Hacker News Reader

Comment by ArtDev | Hacker News Reader

I would buy one just so I could have virtually unlimited monitors for work: http://store.steampowered.com/app/382110/

I thought about that, but if you do the math the resolution is far too low. Maybe after a few generations it'll be a viable replacement for a multi-monitor setup.

That's what zoom in/out is for.

But really, per-monitor is too limiting. This really screams for a window manager that is aware of the display medium, and allows stashing of windows in a virtual desktop space. In a way, old Linux/BSD window managers like TWM and FVWM with their ability to have unconstrained (or very large) desktop spaces that you can move the viewport around is the paradigm we might want (with the addition of zoom).

The point is that since the Rift's 2160 x 1200 screen is so close to your face the pixels are further apart in your field of view. You're very far from Retina levels of smoothness. Small details, such as small text, will always look grainy.

Back to text mode then. Does infinity text consoles do anything better than one retina monitor? We will soon see!

Possibly the sheer real estate available could compensate. I wouldn't use it for e.g. graphic design or the visual parts of web work, but grainy terminal text or web content could be made more legible by just using more of the field of view and blowing everything up by 2x or so.

If you make the text big then you'll always be moving your head around, which is slower and more tiring than just moving your eyeballs. It's like sitting too close at a movie theater.

I guess the question is how small does the text need to be? When zoomed in so it takes most the view, you essentially have the equivalent of a ~40" screen right in front of you. Would that entire viewspace being a single terminal offer enough fidelity? If I could just turn my head to see a second terminal of similar size for secondary info (such as build error output, viewing a log, etc), then I think that might be sufficient.

It's worth noting that I think trying to approximate current desktops or monitors in a VR environment is entirely the wrong way to go, as it's an abstraction of an abstraction, when we now have the ability to get closer to the source that the initial abstraction ever could.

That is, we have multiple windows, or multiple displays with windows to approximate our ability to have multiple things in front of us. But we don't actually look at both things at the same time very often. That is, we might examine two separate papers, and have one in each hand, but we look at one at a time. Having two documents or terminals open in separate windows/displays but both visible is a way for us to approximate this on a display. We have both visible at the same time so we can physically look back and forth between them easily. With a VR display, we might want a view like that, but when interacting with an item, it will likely be zoomed to fill 90% of the visible space, as that's how we generally interact with things in real life. When we don't, that's because of other limitations that prevent it, or because we need to split focus for a specialized task.

Personally, I would love to have a VR system that presented windows/screens of apps in a 3x3 grid (or more likely a 3x3 cross, a center item with items at the four cardinal directions), where the center item was zoomed to 90% of the display or so, but I could look around at the others, and could rotate vertically or horizontally to center (focus) a new item. This would be the "active" applications/views, and other things could be stashed to the side. It might start as a single item, expand to 2x1, items, then 3x1 (so I can look left or right), then 3x1 with one above, then one below, etc.

When you compare VR to multiple screens, you also have to consider that with multiple screens you have the ability to see things happening in the peripheral vision, and react to changes or alerts. Switching from one screen to another is just a matter of shifting your eyeballs and focus, which humans can do extremely quickly. With VR, you are shifting your entire head by a rather large amount, and using a significant amount of energy to do so. Also, if you look at the way people usually work at a workstation, you will see people move their head, stretch their neck, etc. Our eyes can naturally compensate for that. For instance you can move your head from side to side to stretch the neck without having to loose focus of the screen. With VR, that's not going to be possible. Plus, with any slight head movement, your entire screen view would be constantly shifting.

Basically, what it comes down to is that we would need an eye tracking VR experience, not just head tracking, to make it useful as a screen replacement. Of course, I see the day when very high resolution small goggles would be something you put on and use in combination with your phone as a laptop replacement. Kind of the Ubuntu idea of a phone that can run desktop software. But that's different from VR. High quality goggles and AR is the future. VR? Not sure, the application is just too narrow, and from my conversations with die hard gamers, many of them just don't really like it for now.

I think it's more about the experience around it than a one to one swap. The end goal would be something where when you swivel your head/chair 90 deg that you see a different "bank" of monitors and that the response time is so low that it's preferable to doing some keyboard command to flip over to another screen on your physical monitor.

