Confirmed: Philip Rosedale Creating Next Generation Virtual Reality Project, Likely With a 3D Voxel Engine

High Fidelity is the name of Philip Rosedale's new startup, TechCrunch first reported, and it's apparently that "secret project" I blogged about last February. That is to say, it's about, as one of High Fidelity's job descriptions puts it, "prototyping the technology and user experience of a next-generation virtual reality system."Another High Fidelity open job assignment description written by Philip strongly suggests it's going to run on a voxel-based engine:

Theoretically (I think), one could maximize the performance of a voxel-based renderer by passing the compressed representation of a sparse octree directly to the shader, and then building the appropriate cubes to match the data. Create a renderer (if possible) that does this, and report on its frame rate and performance.

That's some heavy geek lingua franca, but in plain English, a voxel-based renderer improved along the lines Philip describes would create real-time 3D graphics of extremely high, well, fidelity. A couple years ago, for instance, a voxel engine called Unlimited Detail got a lot of buzz and also a lot of criticism from folks like John Carmack and Minecraft's Notch, who thought it wasn't really ready for prime time. But the demo video was cool and suggests what Philip's optimized voxel engine might look like:

In 2011, John Carmack told me this kind of technology in a consumer-ready product could possibly happen "several years from now." Which now means quite soon. More from what Carmack told me then: "You can real time ray trace a static world on high end hardware today as a demo, but there is a long path between a demo and something that is competitive with rasterization in a real product. My plan for such technologies has always been to emit a depth buffer as well as color from the voxel/point cloud renderer and continue to use existing technologies for characters/particles/etc." All of which suggests another possibility for High Fidelity: Hire John Carmack.

In any case, several other former Lindens, including my friends Jeska and Ryan who I worked with at Linden Lab back in the day, have also come along with Philip, and the website includes a statement that fits Philip's original vision for Second Life as the next generation of the Internet:

What will the information spaces of the future look like? 20 years ago the web didn't exist. What's next? How will address schemas and navigational metaphors evolve to keep pace with computing power? How can we start designing now for a level of richness and detail that continually extends deeper?

So whatever High Fidelity ultimately makes, it definitely seems to be another attempt to fulfill Philip's aspiration he had back in the 90s (as he told me in The Making of Second Life) -- to be immersed in the Internet and be able to navigate and create content within it in a physical way.

Please share this post with people you like:

Tweet

Comments

[Damien Fate]

Definitely keeping my eye on this one!

[Maria Korolov]

Very cool!

I can't wait to put on my Oculus Rift headset -- maybe climb into this device here: http://www.element14.com/community/community/news/blog/2013/03/21/virtuix-omni-allows-gamers-to-move-freely-in-the-virtual-world -- and walk around a high-def virtual world.

[Beverly Millson (aka Bettina Tizzy)]

Philip Rosedale is a creative genius. He's also a lousy manager.

Philip, I admire and respect you. For this and all future endeavors, please acknowledge your strengths and weaknesses and align yourself with brilliant, visionary managers that build smartly and finish what they start.

Second Life is still one of the best man-made creations ever, but it's languishing for all kinds of management reasons. By the time Rod Humble took over, the damage SL had sustained was irreparable.

If Philip will just stick to creation and leave management to people who are his equals in marketing, finance and operations, then his works (which are totally awesome) will live up to their potential.

[Adeon Writer]

It sounds like Rosedale is working on the software, rather than the hardware, of VR. The Oculus Rift is sensors and display, but ultimately needs a videofeed. Sounds like Rosedale is focusing on THAT end.

I'll be buying my consumer-version of the Rift on day 1.

[Metacam Oh]

Yeah sorry Rosedale, you're virtual world clout is good no longer with me. The last thing I'm going to do is invest in another one of his dead ends, I suggest everyone else think carefully before you get suckered into another virtual world project that he will eventually lose interest in.

[Iggy]

What do you do when your utopian project fails to remake the whole world?

Hey, try try again!

I hope that not too many optimistic, clever, and well meaning folks get hurt this time around, as Philip one again tries to harness the lightning he saw at Burning Man oh so long ago.

[A.J.]

This is very cool.

I never came from divorced parents so I can only imagine what it feels like when you see your dad leave your family in a ditch because he needs to be free and then, two years later, watch his new kids get all of the cool stuff for Christmas.

Philip, your SL family are the very people you're gonna need to buy the Christmas gifts you're promising here. You better make sure there's something real sweet in our stockings next time.

[Melissa Yeuxdoux]

I came to Second Life for the things I could do in a virtual world; I have stayed not only for that, or more accurately for the things others have done, but for the friends I've made.

If a new virtual world comes along in which all these things can be done better--more realistically, more efficiently--then I will try it. I won't leave old friends behind... but perhaps we can move together.

