6 ms·

The VR Winter

28 points by 1cvmask 7y ago

oddity 7y ago
These are very good questions and I absolutely agree with last paragraph but it's far too early to conclude we're in for a VR winter. > There’s nothing fundamentally illogical about any of these ideas, but they do remind me a little of Karl Popper’s criticism of Marxists - that when asked why their supposedly scientific prediction hadn’t happened yet they would always say ‘ah, the historical circumstances aren’t right - you just have to wait a few more years’. This is the necessary position when arguing for the inevitability of something that hasn't happened yet. That the arguments are structurally similar isn't evidence for or against. The arguments themselves are what need to be examined. In the past 20 years, we went from human-narrated MapQuest print-outs from a computer at home to live directions on a tiny computer in our pocket narrated to us via a tiny computer in our ears. For a long time, I held out from upgrading to a smartphone. It was only when I was lost in a city in 2013 and had no idea how to get back home without calling someone many miles away that I finally caved. No one built the entire combination of smartphone, LTE, or voice assistants right away because they wanted a better MapQuest, the pieces came together organically. > if ‘real VR’ needs something that’s ten or twenty years away, we’re in for another VR winter. The modern smartphone was ten or twenty years away from the world wide web, which was ten or twenty years away from the computer. Each of these technologies found a niche, however small, until they took over the world. Is it possible that enthusiasm for VR might wane and the market might die before it has a chance to grow? Absolutely, it's possible. However, the ideal form of the technology being a long way out doesn't make it inevitable. Tech doesn't need a killer-app to gain a permanent foothold. It helps, but it seems that it's the exception rather than the norm. Incrementally expanding niches is a much more predictable way to bring a product to market. To me, the big open question is whether or not the small niches VR has today can be grown enough to reach sustainability before the primary hardware vendors (Valve, Facebook, Microsoft, Apple) lose interest. That's a function of price. Given the strong overlap with the hot tech of today (phones, watches, headphones), price will be driven down regardless of how successful VR is. That lowers the threshold of how successful VR needs to be today. > Meanwhile, it’s instructive that now that we’re all locked up at home, video calls have become a huge consumer phenomenon, but VR has been not. This should have been a VR moment, and it isn’t. This isn't a very convincing argument. Devices that can do video conferencing are everywhere today. Zoom is free. Of course it's a phenomenon. Meanwhile, I can't buy a VR headset today without paying 50% above MSRP or waiting 8 weeks. Even then, I'm one of the lucky ones who hasn't lost their job yet and can thus entertain the idea of spending $$$ on entertainment. I would want to see market data following more high profile releases like HL Alyx or whatever Apple has planned, and look out for a trend in hardware vendors pulling back from VR.
perilunar 7y ago
> here was no killer app for the first PCs either, but they looked useful I thought VisiCalc (and Lotus 123) were the killer apps for early PCs?
blackrock 7y ago
Make a VR operating system for work. Video games will never achieve critical mass for VR. Although, the one video game that I thought could become the killer app for VR, is BeatSaber [1]. [1] https://www.youtube.com/watch?v=ZJcqhvadbkI&t=59 https://www.youtube.com/watch?v=ZJcqhvadbkI&t=59
- Analemma_ 7y ago
  VR content is limited by the size of the addressable audience and thus the potential ROI. If it’s pretty low for games, it’s dreadfully low for an entire operating system.
- eddhead 7y ago
  There's already Windows 10 for both VR and AR, but Microsoft is playing long with this and taking it very slow. It runs most ARM apps flawlessly (on Hololens at least) and is quite focused on the enterprise to approach the industry from there.
artfulhippo 7y ago
The Oculus Quest still has serious flaws in screen resolution, input latency, and controller ergonomics. It’s literally painful for me to use it. I get eye strain, a headache, and sore thumbs. VR is clearly useful but before it becomes mainstream it will have to have the hardware quality that we associate with Apple or at least BlackBerry (ie a high-quality input interface). I hear the Valve Index is better but I haven’t tested it yet.
- Wowfunhappy 7y ago
