• 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.

I worry for Sony's commitment to PSVR2.

I’m more convinced of their commitment this time around.

In their first party we know that they have:
- FireSprite
- Insomniac
- Team Asobi
- London Studio

who make good VR games.

And they’ve continued to built on existing relationships with other game makers (AAA and indies) to provide a wider range of content.

I do think they need a couple more studios to help flesh out things, but then again we don’t know what their plans are for the metaverse.
I'm not. They closed manchester studio one of their vr studios not that long ago. Why when PSVR2 is releasing soon. Makes no sense. Unless they are treating vr2 as vita where they hope third parties will shoulder the burning of keeping it afloat.
 

ZoolNL

Member
It’s an expensive add-on for an expensive device. I’ve never seen an add-on being an enormous success. As third party I love the tech, but I would not want to support the device. Too risky.
 

Hendrick's

If only my penis was as big as my GamerScore!
VR has take a bit across the board. Sony knows this thing isn't going to fly off the shelves.
 

midnightAI

Member
I'm not. They closed manchester studio one of their vr studios not that long ago. Why when PSVR2 is releasing soon. Makes no sense. Unless they are treating vr2 as vita where they hope third parties will shoulder the burning of keeping it afloat.
They opened the studio, were working on a VR game, VR game obviously wasnt working out well, they closed the studio. They were hardly instrumental in Sony's VR strategy.

Now if they closed down Team Asobi or Firesprite (who they have only just purchased) then you'd have a point, but closing down a single VR studio who hadnt even launched a single game and it means Sony are putting PSVR 2 out to die before it even launches? sounds like reaching to me (or trolling, one of the two)
 
Last edited:

Synless

Member
It's dead in the water in my opinion. So many people were burned with psvr1 that they won't be coming back.
Who are these so many people you are speaking for? Sounds like a blanket statement with no data to back it up.

I can do it too. Of the people I know who owned PSVR, they are excited for the second one because of the actual controls designed for it and massively improved resolution and graphics. They will be buying one.

Guess most people weren’t burned after all.
 
They opened the studio, were working on a VR game, VR game obviously wasnt working out well, they closed the studio. They were hardly instrumental in Sony's VR strategy.

Now if they closed down Team Asobi or Firesprite (who they have only just purchased) then you'd have a point, but closing down a single VR studio who hadnt even launched a single game and it means Sony are putting PSVR 2 out to die before it even launches? sounds like reaching to me (or trolling, one of the two)
They aren't releasing it to die. Sony is expecting third parties to shoulder the burden. That's a fact. They need to put one of their premier developers like naughty dog or insomniac to develop a AAA exclusive in the same vein as Half life Alyx for Valve. Will it cost tens of millions. Yes. But they probably already know any huge amounts of money spent on making games for psvr will be a lost they are not willing to take. And thus another reason why psvr2 will fail.
 

th4tguy

Member
I'm not so worried they leave the accessory without games. I'm worried they split their resources too much and struggle to get enough games out for either PS5 or VR system.
 
Who are these so many people you are speaking for? Sounds like a blanket statement with no data to back it up.

I can do it too. Of the people I know who owned PSVR, they are excited for the second one because of the actual controls designed for it and massively improved resolution and graphics. They will be buying one.

Guess most people weren’t burned after all.
There are no empirical data just anedoctal evidence for reading forums and feeling the pulse of hype surrounding psvr. People are mad that psvr2 is still tethered or they are mad that psvr1 games are not backwards compatible. Or they are mad about the price. Do you not see this at all either?
 

yurinka

Member
I'm not. They closed manchester studio one of their vr studios not that long ago. Why when PSVR2 is releasing soon. Makes no sense. Unless they are treating vr2 as vita where they hope third parties will shoulder the burning of keeping it afloat.
Bullshit. They were happy with the PSVR1 performance, so now are betting harder on PSVR2, and will have around twice or more games than PSVR1 for at least the launch, its first month or first one or two years.

They closed Manchester Studio pretty likely because they weren't capable to ship a single small VR game in the 6 years it existed.

London Studio successfully released their VR stuff, they grew that studio and now is working on a AAA title.

