Hacker News Reader

Comment by alexmingoia | Hacker News Reader

VR is great but making any decent 3D model or environment requires enormous amount of labor. Without better tools I don't see VR being used for anything but games since you'll need to throw huge teams and many hours at making anything decent. Most (all?) VR apps and games I've used look like shit.

View on HN

Lots of interesting research being done on this, keep an eye out for stuff from SIGGRAPH later this month.

Here's an example from SIGGRAPH Asia: https://www.youtube.com/watch?v=Oie1ZXWceqM .

You're right. However, VR-native tools for creating 3D environments are going to be vastly easier to use than desktop based ones (so much of the specialize skills required for using e.g. Maya are just because you're using a 2D mouse and a 2D screen to interact with a 3D world), so I feel that this problem will get quite a bit smaller (though not disappear, of course).

This is the real problem of VR right now. Content is much more expensive to produce than it is for the written language, music, or film.

Games are the one thing that takes equivalent development effort as VR.

The first billion dollar VR startup is probably the company that makes it cheaper to produce VR content.

There has already been a first (multi)billion dollar VR startup. Oculus.

One reason why user-generated 3D content in general looks bad is that for a long time, "professional" content typically has lighting applied to surfaces as a texture, and that texture is expensive to compute and requires complicated software. User-generated models on the other hand (like one might find in Second Life) don't look that great in comparison because the lighting is simplistic and none of the models have pre-baked lighting textures.

As real-time global illumination becomes faster and better, I think we could get to a place where user-generated content looks pretty good by default, because it's always being re-calculated. I haven't really kept up with what the latest games are doing these days, so I don't know what the state of the art is, but there have been some pretty cool demos of real-time path-tracing. It may be awhile before that sort of thing is fast enough to meet the latency and resolution requirements of VR without being grainy, but I think we'll get there eventually.

Well, in our group we just started (academic) research testing out possibilities to bring data analysis, visualization and remote collaboration to VR (Vive).

I'm quite curious whether this will work out and what the future will hold for us.

So has our lab! We should really share some knowledge and experiences. If you ever need another academic lab in London to try some stuff out then give us a shout.

I think that would be a really great idea. Sounds like a lot of good ideas could be brought together.

If you want to get in contact feel free to shoot me an email at [redacted].

I, for one, am waiting for GitS style chat rooms:

https://www.youtube.com/watch?v=poKi7YyuamI

Isn't AltspaceVR or Oculus Social or vTime or JanusVR this? I mean, I'm pretty sure I saw an exact replica of this scene in JanusVR.

Just pick your favorite from the dozen or so Second Life forks.

What lab is it? Do you have any footage? How does it work?

It's the [redacted].

We literally got our Vive yesterday, so right now we are still in the concept phase. Hopefully I will have something to post on HN soon :)

Lemme know when you post it. iamwil@gmail.com

Pretty interested in what you guys are doing.

Reply on news.ycombinator.com