  Two notes, of equal importance: 1. The Index is a lot better. The higher resolution and refresh rate make a big difference. The Oculus Quest has a refresh rate of only 72hz, which we've known for years isn't enough for lots of people. 2. Comfort is highly dependent on the software you play. A lot of VR enthusiasts have been pushing for less comfortable modes of play over the past few years (because it allows games to do more), which is great if that's what they like, but I worry it's turning off more susceptible individuals. Using a sufficiently powerful PC to play The Lab on an Index will not make anyone feel ill.
- ChrisLTD 7y ago
  I’m really hoping that Oculus improves the weight balance of the Quest headset for V2. It seems to put all of the weight on my cheekbones no matter how I adjust the straps.
lowdose 7y ago
> Smartphones are broad and universal, whereas consoles are deep and narrow, and deep and narrow is a smaller market. VR is even deeper and even narrower, and so if we can’t work out a form of content that isn’t also deep and narrow. People are now playing those console games on their iPhone. The average paying iOS user spends $9.60 per month on in app purchases. The game market is huge, bigger than movies & tv combined. Maybe in the current form VR & AR is niche but that says nothing about how valuable the market can become. The lesson is you shouldn't predict potential market size of a technology from the first particular hardware implementation.
ChrisLTD 7y ago
I don’t see VR taking off anytime soon, but I also think it’s a bit tiresome to say that a product category isn’t successful unless it nears the ubiquity of the smartphone. By that measure, plenty (most all?) profitable and useful products are failures.
Jemm 7y ago
It is time we stopped trying to make VR for the broadest appeal. There are people who simply cannot handle the experience either from a motion sickness point of view or otherwise. What will save VR is games or experiences that focus on quality and immersion rather than trying to please everybody. The Playstation VR is a good example. It is specifically not for children under 12 yet many of the games seem to be targeted at people with that maturity level.
throwaway34241 7y ago
> I think there are maybe four propositions to think about. > Is it true that we are essentially almost there, and a bit more iteration of the hardware and the developer ecosystem will get us to a tipping point? > Are we where smartphones were before the iPhone? > Is there a fundamental contradiction between a universally accessible experience and a device that you put on your head that shuts out the world around you? > Or, by extension, is this the point - that ‘real’ VR needs some completely different device and that’s what would take it to universality? VR as HMD is narrow but VR as, say, neural lace is not? In addition to his four propositions to think about I would add one more. There's technology to represent other humans in VR (eye tracking, face tracking, body tracking) that basically doesn't exist yet as part of consumer VR systems. But with those technologies already present outside of consumer VR (and continued massive R&D spending by Oculus etc) they seem realistically achievable in the medium term. Will this make a significant difference in the VR adoption rate or enable new applications?
ksec 7y ago
I dont understand how all these Smart people got sold into VR, AR, and AV ( Autonomous Vehicles). We have long entered into the age of TDP Computing as a proxy of relative performance. We could not even make the application Work on a 1000+ Watt PCs. Why is it we think we could make it works in Sub 100W anytime soon. We need Ultra Hi-Res because of how close the pixel are to our eyes, we also need Ultra-High frame rate so we dont have motion sickness. And then we expect Photo Realistic graphics along with Ultra Low Latency which No one have solved yet as we are constantly trading latency for throughput. ( Good old Street Fighter II felt more responsive than today's fighting game ) We can estimate with certain degree of accuracy how long it would take to reach 80% of the project, but experience taught me the last 20% would somehow take the same amount of time as the first 80%. To put this into AR, VR, AV perspective, we are at least another 6-8 years away from seeing its usefulness.
- saurik 7y ago