I did the math a couple years ago, at 4K (next gen?) things get reasonably useful (similar arc resolution to early 90's VGA monitors), and a 8K, you'd get pretty close to the arc resolution of looking at a 1080p screen everywhere you look: https://docs.google.com/spreadsheets/d/1pOx0Jcy6tpMGLxmPYX8n...

Apparently you need 1 real monitor for each 'virtual' monitor.

>The latest version (0.9.5) now supports multiple monitors as long as they are connected to the same GPU as the Rift. i'm working on adding support for multi-GPU and Optimus configurations. Multi-Monitor is only available in curved mode at the moment but I'll add support for flat mode at some point too.

This should help, so long as you have available ports:

http://www.amazon.com/CompuLab-fit-Headless-Display-Emulator...

Although as I said in sister comments, I think the resolution is still too low for most text-based work.

> I think the resolution is still too low for most text-based work

I used a headset with 2x 800x600 screens, but a narrower field of view; I would imagine that the arc spanned per dot is similar to the Rift, and I found it quite usable for text-based work.

This. It is a huge limitation of the software. You cannot have "unlimited monitors".

Finally, I can play "world of world of warcraft"!

How is that any different (functionally) than making new desktop spaces in your windows manager?

Oh it requires an expensive, clunky device strapped on my face you say? Sign me up.

> How is that any different (functionally) than making new desktop spaces in your windows manager?

In theory you can add N number of monitors to your desk and simply look at each one. Being able to simple turn your head and hook is a huge win over multiple, virtual desktop spaces (otherwise why would you buy an additional monitor?).

So a dynamic way of adding and removing monitors form your setup. Too bad this one doesn't quite work that way and wouldn't work as well (especially with, as you said, a huge clunky device strapped to your face).

Yes, that's what I was imagining (literally spawning new monitors in the viewspace) but this simply isn't that. Also with Oculus you need to have the physical monitors to support it (for now):

Exactly. The Oculus maybe opens up some very interesting desktop UI possibilities, but none of the options I've seen seem to fulfill that promise. Yet

It doesn't need screens. Or a desk.

Yes it does.

You also need a computer to run the Rift. So unless you're cool looking like this (http://i.imgur.com/1HMnJoc.jpg) then you'll need a desk to put that on.

The screens sound like a limitation of Windows, I don't think Linux will mind. About the desk, I don't see why you can't plug this in a laptop, apart from the fact that laptop GPUs aren't powerful enough yet.

Oculus is Windows-only from now until the foreseeable future.

And I guess I was being pedantic about the desk but you understand my point.

He said on reddit that all you need to to is plug in cables or DVI-to-HDMI adapters to "fake out" your graphics card thinking monitors are attached, but physical monitors are not required.

Ah! Good to know, thank you.

So, you'd be interfacing with a virtual world of your virtual terminals which are running on a virtual machine in the real world?

I just realized I had not really considered how many possible uses VR could have.

Also to turn my workspace into a virtual corner office overlooking Central Park

Would love this to build a 'situation room' for the more expansive projects I work on. Sometimes inspiration comes when all the project graphs, tables, photos etc are presented in a way that I can just look from one to the other of the tens of key pages with a head movement (previously accomplished with pinboards)

Is anybody actually using this for some serious work?

How easy is it on your eyes compared to regular monitors? Doesen't your neck hurt if you wear Oculus for longer periods of time? How is reading small text compared to regular monitor?

For the preview device at least, the resolution was really bad. Terrible screen door effect.

Dammit, thanks for throwing a monkey wrench into my VR excitement. I just started having neck issues recently. These headsets probably aren't too great for my neck, especially first generation models.

If it's the weight of the headset you're worried about, try hanging a rope from the ceiling with the headset at the end, at the exact height of where you'll be using it. Think of the tennis ball hanging in many people's garages. The weight of the headset will be handled by the rope and you can just rotate naturally.

Doesn't solve up and down motion, though.

If it is an issue I'll consider that, although that's another cord limiting range of motion. There's a good chance I'll be able to handle it for short periods of time (less than an hour) without too many problems. I'm more worried about down the line when my neck is likely to get worse.

Yeah thats my use case as well, but i m holding off to the second version which should be more ergonomic. We need to start inventing portable/wearable keyboards

Of all the stuff you can do with VR, I think wanting "virtual desktops" is one of the least interesting (and useful).

But interesting =/= useful, most of the time.

How would you type? Xbox controller? Leap motion table tapping? Without the ability to input effectively it's a negative hit on productivity for multi monitor use.

