• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Limitations of VR - a virtual reality check

AniHawk

Member
You may say I'm a dreamer, but... I love the comparison to retrogaming. Technical limitations gave us amazing games during the 8-bit and 16-bit eras. VR limitations might usher in a new era of innovation in game design (instead of VR Skyrim, VR Assassins Creed and stuff like that).

i agree with this. i mean there's something to be said for achieving dreams, but a lot of happy accidents happen out of a lack of resources. humans are pretty neat in that they're adaptable and can solve problems.

as for this kind of vr, i liken it to early cd add-ons. i mean, i see this as the precursor to the next generation where everything starts to click in a major way. and 20 years from then we'll look back at how primitive that was.
 
Disagree. Give me interstate 76 graphics, in VR... And I'm lost in the world. VR doesn't have to be some photorealistic pixel pushing realm. Keep it simple, let us get lost in the world, the story, the immersion. I'd love to inhabit polygon land.

interstate-76_intro_2.jpg

Rewind gaming.

Good point, that game was fucking incredible.




Haha
 
If anything all of the pieces for consumer VR have been readily available for 5 or so years now. We've just been waiting for someone to put them together.

Everything else I agree with.

It's been out there, but all with the explicit warning of "wait for the consumer version unless you're a dev!". This next year is when major companies will have a reason to sell us on VR. It's also when the first retail games explicitly made for VR will start popping up. That's going to start a train, and it'll be hard to stop to explain people why they can't play Battlefront on the $200 headset they just bought.
 

Seanspeed

Banned
It's been out there, but all with the explicit warning of "wait for the consumer version unless you're a dev!". This next year is when major companies will have a reason to sell us on VR. It's also when the first retail games explicitly made for VR will start popping up. That's going to start a train, and it'll be hard to stop to explain people why they can't play Battlefront on the $200 headset they just bought.
There will be people who don't understand, but I think the VR experiences we will get will be so compelling that it's not going to cause any huge backlash to it. People are going to be highly, highly impressed with this technology when they get their hands on it.
 
There will be people who don't understand, but I think the VR experiences we will get will be so compelling that it's not going to cause any huge backlash to it. People are going to be highly, highly impressed with this technology when they get their hands on it.

Nah man, I agree with this:

it'll be hard to stop to explain people why they can't play Battlefront on the $200 headset they just bought.

People don't have the imagination and experience of various science-fiction text to be able to understand why this is important. They'll see it in its early incarnation and not be able to understand where it is going to go.

This is going to be hard as fuck to sell to people. It's basically marketing on hard mode.
 
You may say I'm a dreamer, but... I love the comparison to retrogaming. Technical limitations gave us amazing games during the 8-bit and 16-bit eras. VR limitations might usher in a new era of innovation in game design (instead of VR Skyrim, VR Assassins Creed and stuff like that).

This. I want innovation in Game design instead of playing same games we play now on VR. I'm sure we are going to see games that innovate and play different to overcome these limitations in some way. Look at Nintendo how they use art and game design in their games to better use the power of the hardware and hide the technical limitations.
 

Seanspeed

Banned
Nah man, I agree with this:



People don't have the imagination and experience of various science-fiction text to be able to understand why this is important. They'll see it in its early incarnation and not be able to understand where it is going to go.

This is going to be hard as fuck to sell to people. It's basically marketing on hard mode.
I really think once people try it, it wont be a hard sell at all. Having a friend who has it, or a test station somewhere, or at an expo, or just plain word-of-mouth should go a long way in convincing people it's worthwhile. It wont capture 100% of the market, no, but I think it's going to be soooo impactful for those who do try it that a decent level of popularity is pretty much inevitable.

I cant wait til these headsets come out. There is going to be a flood of people gushing on here, I guarantee it.
 

2+2=5

The Amiga Brotherhood
Wow i didn't think VR was that limiting! I thought it was just a little more demanding than a stereoscopic 3d game!
 
