Support NeoGAF

[Glassboy]

Go download the benchmark tool from each manufacturer. Should tell you right away if you're ready or not

Well steam told me I was ready but oculus did not because it did not make the minimum requirements.

 

[Krejlooc]

Please explain to me how I will be able to walk through a free range open world (think Witcher 3 or Skyrim) inside a 18x18 room ( who has a room that size just for VR?).

http://www.cs.unc.edu/techreports/01-007.pdf

there you go

 

[Krejlooc]

Sorry, can you show me the battery slot on a vive pre lighthouse?

I don't have a vive, so no.

 

[Zalusithix]

You don't need to adhere to VESA standards thanks to the earlier mentioned compositor. The compositor will build a VESA-compatible frame at the headset itself.

The process is simple - you wind up sending two tiny frames simultaneously be composited at the headset. You already have the framework worked out in your post. One tiny frame is for everything besides the fovea area of the image - this could be theoretically something tiny, like 320x240 resolution or comparable. You blit this to the compositor using a scaling algorithm - even something as simple as doubling or quadrupling up the pixel resolution, then blit the second tiny frame at it's native resolution on top of the scaled image, at the position the fovea rests.

No, of course this is not free, but blitting is hardly an expensive operation, and the gains made by sending a few tiny frames and composing the image at the headset is going to be way smaller than the cost of transmitting a full-resolution frame.

If you're not adhering to VESA standards, then you're making a new one and requiring the GPU manufactures to play game on the driver level. At that point you better fucking have the support of every headset manufacturer because things will get messy very fast otherwise.

 

[Krejlooc]

If you're not adhering to VESA standards, then you're making a new one and requiring the GPU manufactures to play game on the driver level. At that point you better fucking have the support of every headset manufacturer because things will get messy very fast otherwise.

Or, much more likely, a standard will arrive that will be adopted by headset manufacturers. Just like VESA in the first place.

Foveated rendering will have a profound effect on display standards. Why would you assume no such standard will arise for VR displays? Why would you assume they would continue to stick to display standards built for conventional 2D screens as technology progresses?

 

[mrklaw]

I don't see how "oh i want to walk over there, shit my wall is in the way. now i have to use some weird locomotion method (teleportation) to get around that issue" will ever stop being annoying.

It works fine for stuff where the scene can realistically be constrained to an area that small, but I wan't to play open world stuff.

nobody is saying room scale is the end goal of VR. There are still huge challenges for open world games and wider exploration and freedom of movement. But even just being able to take a step or two to the side is a big step forwards from standing up, which is itself a big step forward from sitting down - for games that represent players walking around. Obviously for cockpit type games then you want to be sitting down anyway

 

[Monger]

Please explain to me how I will be able to walk through a free range open world (think Witcher 3 or Skyrim) inside a 18x18 room ( who has a room that size just for VR?).

You won't unless you want to use redirected walking to walk across an entire city. Unless motion sickness disappears and VR sells 40 million headsets, these aren't games that will be made for VR. VR support may exist for said game and a person could play it with a controller if they want, but VR isn't a replacement for traditional gaming if that's what you're expecting it to be.

 

[Compsiox]

Maybe if more people had the DK2, more people would desire room scale. The tracking space is very limiting and leaves you wanting so much more.

 

[Soi-Fong]

Sorry, can you show me the battery slot on a vive pre lighthouse?

I can prolly get you a picture of it. Give me a sec to take it off the wall mounts we have.

 

[chubigans]

http://www.cs.unc.edu/techreports/01-007.pdf

there you go

Wow, that was a fascinating read. Thanks for sharing!

 

[pj]

I don't have a vive, so no.

Here's a picture of the consumer ones

http://36646d87786feafc0611-0338bbb...dn.com/images/QpelErEKqzLD.878x0.Z-Z96KYq.jpg

Edit: picture of the bottom: http://cdn.mobilesyrup.com/wp-content/uploads/2016/01/HTC-Vive-Pre-8.jpg

Please just circle the area where the 35wh+ battery (assuming the 7 hours you observed is the minimum battery life) goes. Note that 35wh is about the size of an ipad battery, so it's pretty impressive that they can fit something that large in such a tiny cube!

