Microsoft HoloLens [Virtual Reality, Augmented Reality, Holograms]

As I said, the actual game on actual holo lens doesn't use assets as dark as the ones in the video. Which only adds to my point that it was deceptive in the first place.

 

Given that the consistency of the color (and the perception of opacity) is due to the intensity of the projection relative to ambient light, you also can perceive significant color shifting in rendered objects that are overlaid on top of comparably strong colored ambient light (non-white interior lights, reflected sunlight off colored surfaces, etc.) Although I suppose that the brain never truly perceives color in an objective way (it performs its own white balancing of sorts by contrasting different neighboring colors), so the opacity and color presentation may be difficult to capture even with traditional cameras and displays, especially as the exposure and dynamic range won't adequately represent what the eye/brain is experiencing.

I'd still prefer traditional cameras mounted inside the HMD to the composited shots however as the latter completely bypasses the display technology for which the device is named. When you think about it that way you realize that you could stick the SLAM technology from Hololens onto any AR or VR HMD and it would produce identical composited videos, which is why none of the those adverts are particularly interesting to me. The videos can, at most, serve to showcase the tracking/mapping, but judging the performance and precision of that from a couple minutes of video is problematic - there's no way to know whether the animated visual representation of the mapping is reflective of the regularity and precision of the actual mapping or how robust the mapping is.

I suspect that we're just going to have to get used to the idea of the PR and tech presentation of AR/VR playing a bit fast and loose given that it's still very much an idea of the future being sold to investors and developers rather than a product being sold to consumers. Maybe once it gets closer to being a true consumer platform we'll have to move back to real world demonstrations - you can't exactly differentiate between your display tech and someone else's if all your demos are bypassing the displays.

 

Thanks Shifty & Milk for ruining another thread, back to Reddit I go. Spending less & less time here. Have fun bashing everything to kingdom come.

Tommy McClain

 

this is due to the cell phone camera and not the hololens.

Its just like when you take a picture and look at the picture and the actual thing you shot and both look diffrent

 

Watching a lot of vids on youtube from devs, seem pretty positive. I hope I get to try a demo soon.

 

Ok guys. I was wrong. Holo lens can diplay dark objects over a white wall. It can even draw pitch black objects over a sun-lit window. That video is perfectly representative...
Way to miss the point.

 

The actual 3D composite there is pretty awesome at the end.

I don't know if you realise 'deceptive' is the wrong word. Deceptive implies intention to mislead. The word I'd use is maybe 'inaccurate' or 'non-representative'. It's a developer showcasing his work, not trying to deceive people.

Why's it ruined? Discussion is about a display tech and its possibilities and limitations and understanding it. Isn't that exactly what B3D's core reason d'etre is?

In fact, what else are we supposed to talk about in the 'hardware and displays' forum?! Hololens is a bleeding-edge display (coupled with best-in-class motion tracking/3D scanning) that should be discussed in full. And why are milk and myself the bad guys when others are also contributing to this line of discussion by presenting counter arguments?

 

Would you mind linking that thread? I wouldn't mind reading the discussion there as well.

 

If the display was able to black out pixel behind the lit pixels; then it would not appear as a hologram anymore; also there could be a disconnect because you have an 'imperfect' possibly slightly aliased object in view. Which seems real because it's just there. But also not because of the resolution and or focus disconnect for one.
If it's a bit transparent then you accept is as a hologram; and it will be better for the end user.

So for AR purposes; having a transparent display actually seems preferable.

 

The through-the-lens videos of various AR products have exposed the solution which, at least for this generation, is to significantly overpower the real world and use brute force contrast.

We know about the limitations of a purely additive AR which is the only thing that exists today (outside of that sony patent with no product yet, and magic leap claim they are working on it but will face a major hurdle mixing it with a light field). We see this problem exhibited even in the Magic Leap demo, the SEER 100 degrees FOV demo, the leaked Hololens video, the Meta 2 AR, the CastAR product, Vuzix AR, Optivent Ora-S, Sony AR SmartEyeGlass, Epson BT300, etc... We even see this with $10,000 military AR. And all of them are using brute force contrast as a solution because they are all additive.