My position is that I think we are at minimum, five years out from decent consumer level VR. By consumer level, I mean something that
- Meets minimum consumer expectations of what a modern videogame looks like, and is capable of
- Is a comfortable experience to wear & use
- Can be bought for the highest end of a reasonable consumer electronics purchase

But we're forcing this stuff to market way too early. Morpheus & PS4 is not going to be enough to meet consumer expectations of VR. If I sell a headset promising "first person VR" on the machine that plays COD, No Man's Sky, and Fallout - people are going to expect the ability to play those games, with that hardware.
My worry is that Morpheus is going to emerge with a massive marketing campaign & public awareness, only to inevitably disappoint the mass market & re-sour them on VR. So in five years, when that acceptable version IS ready - it's going to be that much harder of a sell.

Maybe Sony can prove me wrong at E3. But from the sounds of this, it will take a miracle.

That might be true of some gamers right now but that is only because they haven't even seen any Morpheus games yet. That is not going to be true after E3. That's why Sony is devoting half their booth to Morpheus. People are going to experience and get first hand account of exactly what Morpheus games will look like and that will set expectations. They will also get numerous accounts of what VR in Morpheus feels like which I expect to come back as glowing reviews based on past accounts. That too will set expectations.

As I've said numerous times when the quality of Sony's VR is brought up. Minecraft sold more that 18 million copies on PC and the consoles have sold even more than that. A Minecraft type game on Morpheus is definitely doable and would be a killer app. Just look at the reception it received when being demoed on Microsoft's HoloLens. The limit to a game's success on Morpheus will be the devs creativity and not the specifications of the hardware.
 
I really think once people try it, it wont be a hard sell at all. Having a friend who has it, or a test station somewhere, or at an expo, or just plain word-of-mouth should go a long way in convincing people it's worthwhile. It wont capture 100% of the market, no, but I think it's going to be soooo impactful for those who do try it that a decent level of popularity is pretty much inevitable.

I cant wait til these headsets come out. There is going to be a flood of people gushing on here, I guarantee it.

It feels like such a futile discussion, at the moment, since we are at the verge of some surely impactful reveals.

I just know it's an uphill battle, but I agree with you, I finally had a chance to try it when I was in Japan last month, and it certainly felt like I was looking at the ghostly future.

I've just seen a lot of promising technologies falter due to consumer disinterest, so i'm a bit jaded.
 
Wow i didn't think VR was that limiting! I thought it was just a little more demanding than a stereoscopic 3d game!

The upside to those high demands is an increased perceived resolution, since the motion of your head blurs the lines of aliasing.

Among all the other benefits of VR, of course
 

Seanspeed

Banned
It feels like such a futile discussion, at the moment, since we are at the verge of some surely impactful reveals.

I just know it's an uphill battle, but I agree with you, I finally had a chance to try it when I was in Japan last month, and it certainly felt like I was looking at the ghostly future.

I've just seen a lot of promising technologies falter due to consumer disinterest, so i'm a bit jaded.
I understand. And people see peripherals like Kinect or Move falter and assume this will happen with Morpheus, too. I get why people make these connections. I just don't think this is going to be the same at all. It's not being done half-assed like before. Developers are actually excited about this stuff. And what it brings to the table goes so far beyond what these other technologies could on their own. It is going to be so unlike anything people have seen before and it will get people's attention. I am 99% sure of it.
 
I understand. And people see peripherals like Kinect or Move falter and assume this will happen with Morpheus, too. I get why people make these connections. I just don't think this is going to be the same at all. It's not being done half-assed like before. Developers are actually excited about this stuff. And what it brings to the table goes so far beyond what these other technologies could on their own. It is going to be so unlike anything people have seen before and it will get people's attention. I am 99% sure of it.

At the least there is a lot of money running behind each player in this arena, so we can expect the kind of long term commitment needed to be able to improve overall mind-share.
 

kyser73

Member
The fact people thought there would actually be Skyrim like games for Morpheus or Vive were kidding themselves. Until the tech has grown and adapted we'll be eating very limited games that rely on the gimmick of virtual reality to sell. I'm actually more excited about augmented reality then virtual reality right now because of how much of an impact that has on real life. Almost like a "no strings attaches" kind of sentiment.

AR is still decades away. We're barely started on the Internet of Things which is a bare minimum for real AR.
 