 

[Zalusithix]

Or, much more likely, a standard will arrive that will be adopted by headset manufacturers. Just like VESA in the first place.

Foveated rendering will have a profound effect on display standards. Why would you assume no such standard will arise for VR displays? Why would you assume they would continue to stick to display standards built for conventional 2D screens as technology progresses?

Generally speaking standards are slow to be ratified and implemented. I wont say it'll never happen, but "soon" isn't a word I'd use to describe my view either. We're currently starting out VR with multiple APIs and having at least one of the two major players seemingly not wanting to play ball. This isn't a good start for creating any universal standard. We don't need a Freesync vs G-Sync thing happening with VR.

 

[ashecitism]

Modbox - Creation mode gameplay: https://www.youtube.com/watch?v=_TFcqtVPljo

#SelfieTennis gameplay: https://www.youtube.com/watch?v=2wgUGGvJCG4&feature=youtu.be

 

[Croatoan]

http://www.cs.unc.edu/techreports/01-007.pdf

there you go

Ok cool, now lets get realistic. Out of the people who will buy VR, how big is the audience going to be for a room big enough to pull that off? I am going to guess pretty small. From that paper it says once you start dropping below the optimal size people start noticing the NASCAR effect. People in this very thread don't have space for Valve's room scale and we are enthusiasts. Houses are not going to magically get bigger in the future (or grow extra rooms).

My wife loves games, but there is no way she is going to let me have an empty room just for VR in our house so I can use redirected walking. Hell most people can't even afford a house period.

No dev will take advantage of a "feature" that so few can actually use. So while the tech is cool, and will make a great experience at Disney Land, I don't see it as a plausible solution for making a mainstream, AAA, or whatever "consumer word" you want to use, open world VR game.

You won't unless you want to use redirected walking to walk across an entire city. Unless motion sickness disappears and VR sells 40 million headsets, these aren't games that will be made for VR. VR support may exist for said game and a person could play it with a controller if they want, but VR isn't a replacement for traditional gaming if that's what you're expecting it to be.

VR will have failed, the same way waggle failed to replace controllers, if it cannot replace traditional gaming. Thirty years from now VR should be the way we consume entertainment.

 

[Fret]

Please explain to me how I will be able to walk through a free range open world (think Witcher 3 or Skyrim) inside a 18x18 room ( who has a room that size just for VR?).

You can't

VR isn't about just grabbing existing games and throwing VR support on them

 

[harSon]

Maybe if more people had the DK2, more people would desire room scale. The tracking space is very limiting and leaves you wanting so much more.

I have a DK2 and I don't have a desire for room scale. I was able to test the HTC VIVE at AutoApp Studio in San Jose about a week and a half ago, and while it's definitely an awesome experience, it's not something that I would personally want to incorporate into my daily gaming. I can see why a lot of people would. And as someone who's paying ridiculous Bay Area rental prices, I simply cannot accommodate that type of a setup within my current living space - regardless of how much I wanted it.

 

[Raticus79]

The frame itself can be massively compressed with effectively no discernible difference. You don't need lossless transmission thanks to foveated rendering - very lossy transmission is fine provided only the small area of the fovea is clear. You can even use multiple resolutions and have a piece of compositor hardware on the headset itself piece it together for you.

You can send a single high resolution small are for your fovea, and a much lower resolution full frame, and you won't notice the difference at all.

Or, much more likely, a standard will arrive that will be adopted by headset manufacturers. Just like VESA in the first place.

Foveated rendering will have a profound effect on display standards. Why would you assume no such standard will arise for VR displays? Why would you assume they would continue to stick to display standards built for conventional 2D screens as technology progresses?

