Hololens Round Two: Augmented Reality at Build 2016 | Hacker News Reader

Hololens Round Two: Augmented Reality at Build 2016

anandtech.com

49 pointsyread13 comments

I had a chance to play with HoloLens for a few hours yesterday and it was a really incredible experience. Yes, the FOV is noticeable and I hope they can address it, but I felt like I was able to easily compensate for that with the visual and audio cues the device can show you.

And speaking of audio cues... I found the sound to be a truly magnificent aspect of the hardware. There are two orange speakers that sit above your ears that are somehow able to produce some really great special sound effects. If something wasn't immediately in my FOV but it made a sound, my mind was actually tricked into thinking it was part of the ambient sound in the room. It was almost disconcerting to remove the headset at that point and have the sound disappear.

It is a little disappointing they were unable to fix the field of view issue in the shipping units. That was the biggest limiting factor between the type of amazing demos Microsoft has shown and my personal experience with the device at last years Build. Maybe that will be something that is eventually fixed before a consumer release, but until it is this device is more of a glimpse into an exciting future rather than a game-changer.

I read somewhere that with the technique they're using for the displays, it's very difficult to get a wider fov, and it's continually been a sore point for Hololens.

Despite this, I still think Microsoft is lightyears ahead of the market in the AR space due to the work and research that's gone into Hololens all the way down the stack. They've already made custom silicon to hit their sensor fusion -> display latency requirements and they're making W10 for AR.

I read somewhere else that some of the FOV tuning/shrinking has been for safety concerns as well. Given the "truly untethered" experience, I can certainly understand if Microsoft is trying to err on the side of caution versus accidentally blocking some of the important sideband navigation information people get from peripheral vision, while they wait for more science and research to be done on such things.

> I read somewhere else that some of the FOV tuning/shrinking has been for safety concerns as well.

This is very wishful thinking on the part of wherever you read that. Microsoft would love to expand the FOV, it's literlaly the number one bit of feedback they get on the product, but the current optics design simply won't allow it.

it's very difficult to get a wider fov

It's essentially impossible with the approach that all of the public AR HMD's are using. Not only that, this approach does not allow for accommodation which is arguably one of the most important things needed for convincing HMD AR.

Could someone ELI5 why FOV is such a difficult problem to solve?

It really bothers me that in all of their demos it's this room filled with AR animations, making it seem like you could see all of that through the HoloLens. When in reality, you will be seeing something like a 35-40" TV in front of you. You won't be "surrounded" by all of that stuff.

At least, in the short demo I got to play with, they've got some software tricks that deal with the field of view issues. Cues at the peripheral do a decent job of signaling where the user needs to direct the device to find content.

It's certainly nothing like looking at a flat screen projected. Artifacts have state in the room space, just aren't visible to you outside, of the admittedly small, window. I was impressed,YMMV though.

Here's the basic crux of my outlook as a general principle, well stated:

>The use cases for AR seem, to me, to be not necessarily the same as VR and both should easily be able to co-exist.

My instinct is that one will be far more adaptable and incorporated into entertainment / business / activities than the other, and at this juncture, I'm pretty convinced it'll be the AR track. It seems more dynamic and this review, on the whole, feels a lot more positive than ones noting the current state of VR (e.g. disorientation). That might change over time, but if this is an indication of things to come, I'm anxious to see what AR can drum up for use cases and software (yes, I did like that Minecraft demo a while back and dream of what a DAW could be like).

In the future, I imagine a single device that can do both. In that world, VR is just AR that occludes your entire vision. Then you use AR when you want to stay in your current environment, and VR when you want to be transported to a very different environment from the one you are physically in.

The HTC Vive has already a front facing cameras, so there you are: a VR headset which can do AR. Of course with only one low-res camera, it's very poor AR. And it's tethered.. When most people talk about AR, they're talking about untethered AR..

The tethered --> wireless --> autonomous transitions won't be easy though (probably needs to track the eyeball to reduce the power needed).

That's something they're working on with CastAR. They're developing an attachment which clips onto the front of the glasses.

http://www.castar.com

Reply on news.ycombinator.com