  I got "sold on VR" because I tried it? Like, I have been super busy recently, and so haven't barely had time to stand up much less use VR, but for most of a year I used VR every single day. This article isn't even arguing that it doesn't work or anything: it is arguing that it might not be mainstream, which is entirely different... this is like asking "I don't understand how all these Smart people got sold into computers" back when computers were mainframes with a very limited market, as if until something is ready for a mainstream market no one should bother owning one or using it for anything :/.
  - whytaka 7y ago
    I'm pretty sure you can read GP's statement to mean "sold on the mainstream applicability of VR". No one doubts that some segment of the market has purchased VR devices.
- baxuz 7y ago
  There are ML driven shortcuts, like foveated rendering, DLSS and other techniques: https://www.youtube.com/watch?v=eTUmmW4ispA https://www.youtube.com/watch?v=eTUmmW4ispA
LordHeini 7y ago
Who whatches Opera? But there are still opera houses and nobody complains about them... Vr in its current state is mostly a toy, it is used to play games and watch porn. Much like a lot of succesffull tech that came out actually. Games drove the pc (and the smartphone) to the powerhouse it is. And do not underestimate the porn argument. This is what all the people i know commented on and one of the few things with an abundance of content. So in my book the basics for some form of success are set. The enthusiast status will slowly fade and vr system will slowly seep into the mainstrean like game consoles did. For certain enthusiats vr is a revelation. If you play racing or flight sims, do yourselv a favor and get a good vr set. This "vr will never make it to mass adoption argument" because communism sucks is just weird by the way. Is there some form of goodwins law for marx?
bni 7y ago
Games did its fair share of making the PC mainstream. Many got a PC to be able to play Doom. Games today is mainstream. Everyone young plays games now, might not talk about it but everyone does. So I def think games can be a driver to make VR mainstream. Social media meets games in the other end. All it takes is VR to be both more immersive and more comfortable than playing games on flat screen.
istorical 7y ago
It always surprises me when people completely overlook the evolution TV and film will likely undergo to become stereo 180 (for directed attention experiences) or stereo 360 (for passive self-directed experiences) when HMDs become cheap enough that they become ubiquitous. Sure you can watch video without color, sure you can watch film without sound, but why would you? By the same sense, for most cases watching video/film/TV with 3D depth is going to be superior to watching without. The argument that 3D TV never took off is less convincing when you consider that content creation for 3D TV was way more costly than it is for stereo 180/360 VR today, and that 3D TV died before the chicken and egg problem was ever solved. (Even if the HMD market contracts 90% it should still be profitable for small Chinese manufacturers to produce crap-tier HMD for porn). Today a good stereo 180 VR experience can be filmed with an Insta360 Evo or a Vuze XR (my preferred option due to out of the box color grading) which can be had for about $400 USD. ZCam K1 pro and K2 pro can create incredible content (as seen in top pornography) for $3-4k entry price. But it's really hard to convince anyone who hasn't tried it, especially someone who hasn't seen something filed with one of the aforementioned cameras. Especially people who haven't tried VR or well-filmed VR but have tried 3D movies and assumes they are the same the same experience.
- renox 7y ago
  >By the same sense, for most cases watching video/film/TV with 3D depth is going to be superior to watching without. Uh? Except that to get 3D you have to wear glasses, which is a huge drawback. As for VR, I've tried it 3 times with demos: the first time I got so seasick that I had to stop and was feeling still quite bad 15 minutes after, the second time I nearly fell and the third time I bounced into a wall..
  - jeegsy 7y ago
    Sounds like your immersion level was increasing with every try
    - renox 6y ago
      It wasn't the same demo: the first time, it was a roller coaster (a really dumb demo!!), the second time walking on a roof and falling (interesting that it created some fear of heights, but I didn't expect to nearly fall for real during the virtual fall due to the change of POV)
wildermuthn 7y ago
VR hardware is good enough to go mainstream, but as pointed out, what is missing is a mainstream utility: a killer app like the spreadsheet or word processor. The 10x advantage that VR hardware provides over a 2D display isn’t 3D space. It is social presence. VR’s killer app exists within the domain of collaboration, communication, and organization.