Yeah, I've thought about this myself and it seems like you'd need the hardware in the headset to handle reconstructing the mixed-resolution stream. Anything in the works with DisplayPort? It seemed to be pretty flexible.

Another idea I've been throwing around - with foveated rendering, Virtual Cinema could be used as a way to get foveated rendering benefits in non-VR games. Basically use the headset as an eye-tracking device and have the GPU API handle the resolution regions (via an injector of some kind?).

 

[pj]

No dev will take advantage of a "feature" that so few can actually use. So while the tech is cool, and will make a great experience at Disney Land, I don't see it as a plausible solution for making a mainstream, AAA, or whatever "consumer word" you want to use, open world VR game.

That's why that type of game won't be made in VR. Or if it is, it will be a hybrid experience such as minecraft VR where you can switch from "theater mode" to full VR at will.

You could use theater mode to do traversal without having to deal with motion sickness, and you could switch to full VR for combat or detailed exploration or whatever.

 

[Krejlooc]

Generally speaking standards are slow to be ratified and implemented. I wont say it'll never happen, but "soon" isn't a word I'd use to describe my view either. We're currently starting out VR with multiple APIs and having at least one of the two major players seemingly not wanting to play ball. This isn't a good start for creating any universal standard. We don't need a Freesync vs G-Sync thing happening with VR.

When people say "soon" with regards to stuff like wireless displays, we are speaking in terms of a decade+, not months.

We use VESA standards currently as a dirty hack. The first foveated rendering solutions will work precisely as you'd expect, building a full VESA frame at the PC and sending it that way. This will not be a wireless display. When wireless VR displays arise, then the new standard will arise, and it'll work the way I am describing - using lower resolution frames dynamically composed at the headset to drastically reduce transmission requirements. There are only 3 conceivable ways to send information faster wirelessly - reduce the amount of information needed to be sent, increase the bandwidth of wireless transmission, or eventually, move to light-based transmission technologies.

 

[Krejlooc]

Ok cool, now lets get realistic. Out of the people who will buy VR, how big is the audience going to be for a room big enough to pull that off?

People will adapt. Just like having rooms dedicated to television viewing back in the early 1900's made no sense, but today nearly everybody has one.

Of the early VR adopters? I would expect the percentage of people willing to dedicate a space to roomscale VR to be quite high, because A) VR isn't going to be selling in the range of tens of millions for many years and B) those jumping into VR at the beginning are more willing to invest in ways beyond the average person.

 

[Soi-Fong]

Here's pictures of the Lighthouse that I just took off the wall. Krej, I'm not exactly sure what you're talking about. It'd be good if you not just put out statements as facts. We devs shouldn't mislead like that. There's no battery slots in this thing. Took these just now.

 

[Croatoan]

That's why that type of game won't be made in VR. Or if it is, it will be a hybrid experience such as minecraft VR where you can switch from "theater mode" to full VR at will.

You could use theater mode to do traversal without having to deal with motion sickness, and you could switch to full VR for combat or detailed exploration or whatever.

Im not talking tomorrow, Im talking 20 years from now. Im sorry, I guess people think I am talking about CV1 lol. The mainstream will not buy into VR for probably 5-10 more years.

 

[AbortedWalrusFetus]

When people say "soon" with regards to stuff like wireless displays, we are speaking in terms of a decade+, not months.

We use VESA standards currently as a dirty hack. The first foveated rendering solutions will work precisely as you'd expect, building a full VESA frame at the PC and sending it that way. This will not be a wireless display. When wireless VR displays arise, then the new standard will arise, and it'll work the way I am describing - using lower resolution frames dynamically composed at the headset to drastically reduce transmission requirements. There are only 3 conceivable ways to send information faster wirelessly - reduce the amount of information needed to be sent, increase the bandwidth of wireless transmission, or eventually, move to light-based transmission technologies.

Do you know what the practical upper bound on wireless data transmission is for this applicaiton? What is the bandwidth need for say, two 3840 × 2160 resolution images at 90-120fps? That seems like it would be maybe the ultimate goal of a VR headset.