Branduil

Member
I think when they get foveated rendering down, VR will really explode. Just because it will make all current hardware seem more powerful by an order of magnitude, without having to actually replace it. You'll be able to make games that compete with the most realistic real-time visuals today, with a much higher framerate and perceived resolution, let alone the the advantages of VR itself.
 

Thrakier

Member
My position is that I think we are at minimum, five years out from decent consumer level VR. By consumer level, I mean something that
- Meets minimum consumer expectations of what a modern videogame looks like, and is capable of
- Is a comfortable experience to wear & use
- Can be bought for the highest end of a reasonable consumer electronics purchase

But we're forcing this stuff to market way too early. Morpheus & PS4 is not going to be enough to meet consumer expectations of VR. If I sell a headset promising "first person VR" on the machine that plays COD, No Man's Sky, and Fallout - people are going to expect the ability to play those games, with that hardware.
My worry is that Morpheus is going to emerge with a massive marketing campaign & public awareness, only to inevitably disappoint the mass market & re-sour them on VR. So in five years, when that acceptable version IS ready - it's going to be that much harder of a sell.

Maybe Sony can prove me wrong at E3. But from the sounds of this, it will take a miracle.

Virtual Reality isn't a gaming technology.
 

efyu_lemonardo

May I have a cookie?
The biggest difference is really 11ms frametimes maximum, always, no exceptions (ever).

That's vastly different from how most games run on consoles, and how most people run their PC games.

Edit: Oh, and all those deep pipelines you use to get better parallel scaling? Forget about those, too much latency.

This seems like a more appropriate thread to post a question a recently posted in the "stupid gaming questions" thread.

I'd love to be educated on the matter, if there's a relatively simple explanation, or pointed in the direction of some reading material if it's more complicated.

Here's my original post:

Question about frame rates

I know next to nothing about this, but recently was able to listen in on a discussion between people much more knowledgable than I about the difficulties of maintaining very high framerates for VR. For the purpose of this question, let's say we're rendering at a resolution of 1080p per eye, and we want a frame rate that averages somewhere between 90 and 120 fps, ideally with a minimum frame rate of 90 fps.

Obviously tradeoffs are necessary in terms of geometrical complexity, lighting, shading etc, but sometimes I get the feeling that part of the difficulty for VR maybe stems from developers and consumers trying to set the graphical bar a bit too high. This is where I'd appreciate it if GAF could help educate me on the matter.

What's bothering me is I seem to remember computers from well over a decade ago being able to achieve ridiculous framerates (300+ fps when uncapped) running Quake 3 or similar titles. Yes, the resolutions were no more than 1K whereas 1080p per eye is 4K. Yes, Quake 3 level graphical complexity is totally unacceptable even for a mobile game today. But if I tried to extrapolate that linearly to account for the drastic improvements in performance since then, I could come to the conclusion that a machine today would be able to run a version of Quake 3 that was properly optimized for today's hardware at 4K at an insane speed, somewhere between 500 and 1000 fps, or 1-2 milliseconds per frame. (Note: Which sounds rather ridiculous, to be clear).

I remember a talk by John Carmack a while ago (before he joined Oculus) where he talked about how even Doom 3 felt amazing in VR, and that it wasn't necessary to have the most complex graphics in order to feel immersion and presence. And with VR adaptors coming to smartphones it would appear others in the industry agree.

So basically my question is this: Has there been some fundamental change in rendering pipelines over the last 5-10 years that has made it that much more difficult to achieve such high framerates without significantly cutting back on important graphical features? As in, maybe many popular graphical features today are much more expensive than popular graphical features were a decade ago?
Or perhaps the bottleneck isn't in the graphics department but in other areas such as game logic, AI, I/O, etc? Because I've never heard of a pc running Crysis at anywhere near the same speeds that people were running Quake 3 a mere 3-4 years after it released, and that game was 8 years ago.

Please help me understand GAF, your assistance is appreciated!
 
Without the heavy reliance on graphics, game design and game mechanics will have to be really on point to sustain interest. There are very few developers in the industry that understand and can make compelling game experiences that don't really on the latest and greatest technology/rendering techniques.

Making the technical constraints tighter for designers will not improve matters as most of them are incapable of producing compelling game experiences with more resources at their disposal.