TL;DR
Are we allowed to discuss basic physics? Or is it considered bashing MS? The repetitive and misleading compositing with alpha blending deserves criticism because it's complete BS. No current AR product can do that.

 

I know I've said it before, but ... darken a layer on the visor to increase contrast. Do it by whole visor or by area. You know where the iris is, you know where the image is being projected, you know how bright external, contrast sapping sources are. You can calculate what needs to be darkened and by how much.

Still won't be perfect, but I suppose what is?

 

I don't know that Hololens needs to be that. It's first gen tech. So yeah, full opaque geometry display would be awesome, but for now accept the limitations and use Hololens as a neon display (and explain it as such!) and create apps that acknowledge the limitations. So no virtual Godzilla overlays except in night-time play. As Iroboto says, think of the display as a portal localised projector in terms of what it can draw into the world. If the idea would work as a projection, it'll work in Hololens.

 

Yes this is my number 1 bugbear, when something is marketed as not what it actually is.
1. Don't do it (preferable)
2. If you do it, add a plain disclaimer

I've had arguments here over the years cause I've complained about deceptive (and outright BS) marketing/statements and for some reason some ppl think its OK, "cause marketing is meant to be lies, just accept it"

 

Maybe the general thought is after you mention it a handful of times it becomes ad nauseam and should stop beating the dead horse or create your own thread to flog the same horse corpse.

 

Never use the word "deceptive". It ruins threads.

 

It does get boring harping on the same point over and over again, but if videos are being posted that aren't actually using the display subsystem of the device then options are pretty limited for what can be observed and unpacked for discussion. It would seem to me that discussing how the composited video is different than actual through-the-lens imagery is far more substantive both from the standpoint of what current tech offers as well as what/when we can realistically expect in the future. What AR can't do is just as much of a defining attribute as what it can do, so things like additive waveguide projection and FOV limitations warrant airtime to reach some sort of consensus on what "AR" actually means for the near future. In the same vein I would say VR's shortcoming is its perceived resolution which I'm sure I've annoyed more than a few people about over the past couple years, but anytime particular content is discussed or a screenshot or video is posted it feels wrong to avoid remarking that wide FOV HMDs have a way of turning even the most refined and intricate looking renders into pixelated mush.

 

Usually I would fully agree with you, but the horse looks alive and well.

Most will gain from a discussion to figure out what is true, what is false, and what the technology can do. It's currently misunderstood because the publicity is misleading. The exact same thing happened with the kinect launch video. I said the point cloud had a much lower resolution than the sensor specs, worse latency than what they show with the fake demos (remember the star wars demo?), and heavy noise making those things impossible... But MS was protected against "bashing" with facts. This is what makes B3D look bad sometimes.

If I posted some mrxmedia crap, you wouldn't prevent others from calling it BS. If I claim the PS4 looks way above 1080p, with arguments like, I know I have one, do you? It would be challenged with facts about the hdmi and TV pixel count. This is similar.

 

And ultimately, people end up so focussed on discussing the discussion and what we're allowed to discuss that we don't discuss the actual topic! Any exclusive tech/software ends up derailed with sides complaining about each other bashing/praising the product. I'd like to see a lot more impartiality with the enthusiasm kept for specific styles of threads. eg. If AzBat just wants to discuss the exciting gaming and application possiblities of Hololens without getting bogged down with a technical teardown of the tech, a thread discussing as such could be started, insulated from the typical technical counter-arguments.

 

This whole conversation started because I made one single sentence remark, on something I thought myself had been discussed enough, yet people kept misunderstanding what I meant and challenging me bringing up completely unrelated things such as fov, dynamic range of the eye, having used one and etc.
More than ever, I feel like its confirmed the misleading (because deceptive is too strong of a word aparently) advertising has lead to misinformation. Even beyond3d users don't grasp certain concepts fully.