
nice one hackernews, downvoting the most important question in this thread...
i guess this is full of MS fanboys that don't mind being scammed out of 3k dollars..
As for the topic, they didn't specified the field of view, How could you tell they didn't fix it?
If they can't sort it for consumer release, that's a much bigger issue of course. My guess (and it's just that) would be that they'll be trying to improve field of vision for the consumer release, as from the reviews I read it was the one major negative people cited.
i mean, we all saw the videos from the announcement, and the "live" demos. and that was just pure lies no matter how you want to paint them.
also even if they didn't lie, don't we need the dev kit to have the same full capabilities as the final product? would you acquire an early dev kit for some new console, only to find out that it has less ram, gpu and/or cpu than the final product? how can we actually test it fully?
But when you're developing a new category of device, that's not always possible.
For a very related example, look at the Oculus Rift approach to VR. they've been iterating on developer kits slowly improving the hardware while letting people feel out how to use that class of device.
Ultimately it'll be down to individual dev. houses whether it makes sense for them to invest at this stage, or wait till it's closer to consumer launch
I guess for MS is important to have a good portfolio of apps before the product is launched to the public. This class of unpolished dev kits are not unheard in MS world, Do you remember when the dev kits for the XBox were Apple Power Macs?
That doesn't sound like a problem. Now if it turns out it is less powerful than the final product, that would be a serious issue.
and now they down vote me.. lol whatever guys go crazy!
Is there some industry standard minimum field of view for AR glasses that this fails to meet? Have the people who tested HoloLens out who complained about the field of view being narrower than they would like said that it made the technology useless, or just that it limited it? Seems like you think it's a showstopper, I'd be interested to know why.
The limitations of their approach are clear and well understood.
ODG uses a similar system and has the same FOV.
You need a wholesale new way of doing displays to get to a wider FOV.
[1]http://www.displaydaily.com/articles/446-sponsored-articles/...
edit: Please don't reference Magic Leap as having a solution to this - I'm super tired of discussing them.
Pretty sure they don't use any holographic technology and it's "just standard AR". US only...meh. There also doesn't seem to be any university licensing program. Could be done quite a bit better.
And Canada :)
[0] http://doc-ok.org/?p=1172
This is basically using the free market to sort out who really needs the kits.
Whereas the Rift dev kits could, and still can be, Ebayed for hundreds more than their purchase price.
Oh, great!
>You are a developer in the United States (including Puerto Rico) or Canada where the Development Edition will first be available.
>Please note that the Development Edition hardware and experiences are in English only.
Oh.
Secondly, you can see the environment around you. For science purposes this may not matter, but for industrial visualization, this makes all the difference in the world.