I expect 99% of VR 'games' to suck and I'm not interested in any of them in the slightest. The ones that do succeed are likely to be heavily cinematic, single playthrough, 'experiences' which I have no interest in either.

Welcome to the 'future'...
So you think that both the studios producing triple AAA games will fail at making games because they can't properly innovate, and you expect the only successes to be non-innovative games made by those same producers.

That seems like an odd position to have.
 
Even playing HL2 VR with its 2004 graphics on the DK1 blew my mind, I was there in the world and the textures and polys being much less than modern barely even registered (the screendoor on the other hand... that was a bother, but that's not going to be a big deal for long). I hope devs don't shy away from developing meaty games like that with lower budget/horsepower graphics.
Huh? I run HL2 VR on my laptop using the DK2 and never experience a dropped frame.

OP said:
Half Life 2 VR? The game we work on? It stresses my PC like hell. Our lead modeler, Jazz, is constantly redesigning things like the gun models to remove additional polygons to get it running acceptably. This is a game from 11 years ago, and it can barely run in VR with a ton of reworking. VR is so hard to work with that people honestly would be surprised what little power you actually have left over once you begin designing your game.
I'm having a tough time squaring these comments.
 

ZanDatsu

Member
These are all short-term problems and are only being worried about by people who are thinking linearly rather than exponentially. It won't take long to get to the point where these issues are a thing of the past, totally forgotten about. And even if these initial devices flounder completely that means nothing, the technology will reach a mass audience one way or another given time.
 

efyu_lemonardo

May I have a cookie?
Uh, MSAA is actually extremely important for virtual reality. Aliasing is hell in VR, it creates shimmers and sparks in your eye. Valve's VR design bible recommends using 8xMSAA if at all possible as a minimum.

This is something I didn't know, and helps explain why achieving consistent high framerates seems to be more difficult in VR. Are there other factors contributing to this issue?
 

Majukun

Member
if it were for some people we would have never eveloped smartphones because the first mobiles were huge,impractical and with poor reception
 

hesido

Member
These are all short-term problems that are being talked about by people who are thinking linearly rather than exponentially. It won't take long to get to the point where these issues are a thing of the past, totally forgotten about. And even if these initial devices flounder completely that means nothing, the technology will reach a mass audience one way or another given time.

There are some people who think a prematurely released VR platform could damage reputation of VR and make the further iterations a harder sell. I think they are right in the fact that a bad VR launch can hamper next iterations but I don't think this would be the case. I think Morpheus will amaze a lot people at E3 (haven't tried ANY vr yet) We'll see about that in a few days.
 
Slightly Mad said about Project Cars - "So getting Project Morpheus up-and-running was relatively painless. Not easy as that would be doing our tech team a disservice, but relatively painless.”

It doesn't reveal what kind of sacrifices they made to get there, but it gives a hint as to what's possible.

I'm not so naive to believe that the graphics in VR will be as good as those on a standard display, but I'm willing to bet the experience will be on a whole new level and that's what I'm really interested in.

http://vrfocus.com/archives/9879/slight-mad-studios-getting-project-morpheus-running-relatively-painless/
 

Mula

Member
I think this virtual reality check needs a reality check;)
I don't think the ps4 can't run a M64 like game in VR
 
A friend of mine put together a pretty crappy looking demo in Unity for a F1 car and then just a bunch of objects you walk around, using the original Rift devkit (I don't know all their fancy code names, but I could basically see the pixels) - and it was still tremendously engaging. One of the objects he'd put in there was an Airbus A380 and I'd never felt a sense of scale that well before. And when I walked through a model of a car with no clipping model, it felt very disturbing. I echo the comments of the people saying that the graphics don't need to be amazing to make incredible experiences. When the hardware catches up, so much the better. But it's not like it'll be some gimped piece of shit until then. For developers used to having, as he said, near infinite resources, I can see why it might be a big concern - but really, I don't think those sorts of high-end graphics are required in irder to wow people in the way it currently is.
 

bobbytkc

ADD New Gen Gamer
I don't know what the OP is going about. Alien Isolation has VR support and it played fine when I tried it on my PC. This is a modern game.
 