Team Asobi also did their VR stuff and grew since then even absorving the rest of internal first party developers from Japan Studio, Polyphony also grew after making their VR stuff and made GT7 now with full VR support, and same with Media Molecule.

They also bought Firesprite, and got Fabrik to integrate it there, and since then also has been growing. They are working on Horizon CoM plus multiple unannounced games.

It's dead in the water in my opinion. So many people were burned with psvr1 that they won't be coming back.
Who? All the cases I know of people with a PSVR1 are hyped with PSVR2.

4000/110 = 36 pixels per degree
3664/90 = 41 pixels per degree
So likely yes, depending on some caveats about how stuff is rendered.
PSVR2 has:
  • Better resolution: 2000x2040 per eye vs 1832 x 1920 per eye
  • Better field of view: 110 degrees vs 90 degrees
  • Better display tech: OLED vs LCD
  • Rendering focused on the point of the screen you're looking at: eye tracking combined with foveated rendering vs nothing
We don't know the size of the display panels, their distance with the eyes or how exactly the eye tracking and the display interact, so we can't know how the pixel density compares with both systems. Also, we don't know if the eye tracking + foveated rendering helps to improve the related experience because it boosts the performance meaning that keeping the same framerate they could increase native resolution to show the same game with a sharper image.

But assuming that the panels have a pretty much similar size and distance with the eyes in both devices, and that they don't move following the eye position, the pixel density should be higher on PSVR2 too.
 
Last edited:

midnightAI

Member
They aren't releasing it to die. Sony is expecting third parties to shoulder the burden. That's a fact. They need to put one of their premier developers like naughty dog or insomniac to develop a AAA exclusive in the same vein as Half life Alyx for Valve. Will it cost tens of millions. Yes. But they probably already know any huge amounts of money spent on making games for psvr will be a lost they are not willing to take. And thus another reason why psvr2 will fail.

Sony are and will be making first party games though, GT7 is proof of that and Horizon is made by Firesprite and Guerrilla (are Polyphony and Guerilla not premier developers?). We dont yet know the plans of the other first party teams but you can bet Team Asobi will be working on something and I think its highly likely so will Insomniac but they have their hands full currently with Spider Man and Wolverine. You are acting like relying on third parties is a bad thing, how would any console do without third party? (heck, thats all MS have relied on the whole of 2022)

Edit: sorry, responded initially to wrong quote
 
Last edited:

midnightAI

Member
There are no empirical data just anedoctal evidence for reading forums and feeling the pulse of hype surrounding psvr. People are mad that psvr2 is still tethered or they are mad that psvr1 games are not backwards compatible. Or they are mad about the price. Do you not see this at all either?
I see a few individuals are mad but these individuals are probably not even interested in the PS5 never mind PSVR 2 (and we know why YouTubers do this, clickbait=views). Some of us defend these things explaining why things are how they are but it falls on deaf ears because they dont want to listen to this as the drama is what they want.

Tethered.... because it needs the bandwidth/low latency to provide a 4K HDR image and allow for foveated rendering.
No back compat.... the tracking is completely different, the controls are completely different (two analogue sticks as standard, no start/options buttons, no tracking of the dualsense), you'd also have to change the games anyway as most games show menu options and images for the old move controllers. Plenty of games are getting updates so I dont see it as much of an issue anyway)
Price.... it will be the cheapest high end headset out there, thats all that needs to be said (don't include the PS5 price as most buying PSVR2 will already have a PS5 and if not they will be buying both so the PS5 can be used for flat screen gaming also)

Could Sony have basically not just made a lower specced headset, wireless, backwards compatible and cheap? of course they could have but then it would basically just be a Quest 2 (or a wireless PSVR1). Instead they are releasing something that will be able to provide AAA quality games in VR (as we are seeing with Horizon, GT7 and Resident Evil 8)
 
I see a few individuals are mad but these individuals are probably not even interested in the PS5 never mind PSVR 2 (and we know why YouTubers do this, clickbait=views). Some of us defend these things explaining why things are how they are but it falls on deaf ears because they dont want to listen to this as the drama is what they want.