28 points by 1cvmask 7y ago

These are very good questions and I absolutely agree with last paragraph but it's far too early to conclude we're in for a VR winter. > There’s nothing fundamentally illogical about any of these ideas, but they do remind me a little of Karl Popper’s criticism of Marxists - that when asked why their supposedly scientific prediction hadn’t happened yet they would always say ‘ah, the historical circumstances aren’t right - you just have to wait a few more years’. This is the necessary position when arguing for the inevitability of something that hasn't happened yet. That the arguments are structurally similar isn't evidence for or against. The arguments themselves are what need to be examined. In the past 20 years, we went from human-narrated MapQuest print-outs from a computer at home to live directions on a tiny computer in our pocket narrated to us via a tiny computer in our ears. For a long time, I held out from upgrading to a smartphone. It was only when I was lost in a city in 2013 and had no idea how to get back home without calling someone many miles away that I finally caved. No one built the entire combination of smartphone, LTE, or voice assistants right away because they wanted a better MapQuest, the pieces came together organically. > if ‘real VR’ needs something that’s ten or twenty years away, we’re in for another VR winter. The modern smartphone was ten or twenty years away from the world wide web, which was ten or twenty years away from the computer. Each of these technologies found a niche, however small, until they took over the world. Is it possible that enthusiasm for VR might wane and the market might die before it has a chance to grow? Absolutely, it's possible. However, the ideal form of the technology being a long way out doesn't make it inevitable. Tech doesn't need a killer-app to gain a permanent foothold. It helps, but it seems that it's the exception rather than the norm. Incrementally expanding niches is a much more predictable way to bring a product to market. To me, the big open question is whether or not the small niches VR has today can be grown enough to reach sustainability before the primary hardware vendors (Valve, Facebook, Microsoft, Apple) lose interest. That's a function of price. Given the strong overlap with the hot tech of today (phones, watches, headphones), price will be driven down regardless of how successful VR is. That lowers the threshold of how successful VR needs to be today. > Meanwhile, it’s instructive that now that we’re all locked up at home, video calls have become a huge consumer phenomenon, but VR has been not. This should have been a VR moment, and it isn’t. This isn't a very convincing argument. Devices that can do video conferencing are everywhere today. Zoom is free. Of course it's a phenomenon. Meanwhile, I can't buy a VR headset today without paying 50% above MSRP or waiting 8 weeks. Even then, I'm one of the lucky ones who hasn't lost their job yet and can thus entertain the idea of spending $$$ on entertainment. I would want to see market data following more high profile releases like HL Alyx or whatever Apple has planned, and look out for a trend in hardware vendors pulling back from VR.

perilunar 7y ago

> here was no killer app for the first PCs either, but they looked useful I thought VisiCalc (and Lotus 123) were the killer apps for early PCs?

blackrock 7y ago

Make a VR operating system for work. Video games will never achieve critical mass for VR. Although, the one video game that I thought could become the killer app for VR, is BeatSaber [1]. [1] https://www.youtube.com/watch?v=ZJcqhvadbkI&t=59 https://www.youtube.com/watch?v=ZJcqhvadbkI&t=59

Analemma_ 7y ago
VR content is limited by the size of the addressable audience and thus the potential ROI. If it’s pretty low for games, it’s dreadfully low for an entire operating system.
eddhead 7y ago
There's already Windows 10 for both VR and AR, but Microsoft is playing long with this and taking it very slow. It runs most ARM apps flawlessly (on Hololens at least) and is quite focused on the enterprise to approach the industry from there.

artfulhippo 7y ago

The Oculus Quest still has serious flaws in screen resolution, input latency, and controller ergonomics. It’s literally painful for me to use it. I get eye strain, a headache, and sore thumbs. VR is clearly useful but before it becomes mainstream it will have to have the hardware quality that we associate with Apple or at least BlackBerry (ie a high-quality input interface). I hear the Valve Index is better but I haven’t tested it yet.