RoadHazard

Gold Member
So would we be willing to say comfortably, that the PS4 could power a game like Wind Waker, at 60hz, in the Morpheus's FOV with little issue? These are the kinds of graphics that I'm mentally preparing for.

Should get much, much better visuals than that. Sony's recommendation to devs is to start out targeting PS3-level visuals at 1080p60 and work their way up from there. Don't know why they would recommend things that aren't feasible.
 

2+2=5

The Amiga Brotherhood
I'm sorry for the evident ignorance, but why there's so much difference between vr and 3d power requirements?

What do vr needs to do rendering-wise that 3d doesn't?
What's the difference between vr and two simple screens showing the same scene from different angles?
 

UrbanRats

Member
I'm sorry for the evident ignorance, but why there's so much difference between vr and 3d power requirements?

What do vr needs to do rendering-wise that 3d doesn't?
What's the difference between vr and two simple screens showing the same scene from different angles?

VR needs to go at very high framerates, for one, and with even minor frame drops, that wouldn't be a problem on a regular screen, you can get nausea, which is a major issue with VR.
 
VR needs to go at very high framerates, for one, and with even minor frame drops, that wouldn't be a problem on a regular screen, you can get nausea, which is a major issue with VR.

Additionally, because of the high frame rate, that reduces the computation time for stuff like AI, physics etc (which the article in the OP notes), so it's not a purely graphical thing.
 

efyu_lemonardo

May I have a cookie?
Additionally, because of the high frame rate, that reduces the computation time for stuff like AI, physics etc (which the article in the OP notes), so it's not a purely graphical thing.

This is probably another thing I missed when asking about 300+ fps in uncapped Quake 3 over a decade ago, as those were probably benchmarks with no AI, game logic, etc.
 

2+2=5

The Amiga Brotherhood
VR needs to go at very high framerates, for one, and with even minor frame drops, that wouldn't be a problem on a regular screen, you can get nausea, which is a major issue with VR.

Additionally, because of the high frame rate, that reduces the computation time for stuff like AI, physics etc (which the article in the OP notes), so it's not a purely graphical thing.
Thans guys! :)
When you say high framerates what do you mean precisely? 120fps? 240? More?

This is probably another thing I missed when asking about 300+ fps in uncapped Quake 3 over a decade ago, as those were probably benchmarks with no AI, game logic, etc.
That's why i say to not trust "in-engine" trailers, they have nothing of the complexity of a real gameplay.
 

efyu_lemonardo

May I have a cookie?
That's why i say to not trust "in-engine" trailers, they have nothing of the complexity of a real gameplay.

True. For some reason when discussing frame rates I automatically think of rendering being the bottleneck. That's just a misconception I should get rid of.
 
Thans guys! :)
When you say high framerates what do you mean precisely? 120fps? 240? More?

90 fps, for the Rift and Vive.

The difficulty scale is this

1. 30 fps games.
2. "60" fps games, which actually they are "up to 60fps", their average framerate is 50-55 fps.
3. True 60 fps games, which they still can have drops.
4. 90 fps games.
5. 90 fps games without drops.

Most games are 1) or 2), now VR needs 5).
 

spekkeh

Banned
I'd argue there's a *very* specific and more powerful form of presence when talking about VR, and that equating it with any sort of 'in-the-moment' sort of feeling when reading a book or even playing a normal game just isn't fair whatsoever, being quite incomparable.

I also think that presence *can* be something that you cross over into. It may not be any specific line, but a lot of people who own and use their VR headsets regularly report that certain experiences can certainly trigger a change where they go from being super immersed to essentially being there. I mean, even something like the flicker fusion threshold, this is an important barrier that needs to be crossed before your brain accepts motion as real. I do think 'presence' will be an ever changing standard though, where if you get used to a certain level of VR, going backwards can result in breaking presence, while going forwards can introduce a new 'presence' that you hadn't known existed before, on and on basically until we get holodecks. But I think we are at a level where we can create convincing enough VR experiences that it really is worth describing it like being transported to someplace entirely new. It's something that could never in a million years be achieved through flatscreen gaming no matter how good the graphics get.