"Plan to throw one away. You will, anyhow. The only question is whether you will deliver the throwaway to your customers." That is, as best I can remember, what Fred Brooks (who, BTW, has an interest in virtual worlds!) wrote in _The Mythical Man-Month_. For something like a virtual world, I don't think you have any choice but to deliver the throwaway--but backwards compatibility is a millstone. At some point, better to start over.

[pixeleen mistral]

Hammie,

I suspect the main issue with any of Philip's ventures straightforward:
How can you yiff with this?

[Drew956]

Here's what I don't understand...

I can download an OpenSim server on my laptop and can play with it however I please on my LAN.
Now can someone explain to me why I can't purchase an offline version of SL with online capabilities?

It wouldn't remove online creation, it would just promote the security of being able to store your purchases in your local virtual home or realm.
And your avatar could still visit the SecondLife grid to do transactions and buy virtual content, or to own virtual land on the grid.
People would still pay for land on the grid because they build attractions and realms collaboratively there and because there is always an economic incentive.
(Oh and IMO it should be hard to collaborate on that same scale using port forwarding lol)
I know plenty of people who would probably be all for offline creation.

IIRC, Minecraft is both a client and a server and the environment runs on both simultaneously? (not 100% sure, I am trying to get into modding and something to this effect was mentioned)

Why can't an offline version of SL be a LAN client and server, and be a viewer for the SL grid also?
Some people probably don't need or care for the grid all that much.
(but as soon as they do they can still connect)
And they would want to spend money knowing their purchased items would be downloaded to their home realm.

I think it would be awesome to be able to invite your friends over to your dojo if you have port forwarding, and to be able to send virtual mail from your home mailbox.
[email protected]
or something snazzy and fuzzily warm feeling (domain name wise)
The lindens could give those who bought it their own email address and there could be inter-realm communication..
Whole SL compliant grid-servers, whole REALMS, could be made by avid SLers if people were allowed to host their own SL servers, but alas the infrastructure is too closed currently.

Hopefully if SL does fail it will become Open Source and OpenSim will benefit.
There is just so much that could have been done with SL...
Oh what a pity.
(assuming this is the beginning of the end, but I'll wait and see...)

I've never had much time to SL and my laptop graphics are quite bad and laggy, but those few times I did visit were awesome and magical.
And I did download an Opensim server as I mentioned, on which my sisters and I built "whales" out of prims lol.
Good times...

Bottom line is: an offline SL would be what I look for in Minecraft but do not see: unlimited creation potential.

Minecraft has preset creation possibilities, SL has preset creation tools.
(modding MC doesnt count because it's not intuitive, and not part of Minecraft itself (although I hear they are working on an API for MC))

Maybe an offline SL would help LL get back on it's feet?
What do you guys think?
Would you buy it? Why why not?
You guys are the real SLers, I'm just an intrigued fan :)

[Troy McConaghy]

The High Fidelity YouTube channel has a few interesting videos. It seems they're also prototyping alternative input & control technologies:

http://www.youtube.com/user/HighFidelityio

I'm reminded of "The Rig."

[cyberserenity]

Maybe this technology could be used in Opensim?

[Robertus]

Very interresting indeed.

Imagine this Voxel engine in combination with a tool like Unity and when you want to use a 3D scanner also.

(I was to late for the already successful funded CADScan3D
http://www.kickstarter.com/projects/621838643/desktop-3d-scanner)

And of course: Virtual worlds like SL / OpenSim with this Voxel engine and the Rift ...wow

Btw: I just discovered this NWN blog and I compliment about the many interresting / quality subjects !

[Wolf Baginski]

Voxels seem to have been around a long time, but whether there's any useful connection between the voxel graphics of the last century and what PR is involved with, I don't know.

I'm guessing that the big thing will be getting good images with low bandwidth. I'm afraid I see some people talking about wireless solutions with no apparent thought for what will happen. The coverage in rural areas is patchy, while urban areas are becoming overloaded. I would be better off with a length of wet string than with the mobile data service I can get here.

[WhiteStar Magic]

Excellent for Philip !

Voxel Technology has such great potential and while it is still in mid level development and the naysayer's are bashing away because they don't want to admit/accept the potentiality, it is in fact very real. There are several organizations tinkering with Voxel now and acceptance is slowly growing, hopefully soon we will see a product using Voxel Tech.

People insisted (corporates & scientists) that Quantum Computing wasn't possible yet but it's happening now, today. Lockheed-Martin Aerospace has just purchased it's Second Generation Quantum Computer from D-Wave in Canada, to upgrade their original First Generation system which they used for a few years. There are other serious companies looking at at this technology / aerospace looking into it as it has now been proven to deliver on it's promise... Yet the naysayer's are still arguing it is not possible. Intel / IBM or AMD should partner with D-Wave and further accelerate the development and who knows when that tech would be scaled down and made more affordable... $10 Million for a Quantum Mainframe = lots but it will get cheaper as development continues.