Tethered.... because it needs the bandwidth/low latency to provide a 4K HDR image and allow for foveated rendering.
No back compat.... the tracking is completely different, the controls are completely different (two analogue sticks as standard, no start/options buttons, no tracking of the dualsense), you'd also have to change the games anyway as most games show menu options and images for the old move controllers. Plenty of games are getting updates so I dont see it as much of an issue anyway)
Price.... it will be the cheapest high end headset out there, thats all that needs to be said (don't include the PS5 price as most buying PSVR2 will already have a PS5 and if not they will be buying both so the PS5 can be used for flat screen gaming also)

Could Sony have basically not just made a lower specced headset, wireless, backwards compatible and cheap? of course they could have but then it would basically just be a Quest 2 (or a wireless PSVR1). Instead they are releasing something that will be able to provide AAA quality games in VR (as we are seeing with Horizon, GT7 and Resident Evil 8)
Tether: Why not provide the option and go will dual use like meta quest 2. Ok, so sony didn't want to add to the cost..Its already 600 dollars (sorry 549 plus tax)
Backwards compatible: Trust me. They could fix the issues if they wanted to but like the first its all about money. They not gonna offer this good will to those that purchased psvr games. Basically saying. Tough. Hope your ps4 doesn't break.
Price 600 dollars is high end for sure (sorry 549 plus tax) Not to mention the cost of ps5. That's leaves out a huge mainstream of the populace. But pcs are also a cost you might say. Pcs have multiple uses besides gaming though.
 

Synless

Member
There are no empirical data just anedoctal evidence for reading forums and feeling the pulse of hype surrounding psvr. People are mad that psvr2 is still tethered or they are mad that psvr1 games are not backwards compatible. Or they are mad about the price. Do you not see this at all either?
People were pissed ps4 and ps5 didn’t support legacy games yet both consoles sold insane numbers. it’s tethered? very few will give a shit outside of being a minor annoyance.

The only thing going against this platform this go round is that it’s expensive.
 
People were pissed ps4 and ps5 didn’t support legacy games yet both consoles sold insane numbers. it’s tethered? very few will give a shit outside of being a minor annoyance.

The only thing going against this platform this go round is that it’s expensive.
So you believe PSVR2 will sell insane numbers? Curious what do you predict sales will be in the first year? How about lifetime?
 

Romulus

Member
I see a few individuals are mad but these individuals are probably not even interested in the PS5 never mind PSVR 2 (and we know why YouTubers do this, clickbait=views). Some of us defend these things explaining why things are how they are but it falls on deaf ears because they dont want to listen to this as the drama is what they want.

Tethered.... because it needs the bandwidth/low latency to provide a 4K HDR image and allow for foveated rendering.
No back compat.... the tracking is completely different, the controls are completely different (two analogue sticks as standard, no start/options buttons, no tracking of the dualsense), you'd also have to change the games anyway as most games show menu options and images for the old move controllers. Plenty of games are getting updates so I dont see it as much of an issue anyway)
Price.... it will be the cheapest high end headset out there, thats all that needs to be said (don't include the PS5 price as most buying PSVR2 will already have a PS5 and if not they will be buying both so the PS5 can be used for flat screen gaming also)

Could Sony have basically not just made a lower specced headset, wireless, backwards compatible and cheap? of course they could have but then it would basically just be a Quest 2 (or a wireless PSVR1). Instead they are releasing something that will be able to provide AAA quality games in VR (as we are seeing with Horizon, GT7 and Resident Evil 8)

I don't think it matters why it's not BC, it's just not. I have around 40 psvr1 games that are lost. In terms of it being like Quest 2 and not able to play AAA games just because its wireless, that's not exactly right. I can tether my quest 2 for perfect latency or play half life alyx wirelessly for latency I don't notice. Options. And play Half life 2 in vr, or turn around and play half life 1 in VR natively without a PC.
 
Last edited:

Synless

Member
So you believe PSVR2 will sell insane numbers? Curious what do you predict sales will be in the first year? How about lifetime?
That’s not what I said. I’m saying the same people bitched about that on ps4/5 and will bitch about it here. They are not the people to make or break the platforms sales, just noise.