Wowfunhappy 7y ago
Two notes, of equal importance: 1. The Index is a lot better. The higher resolution and refresh rate make a big difference. The Oculus Quest has a refresh rate of only 72hz, which we've known for years isn't enough for lots of people. 2. Comfort is highly dependent on the software you play. A lot of VR enthusiasts have been pushing for less comfortable modes of play over the past few years (because it allows games to do more), which is great if that's what they like, but I worry it's turning off more susceptible individuals. Using a sufficiently powerful PC to play The Lab on an Index will not make anyone feel ill.
ChrisLTD 7y ago
I’m really hoping that Oculus improves the weight balance of the Quest headset for V2. It seems to put all of the weight on my cheekbones no matter how I adjust the straps.

lowdose 7y ago

> Smartphones are broad and universal, whereas consoles are deep and narrow, and deep and narrow is a smaller market. VR is even deeper and even narrower, and so if we can’t work out a form of content that isn’t also deep and narrow. People are now playing those console games on their iPhone. The average paying iOS user spends $9.60 per month on in app purchases. The game market is huge, bigger than movies & tv combined. Maybe in the current form VR & AR is niche but that says nothing about how valuable the market can become. The lesson is you shouldn't predict potential market size of a technology from the first particular hardware implementation.

ChrisLTD 7y ago

I don’t see VR taking off anytime soon, but I also think it’s a bit tiresome to say that a product category isn’t successful unless it nears the ubiquity of the smartphone. By that measure, plenty (most all?) profitable and useful products are failures.

Jemm 7y ago

It is time we stopped trying to make VR for the broadest appeal. There are people who simply cannot handle the experience either from a motion sickness point of view or otherwise. What will save VR is games or experiences that focus on quality and immersion rather than trying to please everybody. The Playstation VR is a good example. It is specifically not for children under 12 yet many of the games seem to be targeted at people with that maturity level.

throwaway34241 7y ago

> I think there are maybe four propositions to think about. > Is it true that we are essentially almost there, and a bit more iteration of the hardware and the developer ecosystem will get us to a tipping point? > Are we where smartphones were before the iPhone? > Is there a fundamental contradiction between a universally accessible experience and a device that you put on your head that shuts out the world around you? > Or, by extension, is this the point - that ‘real’ VR needs some completely different device and that’s what would take it to universality? VR as HMD is narrow but VR as, say, neural lace is not? In addition to his four propositions to think about I would add one more. There's technology to represent other humans in VR (eye tracking, face tracking, body tracking) that basically doesn't exist yet as part of consumer VR systems. But with those technologies already present outside of consumer VR (and continued massive R&D spending by Oculus etc) they seem realistically achievable in the medium term. Will this make a significant difference in the VR adoption rate or enable new applications?

ksec 7y ago

I dont understand how all these Smart people got sold into VR, AR, and AV ( Autonomous Vehicles). We have long entered into the age of TDP Computing as a proxy of relative performance. We could not even make the application Work on a 1000+ Watt PCs. Why is it we think we could make it works in Sub 100W anytime soon. We need Ultra Hi-Res because of how close the pixel are to our eyes, we also need Ultra-High frame rate so we dont have motion sickness. And then we expect Photo Realistic graphics along with Ultra Low Latency which No one have solved yet as we are constantly trading latency for throughput. ( Good old Street Fighter II felt more responsive than today's fighting game ) We can estimate with certain degree of accuracy how long it would take to reach 80% of the project, but experience taught me the last 20% would somehow take the same amount of time as the first 80%. To put this into AR, VR, AV perspective, we are at least another 6-8 years away from seeing its usefulness.

