I’m looking to upgrade monitors and wondering if a VR headset might not be a good alternative. From what I understand, they have OLED displays for great colors, high variable refresh rates and potentially infinite virtual space.
What are VR headsets like for watching movies, browsing or using software like Photoshop or Blender?
They are not good for any of these uses except maybe watching movies. The resolution is low, though, so you will not enjoy HD quality. With the older/cheaper models it’s like watching videos through a screen door.
The headset uses enough computer resources that I’m not convinced it could easily handle Photoshop work, and on top of that, I cannot imagine dealing with Photoshop’s interface in a VR environment. It would be the hugest pain in the ass I can imagine. Photoshop’s interface is already minuscule and I have a hard enough time accurately clicking on things with my tablet pen, much less a hand-held controller in VR space. Sure, you can zoom in, but then you’re constantly zooming in and out of the program window on top of having to do the same thing to the image you’re working on, which is double the difficulty. Then that doesn’t even take into account all the keyboard keys you need to use to operate Photoshop effectively, that I’m not sure can be mapped to the controllers. Mapping in a keyboard for use in VR is simply a headache - would PS even be able to use a virtual keyboard in this case, or would you have to blindly flail for your real keyboard? Putting in a pass-through video window to see your actual physical desktop to use the keyboard while also running Photoshop in VR…that sounds awful. On top of that, using a tablet would be like using an intuos blind. I literally cannot imagine a workflow where I could use Photoshop in VR without taking 50x longer than necessary and giving myself a headache in the meantime. I don’t even know if programs run in VR can even accept inputs from a tablet. You’d be stuck with the hand controllers, and there’s no precision or pressure sensitivity there. It simply does not sound feasible.
All the same things apply for Blender except traditional 3D software is even MORE taxing on a computer system. I doubt that you could interface the hand controllers with Blender in any useful way, and would quickly make a mess of your model. It’s not like the hand controllers have any sensitivity slowdown and the ability to do precise clicks is basically nonexistent. There would definitely not be pressure sensitivity like you would get with a tablet. You can use one of the 3D modeling programs (like Medium) that was actually made for VR use, but it’s not like using Blender whatsoever and does not make models that can be used in traditional methods. I’ve done it. The amount of tris was astronomical and learning to use it is completely divorced from any method you’d use in Blender. What you can make in Medium is just a hobby thing, at best.
Browsing the web is also in low resolution. I hope you like straining your eyes.
There are some cases where I know people have watched anime in VR programs like bigscreen, to have the experience of watching something in your living room with a bunch of friends. It’s still not a great way to watch stuff, but in this case animation lends itself to being watched in this way, and the handoff of being able to do it with other people makes it a good experience. But doing it on your own? Bad and awful.
In short, VR is not designed for this use and the technology is still so new that your idea is dead in the water. You need specially designed programs for use in VR to get any use out of it in this way, and the programs are all “baby versions” of actual workhorse programs by virtue of the interface and controllers.
A SLIGHTLY more positive take: it’s not there YET. Watching movies in VR already can be a good experience, although, as mentioned, it will be slightly below HD even on higher-end headsets. I DO think it’s currently the best way to watch 3D movies at home, though – a 3D virtual theater actually feels better than watching a 3D TV with glasses. More immersive, bigger screen, and you’re always in the “sweet spot” for 3D.
Things like photoshop and blender aren’t going to be good “out of the box” experiences, but VR/AR-enabled versions or competitors could be viable at some point. I think they would likely require specialized controllers, though. I personally think an AR Photoshop that can essentially map a tabletop to function like a giant touchscreen paired with a pressure-sensitive standalone stylus will make for a fantastic editing setup someday. But…today isn’t someday. AR is EXTREMELY undeveloped and not at all ready for this type of use.
I own an HTC Vive, HP Windows Mixed Reality headset and a Google Daydream headset for my Pixel phone. I’ve also owned an Oculus Go.
These things are awesome for certain types of games & VR-specific software (Google Earth is fantastic), but are completely useless for general computing. 2160×1200 might sound like reasonably good resolution, but not when it’s magnified to fill a 100-degree field of view. It’s OK for gaming because the immersive feel makes you forget it, but whenever you try to read text or make out fine detail, you realize how fuzzy the image is.