I think it will do better in lifetime sales than the first.
 

midnightAI

Member
Tether: Why not provide the option and go will dual use like meta quest 2. Ok, so sony didn't want to add to the cost..Its already 600 dollars (sorry 549 plus tax)
Backwards compatible: Trust me. They could fix the issues if they wanted to but like the first its all about money. They not gonna offer this good will to those that purchased psvr games. Basically saying. Tough. Hope your ps4 doesn't break.
Price 600 dollars is high end for sure (sorry 549 plus tax) Not to mention the cost of ps5. That's leaves out a huge mainstream of the populace. But pcs are also a cost you might say. Pcs have multiple uses besides gaming though.
You just proved my point, well done
 

midnightAI

Member
I don't think it matters why it's not BC, it's just not. I have around 40 psvr1 games that are lost. In terms of it being like Quest 2 and not able to play AAA games just because its wireless, that's not exactly right. I can tether my quest 2 for perfect latency or play half life alyx wirelessly for latency I don't notice. Options. And play Half life 2 in vr, or turn around and play half life 1 in VR natively without a PC.
And does Quest 2 have 4K HDR? foveated rendering? no, so comparing the latency to Quest 2 is irrelevant

I get it about back compat, I really do, I have loads of PSVR 1 games also, but there is a reason for not having it whether you like it or not... would you rather the tracking still use the horrible move controllers just so you can play some old PSVR 1 games? There are already several games announced to be getting updates (most are free), ts up to the developers to offer updates to their games.
 

midnightAI

Member
No, I dont see them as issues if you want a high end VR experience.

How many first year? at a guess (and it is a guess), 3 mill first year? will it beat PSVR 1 in my opinion? definitely.

You say PSVR 1 soured them, but people have social media, they can see that PSVR 2 is getting a fair bit of buzz around it, impressions have been very good, even the Occulus founder was 'blown away' by it. Most of the issues with PSVR 1 have been solved, the only real issue I had with it at the time was the horrible Move controllers not tracking well.
 

Romulus

Member
And does Quest 2 have 4K HDR? foveated rendering? no, so comparing the latency to Quest 2 is irrelevant

I get it about back compat, I really do, I have loads of PSVR 1 games also, but there is a reason for not having it whether you like it or not... would you rather the tracking still use the horrible move controllers just so you can play some old PSVR 1 games? There are already several games announced to be getting updates (most are free), ts up to the developers to offer updates to their games.


I'm cool with the updates for psvr2, could mean we get improvements in some cases that otherwise might not have ever happened. For example the hit-and-run control scheme was smartly done considering the limitations but it would be so much better with psvr2 style controllers.

I'm not sure about the resolution of AAA quest 2 games when I played them on PCVR. Its older but I don't see why stepping up the resolution up to psvr2 specs would matter much. It just works so damn well there seems to be alot of headroom. Eye tracking, I know alot of the psvr2 games aren't even using it those far, switchback is an example that is. But even then quest would probably have the option to take a slight resolution hit for wireless if it did matter.
 
Last edited:

RoboFu

One of the green rats
uIuICoM.jpg
Now do a ps5 vr launch games vs what’s been available for quest 2 for years. 😂
 

midnightAI

Member
I'm cool with the updates for psvr2, could mean we get improvements in some cases that otherwise might not have ever happened. For example the hit-and-run control scheme was smartly done considering the limitations but it would be so much better with psvr2 style controllers.

I'm not sure about the resolution of AAA quest 2 games when I played them on PCVR. Its older but I don't see why stepping up the resolution up to psvr2 specs would matter much. It just works so damn well there seems to be alot of headroom. Eye tracking, I know alot of the psvr2 games aren't even using it those far, switchback is an example that is. But even then quest would probably have the option to take a slight resolution hit for wireless if it did matter.
I'm not sure you understand the foveated rendering/eye tracking? Horizon uses it, GT7 will use it, Resident Evil 8 will use it. It's not just about tracking whether the eyes close or not like Switchback does for its blink/weeping angels type gameplay, its about tracking where the eyes are looking. What the eyes look at is rendered in full resolution, then from that point outwards you render different resolutions going lower as you go outwards (sometimes this is two addition resolutions, one in the peripheral that is rather lower resolution then one between the two in about a mid level resolution just so there is no large step between the two. Then there are various different solutions to blue between them, again, just to soften things out so there are no large steps.