 

[iceatcs]

What's about outdoor? For Vive, will it work for bright light?

I don't really have an empty space in my small garden flat for full potential. But I do have a quite big garden. 6ft x 8ft or 8ft x 14 ft flat area.

 

[Krejlooc]

Here's pictures of the Lighthouse that I just took off the wall. Krej, I'm not exactly sure what you're talking about. It'd be good if you not just put out statements as facts. We devs shouldn't mislead like that. There's no battery slots in this thing. Took these just now.

Where did I say the words "battery slots"? I said these could be run off of batteries, and they can, and are advertised as being able to do so. And have been demonstrated that way as well.

 

[pj]

Im not talking tomorrow, Im talking 20 years from now. Im sorry, I guess people think I am talking about CV1 lol. The mainstream will not buy into VR for probably 5-10 more years.

I don't think anyone has a good idea of what VR will be in 5 years, let alone 20

 

[Wrestlemania]

Where did I say the words "battery slots"? I said these could be run off of batteries, and they can, and are advertised as being able to do so. And have been demonstrated that way as well.

You said they "slot in normal batteries".

https://www.reddit.com/r/Vive/comments/42271m/slug/cz74ynb

If you want to run them off normal batteries you're going to need a lot of them, they're not "slotting in" and the battery life is not going to be fantastic.

 

[Croatoan]

What's about outdoor? For Vive, will it work for bright light?

I don't really have empty space in my small garden flat for full potential. But I do have quite big garden. 6ft x 8ft or 8ft x 14 ft flat area.

That...is actually an incredible idea. Instead of a room, get a tent you can put up and take down .

 

[Zalusithix]

When people say "soon" with regards to stuff like wireless displays, we are speaking in terms of a decade+, not months.

The conversation started by somebody saying wireless would maybe happen 3 generations down the line, and another person saying to expect it sooner than that. Assuming 2 year refresh periods, that means "soon" in context of the original statement is ~6 years to wireless or less. I'm not going to contest things a decade+ out.

 

[Cartman86]

#1 Rule of VR

Don't bring existing game design baggage to the discussion. You don't know what you want or what works until you experience it.

 

[Soi-Fong]

Where did I say the words "battery slots"? I said these could be run off of batteries, and they can, and are advertised as being able to do so. And have been demonstrated that way as well.

I have personally seen lighthouse running on normal batteries.



It isn't, the ones with the pre are much smaller and sleeker, and slot in normal batteries. That's from the early dev kits, the ones that also had Vive controllers that only lasted like an hour on a full charge.

It's right there on your post from the previous page. I'm not sure where you're getting at, but that's what you said on your post.

 

[pj]

Where did I say the words "battery slots"? I said these could be run off of batteries, and they can, and are advertised as being able to do so. And have been demonstrated that way as well.

"It isn't, the ones with the pre are much smaller and sleeker, and slot in normal batteries. "

If there isn't a battery slot, what is the battery slotting into???

 

[Croatoan]

I don't think anyone has a good idea of what VR will be in 5 years, let alone 20

Was hoping someone could point me towards stuff like the redirected walking study, or the stanford/nvidia near light field displays. I already know current VR is nowhere near what I expected, but I was looking for hope in the future.

#1 Rule of VR

Don't bring existing game design baggage to the discussion. You don't know what you want or what works until you experience it.

You telling me people are not going to want to move around in a giant world and kill stuff? That is like the most human thing to do in VR. LOL

 

[Krejlooc]

Do you know what the practical upper bound on wireless data transmission is for this applicaiton? What is the bandwidth need for say, two 3840 × 2160 resolution images at 90-120fps? That seems like it would be maybe the ultimate goal of a VR headset.

That's not the ultimate goal of a VR headset, actually. The bandwidth requirements for displays of that size, for the framerates people talk about (and realistically, we're going to increase our framerates going forward) is too high for any conventional wireless technology.