A room full of movies playing is fun, until you see through it.

It's not too hard when seated on your desk with a keyboard and mouse. I wonder how easy it is on the eyes though.

Why wouldn't you just track the keyboard, and superimpose it over your virtual desktop when needed? You would want something to track keyboard position and orientation, and your hands. The winner at that point is the ability to phase different aspects of your workspace in and out of view at will, or extra contextual information based on your task, such as an overlay on the keyboard for shortcuts in the current application.

I'm hoping to try this approach with a digital piano keyboard. Will it be usable if I can see the keys but not see my hands? I'll find out. (I'm not a very good player, I can play without looking at the keys but I still need to glance at them, e.g. when changing position or playing a wide chord)

edit to note: I do have a Leap Motion, and I'll try that, too, but right now my plan is to use it above the keyboard as a gesture controller rather than facing down at the keys. Lots of experiments to try. :)

A leap motion integrated into this to show your hands might work really well. At some point we'll know what's required for a good environment like this, and have software that supports it all. For now, it's new territory.

For seeing your keyboard, simply turning on the AR layer on head tilt works pretty well. Here's a video of a demo I made last year using that: https://www.youtube.com/watch?v=jNtZmCDyEyU&feature=youtu.be...

Doing virtual tracking would be ideal but neither the Lighthouse or Constellation protocols have been published (you could tape a Vive controller to the keyboard if you wanted to simulate that)

Yeah, something that tried to do a visual mathcing of keyboard and hands and only showed those when you looked down would be idea for data entry (but looking down to see reality has other uses).

> Why wouldn't you just track the keyboard, and superimpose it over your virtual desktop when needed?

There's no substitute for having physical keys your fingers touch. I can't touch-type on a virtual keyboard.

No, I'm talking about representing your real keyboard, in it's correct relative space, in your virtual space, so you can correctly use it in reality. If there was a front facing camera on the VR set, you would just pass the image of keyboard and your hands through when in that entry mode, but without that you approximate the keyboard position and the position of your hands (with extra hardware, or a leap motion, etc), and represent them correctly in VR so they can be used as expected.

Ah, I see. Yeah, that makes sense. Doesn't seem necessary if you touch-type, but definitely useful for everyone who doesn't. You'd have to accurately represent your fingers, too, at that point, and with a lot of accuracy to match the user's proprioception.

(By comparison, you definitely want a representation of a controller or wand or similar manipulator in-world, so that you can position it spatially.)

if you are a touch typer this is a non issue. I've typed a bunch of rift on and if I really have to see something I just peek under. With that said though there are some projects out there where people try to track fingers on a keyboard with camera or leap motion so you can see they keyboard and type in vr.

What would the effective resolution be? Any good for detailed work?

This is actually the best "available for everyone" (eg, not remote surgical assistance etc...) use case in my opinion.

Especially for people working in open-floor plan offices.

Still too low res for work use. Next gen, maybe.

Have you tried it on CV1? I mean actually you couldn't have cause it's not out. (edit: this reads snarky, sorry, was just a bad edit.)

I haven't but from what I understand it's good enough quality and ergonomics for all day use.

I have not, but I have tried a DK2 and a Vive. Good enough quality for games, yes, but I think it's pretty clear that most will not want to be reading text on a virtual monitor for very long.

This is my main use case for VR right now, and having owned a DK2, the resolution being too poor to clearly read text is a dealbreaker, and that's before the screen door effect is brought in.

The DK2 is 960 x 1080 per eye, CV1 is 1080x1200 - a modest increase at best. An extra 120 vertical and horizontal pixels just isn't enough.

My CV1 just arrived. Resolution is adequate for gaming, but it's a long, long way off from being usable for text-based activity.

In a world where moving from a standard-resolution to a retina resolution for programming was a major eyestrain relief for me, I can't see VR catching up for another decade at least (though I'd be happy to be wrong.)

I presume you would get tired of wearing that headset on your head after about 2 hours let alone 8.

you know people will be wearing them till they get sores. They will have scars - the addicts will.

I've never used a VR headset, but I imagine that most of any discomfort would come from the headset's fitting closely around the face to block out ambient light. I'd think a true addict might instead just use a loosely-fitting non-light-blocking headset in a completely dark room.

Sounds like neck pain.

Reply on news.ycombinator.com