To do this you have to track the eyes and get what they are looking at, then, you have to send that to the PS5, the PS5 then renders the output with the above foveated view and then sends that back to the headset, So basically, its a two tway process not just one way like other headsets without eye tracking. Latency for foveated rendering is a killer, the round trip latency I believe should ideally be sub 40ms with around 60-70ms as high as you would want ever go (there is a paper by NVidia about foveated rendering latency)
Edit: found it: https://research.nvidia.com/sites/default/files/pubs/2017-09_Latency-Requirements-for/a25-albert.pdf

Moss is one game that is getting a lot of updates/upgrades, (one of which is using the eye tracking for object selection), this is a paid upgrade though but I can understand that to a degree with the updates they are making, still would have been nice if free though.
 
Last edited:

Romulus

Member
I'm not sure you understand the foveated rendering/eye tracking? Horizon uses it, GT7 will use it, Resident Evil 8 will use it. It's not just about tracking whether the eyes close or not like Switchback does for its blink/weeping angels type gameplay, its about tracking where the eyes are looking. What the eyes look at is rendered in full resolution, then from that point outwards you render different resolutions going lower as you go outwards (sometimes this is two addition resolutions, one in the peripheral that is rather lower resolution then one between the two in about a mid level resolution just so there is no large step between the two. Then there are various different solutions to blue between them, again, just to soften things out so there are no large steps.

To do this you have to track the eyes and get what they are looking at, then, you have to send that to the PS5, the PS5 then renders the output with the above foveated view and then sends that back to the headset, So basically, its a two tway process not just one way like other headsets without eye tracking. Latency for foveated rendering is a killer, the round trip latency I believe should ideally be sub 40ms with around 60-70ms as high as you would want ever go (there is a paper by NVidia about foveated rendering latency)
Edit: found it: https://research.nvidia.com/sites/default/files/pubs/2017-09_Latency-Requirements-for/a25-albert.pdf

Moss is one game that is getting a lot of updates/upgrades, (one of which is using the eye tracking for object selection), this is a paid upgrade though but I can understand that to a degree with the updates they are making, still would have been nice if free though.

That's an ancient article in terms of VR technology. PSVR1 was barely out 8 months at that time. If you were to tell me I could stream something like Half Life Alyx with no perceivable latency in 2017 I would tell you that's complete bullshit. It's not though. But everyone at that time was saying wireless VR streaming high-fidelity games was impossible and would make everyone vomit. Having said that, their opinion could have changed a great deal on rendering-saving techniques and streaming.
Even if we were cemented in 2017 tech, I still think it's an easy solution. You could just drop the rendering tech for wireless streaming. Any capable PC will still run it well enough to get used to it within a few minutes, but if you really wanted the rendering-saving tech, tethered is an option.
I'm really going to be looking at how much this tech actually saves. I've played the RE8 mod and it was great. But that's just an unofficial mod by a couple of guys. I would hope capcom could do better and it seems they have, but in the end, I don't expect it to be that big of a jump.
 
Last edited:

midnightAI

Member
That's an ancient article in terms of VR technology. PSVR1 was barely out 8 months at that time. If you were to tell me I could stream something like Half Life Alyx with no perceivable latency in 2017 I would tell you that's complete bullshit. It's not though. But everyone at that time was saying wireless VR streaming high-fidelity games was impossible and would make everyone vomit. Having said that, their opinion could have changed a great deal on rendering-saving techniques and streaming.
Even if we were cemented in 2017 tech, I still think it's an easy solution. You could just drop the rendering tech for wireless streaming. Any capable PC will still run it well enough to get used to it within a few minutes, but if you really wanted the rendering-saving tech, tethered is an option.
I'm really going to be looking at how much this tech actually saves. I've played the RE8 mod and it was great. But that's just an official mod by a couple of guys. I would hope capcom could do better and it seems they have.
What has the age of the article got to do with anything? It's about latency with foveated rendering, its to do with how our eyes work, biology, the time it takes for our eyes to move and focus from one point to the next (saccade). It's this saccade that governs the maximum latency required for foveated rendering to work, it isnt bullshit as you say.
They want to make AAA games in VR, they need a way of doing that and foveated rendering is the technique that allows it. Could they drop dynamic foveated rendering? sure, but then they wouldnt have the visuals that are allowed by foveated across the display. It was a choice, foveated or wireless and in my opinion they chose the correct path, better visuals over wireless any day. But still, this is a different argument to your original post which just alluded to the only game taking advantage of eye tracking was Switchback, but thats simply not true as eye tracking is required for foveated rendering.
People ask why PSVR 2 would sell (bringing up Quest 2 all the time), the answer is the software (mostly first party with some exceptions like Resi Evil) in combination with what the headset can do.