There’s no doubt that the industry is really flat right now. Part of the problem is that the sore spot for VR is still the screens, and they’ve basically maxed out current-gen technology and the next gen isn’t ready for release. That next-gen stuff exists – higher DPI screens and screens that eliminate the “screen door” effect exist now but aren’t commercially available. Screens with multiple depth layers for more natural focusing and MUCH higher resolutions exist in testing but are probably years away.
The Oculus Quest is pretty much the only interesting thing on the horizon, and that isn’t pushing ANY new ground. It’s just trying to find a new sweet spot for price, comfort, and usefulness.
There are lots of little improvements going on, and adding up to better experiences all the time.
According to reviews, the new Samsung Odyssey+ headset mostly eliminates the screen-door effect. It’s probably an optical lowpass filter of some kind, but that’s still an improvement.
Pimax has started shipping their “8k” (actually 4k per eye) headsets, reviews say is a noticeable improvement. The GPU becomes the bottleneck, however. I hear it maxes out an RTX2080.
HTC just announced a version of the Vive Pro with eye-tracking, which is one way to eliminate the GPU bottleneck - it detects where you are looking, and render just that part of the scene at high resolution, and the rest of the scene in much lower resolution.
And the Oculus Quest may not be a huge technological leap, but I think it will be the first stand-alone headset with 6DOF tracking and 6DOF controllers. Which means it’s the first stand-alone VR system that can play room-scale VR software. That’s a significant leap in terms of usefulness.
That’s a good summary! I was kinda intentionally glossing over thing because I know there’s a sense even within the industry that the next generation gear will be a more interesting leap, but you’re absolutely right that there’s good iterative improvements even now.
At what points in terms of resolution and framerate is VR seamless? 4K/eye at 120fps? Is there anything like that expected to come to market in the next few years?
Well there are a number of dimensions and there are presently tradeoffs.
High frame rates make VR feel much more seamless. Not sure if 120 fps is necessary, but 90 definitely is. But higher resolutions are harder to render and cost framerate.
Not having a cable to the headset makes VR feel more seamless…unless the wireless link experiences interference or the batteries run down. And the higher the resolution and framerate, the higher the wireless bandwidth required, and the more difficult this is to even accomplish. (and you have to use compression which lowers the quality)
Fovea tracking is a huge improvement because it lets you chop down the pixels rendered, increasing frame rates, resolution, and reducing wireless bandwidth. This is one tech that improves it all.
So what I am saying here is, there is not going to be a VR headset that is optimal in every dimension in the next 5-10 years. Maybe 20 if VR doesn’t die out again. But better headsets - higher resolutions than the Rift and Vive, possibly wider fields of view, improved lenses, and fovea tracking - are expected to be released soon. Probably this year.
And note that even in 10-20 years we’re not talking about a headset that’s Star Trek and absolutely flawless, just sharp enough that it isn’t perceptible, pretty much covers the FOV of an average adult, and is wireless and at least 90 fps and lightweight with a form factor and like anti sweat fans or something.
And google VR does have prototypes for the real headsets, ones with crazy resolutions (at least 4k per eye, rumor is higher). At these extreme resolutions you will be tethered to a cable for years, but you can actually plausibly build a work environment where you could use VR productively.
Yeah, fovea tracking is potentially a bit of a game-changer. It actually has multiple benefits, since not only does it lighten the computation load (since you don’t have to render the whole frame in full detail) but also having areas that you’re not looking at defocused can supposedly be more comfortable for your eyes.
The truth is that even full 4k-per-eye resolution isn’t “enough,” because, if you think about it, that means that looking at a screen within VR will still be substantially less than 4k (unless your virtual face is right up to the virtual screen). BUT 4k-per-eye might be enough for a lot of actual uses, since it could render a 1080p HD screen fairly accurately. Pushing 4K per eye at 90+ FPS with HDR color depth takes some horsepower, but its not unreachable. Still likely to be pretty niche for a while yet; the trick will be for VR companies to find ENOUGH market to stay alive and keep software coming. The latter is the big one – there’s already complaints that good new VR content has been slowing down.