saurik 7y ago
I got "sold on VR" because I tried it? Like, I have been super busy recently, and so haven't barely had time to stand up much less use VR, but for most of a year I used VR every single day. This article isn't even arguing that it doesn't work or anything: it is arguing that it might not be mainstream, which is entirely different... this is like asking "I don't understand how all these Smart people got sold into computers" back when computers were mainframes with a very limited market, as if until something is ready for a mainstream market no one should bother owning one or using it for anything :/.
- whytaka 7y ago
  I'm pretty sure you can read GP's statement to mean "sold on the mainstream applicability of VR". No one doubts that some segment of the market has purchased VR devices.
baxuz 7y ago
There are ML driven shortcuts, like foveated rendering, DLSS and other techniques: https://www.youtube.com/watch?v=eTUmmW4ispA https://www.youtube.com/watch?v=eTUmmW4ispA

LordHeini 7y ago

Who whatches Opera? But there are still opera houses and nobody complains about them... Vr in its current state is mostly a toy, it is used to play games and watch porn. Much like a lot of succesffull tech that came out actually. Games drove the pc (and the smartphone) to the powerhouse it is. And do not underestimate the porn argument. This is what all the people i know commented on and one of the few things with an abundance of content. So in my book the basics for some form of success are set. The enthusiast status will slowly fade and vr system will slowly seep into the mainstrean like game consoles did. For certain enthusiats vr is a revelation. If you play racing or flight sims, do yourselv a favor and get a good vr set. This "vr will never make it to mass adoption argument" because communism sucks is just weird by the way. Is there some form of goodwins law for marx?

bni 7y ago

Games did its fair share of making the PC mainstream. Many got a PC to be able to play Doom. Games today is mainstream. Everyone young plays games now, might not talk about it but everyone does. So I def think games can be a driver to make VR mainstream. Social media meets games in the other end. All it takes is VR to be both more immersive and more comfortable than playing games on flat screen.

istorical 7y ago

It always surprises me when people completely overlook the evolution TV and film will likely undergo to become stereo 180 (for directed attention experiences) or stereo 360 (for passive self-directed experiences) when HMDs become cheap enough that they become ubiquitous. Sure you can watch video without color, sure you can watch film without sound, but why would you? By the same sense, for most cases watching video/film/TV with 3D depth is going to be superior to watching without. The argument that 3D TV never took off is less convincing when you consider that content creation for 3D TV was way more costly than it is for stereo 180/360 VR today, and that 3D TV died before the chicken and egg problem was ever solved. (Even if the HMD market contracts 90% it should still be profitable for small Chinese manufacturers to produce crap-tier HMD for porn). Today a good stereo 180 VR experience can be filmed with an Insta360 Evo or a Vuze XR (my preferred option due to out of the box color grading) which can be had for about $400 USD. ZCam K1 pro and K2 pro can create incredible content (as seen in top pornography) for $3-4k entry price. But it's really hard to convince anyone who hasn't tried it, especially someone who hasn't seen something filed with one of the aforementioned cameras. Especially people who haven't tried VR or well-filmed VR but have tried 3D movies and assumes they are the same the same experience.

renox 7y ago
>By the same sense, for most cases watching video/film/TV with 3D depth is going to be superior to watching without. Uh? Except that to get 3D you have to wear glasses, which is a huge drawback. As for VR, I've tried it 3 times with demos: the first time I got so seasick that I had to stop and was feeling still quite bad 15 minutes after, the second time I nearly fell and the third time I bounced into a wall..
- jeegsy 7y ago
  Sounds like your immersion level was increasing with every try
  - renox 6y ago
    It wasn't the same demo: the first time, it was a roller coaster (a really dumb demo!!), the second time walking on a roof and falling (interesting that it created some fear of heights, but I didn't expect to nearly fall for real during the virtual fall due to the change of POV)

wildermuthn 7y ago

VR hardware is good enough to go mainstream, but as pointed out, what is missing is a mainstream utility: a killer app like the spreadsheet or word processor. The 10x advantage that VR hardware provides over a 2D display isn’t 3D space. It is social presence. VR’s killer app exists within the domain of collaboration, communication, and organization.