I'm not sure how you'd differentiate the effect (and even though I haven't kept up with recent presence research, I'd hazard a guess no scientist would). Some people said a physical reaction that your lizardbrain gets to the imagery, but for sure we have suspension of disbelief in many entertainment media, leading us to become aroused in action sequences, queasy or anxious when there's e.g. gore in a horror movie etc., and you also have the wellknown story of people running out of the cinema during L'Arrivée d'un train en gare de La Ciotat, because they thought they were going to get run over by a train.
 
This is probably another thing I missed when asking about 300+ fps in uncapped Quake 3 over a decade ago, as those were probably benchmarks with no AI, game logic, etc.

The more I think about it, the more I'm sure some of that stuff could be tweened. It would reduce accuracy a little, but if a barrel is flying around because a grenade went off next to it, it could opt to only do a collision check every other frame. On the non-checked frames, it just continues with its previous velocity and vector. This would lead to the possibility of an object intersecting with some geometry it should bounce off (or otherwise hit) on its "off" frame, but given the high frame rate that might not be a problem unless it's an incredibly fast moving object. You may be able to do this with AI too, because I can't think of too many actions that will require 90 alterations a second (ie if a guard reacts to your loud gunshot 1/90th of a second later than they should, that might not be a problem), but who knows.
 
The most interesting part of VR isn't trying to replicate what we already have but what new experiences VR can deliver. Restrictions on a medium is a good thing as it forces people to get out of their comfort zone and rethink traditional approaches, it is when people hit a brick wall they come up with solutions to navigate it. This is why you hire the creative minded to come up with solutions to problems.

Sony has been straight forward with the audience since revealing the Morpheus that they think traditional game experiences do not work as well in VR, I personally agree after my small amount of time with the Rift DK1 especially considering the limitations of the device. For me it isn't trying to replicate walking around an expansive world as essentially I will be sitting down with a controller in my hand (I personally would like to see a VR experience tackle disability due to this), but rather now there is the ability to bring the sense of immersion how is this then capitalised upon with the limitations. The possibilities are endless but yes they are going to be different than what you are used to and something you fantasise about might just be a crap experience just like everyone that wanted to wield a sword/lightsaber with a Wii controller or Kinect found out.
 
I'm not sure how you'd differentiate the effect (and even though I haven't kept up with recent presence research, I'd hazard a guess no scientist would). Some people said a physical reaction that your lizardbrain gets to the imagery, but for sure we have suspension of disbelief in many entertainment media, leading us to become aroused in action sequences, queasy or anxious when there's e.g. gore in a horror movie etc., and you also have the wellknown story of people running out of the cinema during L'Arrivée d'un train en gare de La Ciotat, because they thought they were going to get run over by a train.

This is all very true. And it's quite interesting how a lot of people say that, when they watch stereo films, after the first few minutes they basically forget it's there. Does that mean they're nice and engrossed in the film and the effect is comfortable that they're "transported" sufficiently that it ceases to be something they think about? Or does it mean that it's such a meaningless effect that they simply don't consider it at all beyond the initial spectacle? And, really, is there any difference between these two positions? Given how readily we find older games harder to get into due to their primative graphics (even though we didn't at the time), it does seem to me that there's a linear march that it seems unhelpful to make distinctions within. The only thing that seems relatively obvious to me is that the "presence" we feel today with a DK2 probably isn't going to be exciting anyone in 15 years.
 

ASIS

Member
yeah I have to say all this talk makes me more excited for VR than before.

of course there will be limitations, of course the first gen will not be perfect by a Longshot.

it means that this will be a long term pln with noticable improvements in our life time. to me that's a great thing!
 

pottuvoi

Banned
Disagree. Give me interstate 76 graphics, in VR... And I'm lost in the world. VR doesn't have to be some photorealistic pixel pushing realm. Keep it simple, let us get lost in the world, the story, the immersion. I'd love to inhabit polygon land.

interstate-76_intro_2.jpg

Rewind gaming.
+1
Mech Warrior 2 would be incredible as well.

The quality of early VR will be quite bit different from what he is suggesting though.
Low level APIs will help a lot.
 
Top Bottom