There are always naysayer's:
* When Radial Tires were developed the companies making Bias-Ply said No Way, No Good! Anyone using Bias-ply now ?
* When desktop PC's first came out, many said, "They will never be used in serious business" So what happened to all the Main Frames ?
* When anti-smoking groups said Cigarettes are bad for you, a Myriad of people, companies & scientists said NO WAY, It's not bad for you... We know the result of that, don't we...

[Melissa Yeuxdoux]

So here's what I'd hope for:

1. Some sort of converter that can turn SL objects into High Fidelity objects.

2. Some agreement by which SL creators can export their SL work to HF.

[Fleep Tuque]

Color me skeptical.

I'll watch the progress like everyone else, but I still think the biggest challenges for the future metaverse are social/cultural, not technical. Prettier graphics won't solve the social barriers.

I thought Coffee & Power was an attempt to investigate some of those social issues and was looking forward to seeing what Philip and crew would do with it, but I'm a little bummed if they've given up on the social and are going back to focusing on the technical. I guess we'll see.

[joker]

its easier to dress up and sell a mechanical pig.

[Kim Anubis]

I had hoped Philip would give VW's another shot, and he has good people working with him, too. Philip knows the difference between creating a product and creating a world, and he got a lot right last time. I'm looking forward ....

[elizabeth (16)]

I think the difficulty faced by Coffee&Power was that it was a bit to much open-ended and undirected. it essentially mimic a general job quoting system in practice

+

High Fidelity seems to be more directed in terms of what Mr Rosedale wants from it. seems to me that is the same collab principles as C&P being applied. just more focused on a end goal this time

[Arcadia Codesmith]

The question is as it has always been; can you push enough data through the pipe fast enough that a dynamic world stored on the host renders as smoothly (or close enough that the end user sees no difference) as a static world stored on the client machine?

I think the approach in that job description is a productive avenue to explore. It's just one aspect of building a viable virtual world, but if it pans out, it could be huge for mass market acceptance.

If a dynamic VW can't outperform a client-stored MMO, then it can be light-years ahead under the hood and still not make a dent. The user experience has to be superior out of the box.

Good luck to Philip. The industry could use a tectonic event to shake things up.

[Iggy]

/me high-fives Pixeleen and asks for more Alphaville Herald please. Time for a column there about Third Life, PR's next Big Shiny New Thing?

[Gwyneth Llewelyn]

Ok. Uh. Ah. Oh. Well, I don't know.

I'm too open-minded to be too skeptic, but I can only say: this will surely attract techies. Sure, they will come. If you need a $3000 computer to render things like what HF is promising to do, then a million or two hard-core gamers and 3D modellers will most definitely jump onto HF.

What about the rest of the world?

Well... we thought they would come to Second Life. They didn't. So perhaps Philip's new venture works out fine: a better SL. Sure. Why not? So many people have tried to do that and fail, why shouldn't SL's inventor out-invent his own creation? Einstein also dropped special relativity and created general relativity instead. Sure, Philip's not Einstein, but out-performing one's own inventions is cool, right?

They didn't hire John Carmack. Nor, of course, Cory, who is still enthusiastic about Cloud Party, will not budge. But Philip hired Jeska again. That sounds good. Doesn't it? :)

I only wonder about conflicts of interest as Philip sits on both boards :)

Unless, uh, SL is going to migrate to HF?

Hmmm. That would mean that they'd blow up all future competition forever.

[sps]

Maybe this voxel engine isn't the "Unlimited Detail" kind but more like the voxel based engine/tool which Cloud Party recently introduced and which James recently blogged about at http://nwn.blogs.com/nwn/2013/04/cloud-party-building-tools-and-contest.html

[pingsting]

It looks incredible, the potential to be truely breathtaking and so very brilliantly immersive.

With regard to virtual worlds, I hope Philip stays true to his 'Burning Man' vision of a place where people can be whatever they want to be (but we'll print our own money and you must use it!).
SL gave LindenLab a cashcow of retards that LIVE their dreams. That currency choked and killed it.
Make it free, make it for everyone. Build it, and they will come.
Cheers!

[Mera]

Yesh... Cloud Party is already building with voxel tech. Its totally ossum *_* When i visit SL nowadays it feels very old and the graphic sucks bad when it comes to avies and moving animals. Cloud Party is a mesh/voxel world and it has spoiled me. Still its a new world so it still needs a lot of patience from the users and growth <3

[Alan Seeger]

Second Life isn't dead, but it's not exactly robust either. I'm interested in seeing how things go this time around. I'm not counting Rosedale out yet, not by a long shot.

[Lingerie sexy]

It looks incredible, the potential to be truely breathtaking and so very brilliantly immersive.