But we don't need to send full frames, because once we can track our fovea at the headset, we can significantly dismiss the vast majority of the resolution of the image being sent. We only see a tiny dot of our vision in crystal clarity. If we can send that single tiny dot at a high resolution (which would still be a tiny image), then we could send the rest of the frame at a much lower resolution to get the bandwidth requirements down.

 

[Compsiox]

"It isn't, the ones with the pre are much smaller and sleeker, and slot in normal batteries. "

If there isn't a battery slot, what is the battery slotting into???

Speaking on behalf of Krejlooc it was obviously a simple mistake

Spoiler

I love you Krejlooc. Just pokin fun

 

[Krejlooc]

You said they "slot in normal batteries".

https://www.reddit.com/r/Vive/comments/42271m/slug/cz74ynb

If you want to run them off normal batteries you're going to need a lot of them, they're not "slotting in" and the battery life is not going to be fantastic.

Then I misspoke. "Slot in" as in "plug in." We ran the vive demo on NiMH batteries for the entire several hours.

 

[iceatcs]

That...is actually an incredible idea. Instead of a room, get a tent you can put up and take down .

True, I might get something like that shade tent if the sunlight clash with Vive. The lighthouse attaching on the legs

Good thing that I live on the ground floor, my computer is next to windows can lead for the cables to outside.

 

[AbortedWalrusFetus]

That's not the ultimate goal of a VR headset, actually. The bandwidth requirements for displays of that size, for the framerates people talk about (and realistically, we're going to increase our framerates going forward) is too high for any conventional wireless technology.

But we don't need to send full frames, because once we can track our fovea at the headset, we can significantly dismiss the vast majority of the resolution of the image being sent. We only see a tiny dot of our vision in crystal clarity. If we can send that single tiny dot at a high resolution (which would still be a tiny image), then we could send the rest of the frame at a much lower resolution to get the bandwidth requirements down.

So with foveated rendering and eye tracking the bandwidth requirements would be vastly lower then. I was thinking about the bitrates of 8k video and I thought I had read that they were at something like 700Mbit/s with a low compression ratio, and I know we have gigabit wifi at this point, I just wasn't sure what was the art of the possible.

 

[Monger]

What's about outdoor? For Vive, will it work for bright light?

I don't really have an empty space in my small garden flat for full potential. But I do have a quite big garden. 6ft x 8ft or 8ft x 14 ft flat area.

It's not a recommended setup but I saw a picture of it being done. Just be careful not to fry your screen by exposing the lenses to the sunlight.

 

[Wrestlemania]

Then I misspoke. "Slot in" as in "plug in." We ran the vive demo on NiMH batteries for the entire several hours.

"Several hours" is really not very long. Not long enough to make it worth avoiding finding a wired power solution at home, even if it does involve a bit of annoying cabling.

 

[Krejlooc]

You telling me people are not going to want to move around in a giant world and kill stuff? That is like the most human thing to do in VR. LOL

If doing that stuff doesn't work well with VR, then no. People will want to do what works well in VR, and will want to avoid what does not work. There are many ways to build an enjoyable experience. Open World COD clones are not the only thing people can and will do for fun.

I expect many passive VR experiences to become popular, things like Sightline the chair, or streaming sports experiences.

 

[Krejlooc]

"Several hours" is really not very long. Not long enough to make it worth avoiding finding a wired power solution at home, even if it does involve a bit of annoying cabling.

I will be running my Lighthouse beacons wirelessly, and keeping them on camera stands at that.

 

[collige]

Ok cool, now lets get realistic. Out of the people who will buy VR, how big is the audience going to be for a room big enough to pull that off? I am going to guess pretty small. From that paper it says once you start dropping below the optimal size people start noticing the NASCAR effect. People in this very thread don't have space for Valve's room scale and we are enthusiasts. Houses are not going to magically get bigger in the future (or grow extra rooms).

Note that the paper is 15 years old.

