pouët.net

Are there any openvr/steamvr based demos yet?

People just realized how much anything VR sucks.

I suspect part of it is that it's hard to show VR demos to large audiences =)

ha, nice talk.

Last I checked, demoscene didn't care about revenue. ;)

It's not, but revenue drives technology, which is why Oculus is releasing its second mobile VR headset that's roughly as good as an Android phone whereas the actual desktop VR is pretty much over.

VR takes away the camera-work aspect, which is generally pretty critical for demos.

That's not to say you couldn't use elements of demos as VR toys/experiences.

I wonder how many people install some emulator for some calculator-like-computer, which wasn´t mainstream anywhere, anywhen...just to watch that "first demo" that appeared on it, finally! Oh, i guess just noone does, as there´s youtube! ;)

The problem with AR/VR-demos is missing Devices (youtube can´t emulate these) on users end, just as missing emulators on users end! ;) And emulators are even free of cost to download/install (which takes 5 mins, including googling them up first!)!

The Emulator-problem is a problem of humanity, if humans don´t get fed with a golden spoon they don´t want to get fed, but still consume!

The AR/VR-problem is a problem of humanity, if humans don´t get fed with enough money to buy proper devices they won´t be fed, but still want to consume!

--> AR/VR is still the future...atleast to the most of us humans, but it IS THE FUTURE! ;) (It won´t die off, believe me, too useful and even fun in the end!)

Yeah, Android phone VR is enough for porn, I guess.



This is true, and it's true that VR thingies are better described as "experiences" than "movies". Google has made a bunch of "demos" that can be found on steam ("google spotlight stories") which tell you to sit down and enjoy the ride with no controls. I haven't checked them all out, but Pearl at least is nice. You can sit down and stand up in that at least, haven't tried moving around much (but the demo requests you to stay put).

As for camera work etc, I think the VR space is still trying to figure out things. Maybe the 180 degree camera approach would work with demos; no walking around, just sit down and watch. To avoid sickness it's probably best to keep the camera still.


This is pretty much the same problem as with showing VR demos at a party - you simply can't show it to a few hundred people at once unless you have those few hundred headsets, too..



People have been trying to push 3d user interfaces forever (on 2d screens) and the problem there has been that there's always a more convenient, faster, easier and more powerful way to do the same thing in 2d. The same applies to VR. VR is more immersive - it's really an experience - but basically anything you can do in VR you can do in 2d, easier, more conveniently, more accurately, cheaper, faster. That's why "everything VR sucks".

I haven't tried a lot of the content creation tools (like sculpting) since they, well, cost money, but the ones I've tried have quickly shown the limitations of the tracking accuracy. The tracking is clearly better than with, say, nintendo wii, but there's limits to it as well as limits to how stable my hands are when moving in free 3d space as opposed to moving a mouse on the desk. Scaling up would definitely help, but that would require more space to walk around in, which then leads to the whole cable discussion (not to mention requiring the actual free space to walk around in..).

Anyway; the primary reason, I think, for the lack of VR demos is the lack of audience. Or at least lack of traditional audience; putting the demos up as VR experiences would gain some audience but not the demo making ones. (Then again, there are people who make demos for traditional telephones and flip screen displays, soooo...)

Doing the demos as 180' videos would also - partially - fix the youtube limitation as you can upload 180' videos on youtube and then people can watch them with their phones taped to their heads if they want to.

..Oh, and finally, paying 100 bucks to release your demo on steam to reach the largest possible audience is also somewhat of a limiter.

Here's the talks Smash and I did that covers this:
https://www.youtube.com/watch?v=fph_4-TyiA8
https://www.youtube.com/watch?v=ft5OXHnVKac

Thanks!

It's kinda funny how I managed to hit many of the same notes as you did in the talk(s), but it was also nice to hear about actual experiences of doing things in VR. I haven't, as of yet, had the time to try things out in practice.

Another possible demo opportunity would be to have your demo-thing as a tracked object linked to the motion controllers, where the user would rotate it freely in their hands.

Watched smashs talk, here are my 2 cents/additions:

- There are some big studios creating VR content, especially on Playstation VR. I've played "Until Dawn - Rush of Blood" a few times, which is a very capable (and fun) Rail Shooter.
- Imho, VR is still a hassle. Setting it up, putting it on. No comparison with "look at a screen". I even found Gear VR a bit cumbersome.
- Plastic are still at it, check out "Bound" which is a nice hybrid of a demo, game and "experience", but once more this is only available for PS VR.
- VR can be used to enhance existing experiences, think amusement park attractions. AEG^smashdesigns is doing some stuff in this field, it appears to be a promising market.
- I don't see VR having any relevance within the demoscene in the near future, because it's just too inaccessible and it takes the fun out of "staring at a bigscreen together"
- Despite all that, I love VR and I'm extremely happy that this stuff exists now

- while Unity/Unreal are associated with a certain "look", imho this is mostly due to their easy accessibility and lazy, less competent devs, particularly with Unity. The best Unity games are the ones where you can't guess the engine. In the end it's about pushing data to the GPU and using the tools provided in a meaningful way. Among the most successful Unity games I've worked on is one that has exactly one scene, one object and one script called "bootstrap.cs" :D

- streaming a demoparty in 360° stereoscopic VR would be ultra nice for sofascening!

that's maybe not impossible :)

Hey, turns out you CAN do OpenGL 1.2 stuff with VR. Still very much a work in progress, many parts are simply broken, etc, but it was surprising to find that some bits work as is.

Now that I've done some experiments (the new year demo will come in a openvr variant too) I have to say that my original idea of rotation-only-tracking-vr doesn't work, it just doesn't feel good.

Scale of things also matters. Like, a lot. I don't remember if that was mentioned in smash's talk, but there we go.

Anyhoo, I heartily recommend trying out making some vr stuff. It's one thing to have a 3d model on a screen and another to have it be right there in space next to you.

The openvr sdk is dead simple to use, basically you call your (2d) render stuff twice, applying hmd/eye matrix each pass.

don't excuse yourselves with camerawork, it's not necessary to provide interactive 360° view - it's enough if there's stereovision and that the scene covers whole field of view - wouldn't mind that moving my head doesn't affect camera.

That's not actually true. A lot of the depth perception comes from head parallax. That's why if you watch a 360 video, even if it's stereo, you can very clearly tell it's video, because the smallest head motion breaks the illusion.

It surely is better to have it than not, but it's not a showbreaker. Perhaps a margin of few degrees of head movement would be enough to keep the illusion, something that everyone should do anyway to cover different aspect ratios or fields of view.

A title for no head parallax demo: Sodom & Gomorrah - Don't Look Back :D

It's also downright fun to go and take a closer look at details.