 

[The Elite]

Maybe if more people had the DK2, more people would desire room scale. The tracking space is very limiting and leaves you wanting so much more.

This is less about desire and more about realistic expectations of the user play space. I have watched the hover junkie vid with them showing off all sorts of limited environments and I am still not convinced that any of those represent the multitude (thousands) of possible scenarios people could face with adapting their enviornment to room scale.

Also, I don't want this to be an Oculus vs Vive thing because this discussion needs to be more forward thinking than that.

Some of the conversations about Room Scale have me extremely excited I Just feel like there are those who aren't really seeing the potential limitations of real world situations. That's not a knock on the tech, that's just reality.

 

[bigwinnerx]

So I have my computer setup in our 5th bedroom and looks like I have probably a 9x6 rectangle of space without shifting things around (except for moving a chair to the side)

I have electrical sockets under 2 diagonal corners where I can put the Lighthouses.

Think I'm ready.

 

[pj]

Then I misspoke. "Slot in" as in "plug in." We ran the vive demo on NiMH batteries for the entire several hours.

How did vive demo on nimh batteries for several hours become the "you can run lighthouses off a single normal battery for weeks at a time"?

You need to be more precise with your language or people will get the wrong idea (as Wrestlemania did) and end up disappointed.

The fact is that it is not really practical to run lighthouses off of batteries unless you need to move them around a lot.
1. You still have a cable running from the lighthouse to the floor
2. Any battery cable of powering the lighthouses for a significant amount of time is going to be quite large/expensive
3. You still have to have power nearby to charge the batteries

It's a lot of added expense and hassle for no real benefit to 99% of people.

 

[iceatcs]

It's not a recommended setup but I saw a picture of it being done. Just be careful not to fry your screen by exposing the lenses to the sunlight.

Live in London, and it will be in shadowed all the time because of 3 storeys building surround but opened at north which away from the sun. I will get a tent in case sudden raining heh.

Wind breeze going to be fun.

 

[Croatoan]

If doing that stuff doesn't work well with VR, then no. People will want to do what works well in VR, and will want to avoid what does not work. There are many ways to build an enjoyable experience. Open World COD clones are not the only thing people can and will do for fun.

I expect many passive VR experiences to become popular, things like Sightline the chair, or streaming sports experiences.

Nvidia claims Near Light Field Displays could dramatically decrease the chance of sickness. Won't foveated rendering also help?

We could get rid of it, or hell, get use to it. The human body is able to adapt to changing conditions. We got use to cars, why wouldn't we get use to VR?

You seem to claim we will never get these experiences. I think we will and we will get them with a traditional style joystick on a controller held in our left (or right) hand. Once this "Room scale tracking" stuff goes the way of the dodo.

Either that or we will all start buying omni-directional treadmills. Would do wonders for America's obesity problem (I plan on getting one once the tech matures some).

The way I see it you would close off a 6 x 6 space in your house, hang a harness from the ceiling and put the treadmill on the floor. The harness will keep you in the middle of the treadmill and keep you from falling. Haptics on your hands and body will allow you to feel the world around you.

 

[viveks86]

Then I misspoke. "Slot in" as in "plug in." We ran the vive demo on NiMH batteries for the entire several hours.

Several hours? Wow. Might consider these, then.

Got pics of these batteries?

 

[Cartman86]

Was hoping someone could point me towards stuff like the redirected walking study, or the stanford/nvidia near light field displays. I already know current VR is nowhere near what I expected, but I was looking for hope in the future.



You telling me people are not going to want to move around in a giant world and kill stuff? That is like the most human thing to do in VR. LOL

People will want to do all sorts of things I imagine at some point. In my opinion what we've managed to do right now may be enough to start with. The second you have the experience that gives you a thousand foot open free space then yeah that will be the standard we can't come back from, but I think the experiences people are creating around that comparatively limited space are interesting enough that most people who've tried a Vive can see themselves getting one in the not to distant future. I don't think that impulse involves pondering about technologies from 15+ years in the future. They are thinking of money.