If you want wireless, stick to Quest 2 or wait for Quest 3 (or any other wireless headset), no point complaining about PSVR 2 not having it as there are other options and there are reasons Sony have done it the way they have, simple really.
 
No, I dont see them as issues if you want a high end VR experience.

How many first year? at a guess (and it is a guess), 3 mill first year? will it beat PSVR 1 in my opinion? definitely.

You say PSVR 1 soured them, but people have social media, they can see that PSVR 2 is getting a fair bit of buzz around it, impressions have been very good, even the Occulus founder was 'blown away' by it. Most of the issues with PSVR 1 have been solved, the only real issue I had with it at the time was the horrible Move controllers not tracking well.

I think for most people that played PSVR1, it wasn't that they were soured - quite the opposite. PSVR1 proved it could be amazing. It's just PSVR1's shortcomings meant that it was too much of a hassle to set up, too janky for the control interfaces, and not clear enough to really pull out of the closet very often.

PSVR2 solves all of those issues and is a genuine platform, so as someone that was excited about the prospect of what VR could achieve, PSVR2 is essentially the culmination of numerous improvements that will lead to a dramatic re-invigoration of VR enthusiasm
 

Romulus

Member
What has the age of the article got to do with anything? It's about latency with foveated rendering, its to do with how our eyes work, biology, the time it takes for our eyes to move and focus from one point to the next (saccade). It's this saccade that governs the maximum latency required for foveated rendering to work, it isnt bullshit as you say.
They want to make AAA games in VR, they need a way of doing that and foveated rendering is the technique that allows it. Could they drop dynamic foveated rendering? sure, but then they wouldnt have the visuals that are allowed by foveated across the display. It was a choice, foveated or wireless and in my opinion they chose the correct path, better visuals over wireless any day. But still, this is a different argument to your original post which just alluded to the only game taking advantage of eye tracking was Switchback, but thats simply not true as eye tracking is required for foveated rendering.
People ask why PSVR 2 would sell (bringing up Quest 2 all the time), the answer is the software (mostly first party with some exceptions like Resi Evil) in combination with what the headset can do.

If you want wireless, stick to Quest 2 or wait for Quest 3 (or any other wireless headset), no point complaining about PSVR 2 not having it as there are other options and there are reasons Sony have done it the way they have, simple really.

What does age have to do with it? You posted an article from a time when wirelessly streaming VR games was thought to cause vomit-inducing latency. Meanwhile, millions of people are doing it now, including myself. So, it's really difficult to take that article seriously considering so much has changed on the VR streaming front. Not saying it would be possible or not, just saying I don't consider the article valid anymore. I never said the article was bullshit, you misread my post. I think the path sony went was fine, they only have 10TF to work with when we're moving towards 90+ in the PS5's lifetime. So you're looking at 10x the horsepower. And I'm not complaining at all, I'm just having a discussion. I think when people like something, negative comments might come off as complaining. It's not in my case. I just mentioned I can play AAA games wirelessly or tethered with Quest 2. That was it.
 

Gambit2483

Member
Sony's pricing strategy killed Vita. It will be the same with PSVR2.

They will get the Hardcore/early adopters in the first 6-12 months but after that they need to drop the price or it WILL die an early death.
 
I think the path sony went was fine, they only have 10TF to work with when we're moving towards 90+ in the PS5's lifetime. So you're looking at 10x the horsepower. And I'm not complaining at all, I'm just having a discussion. I think when people like something, negative comments might come off as complaining. It's not in my case. I just mentioned I can play AAA games wirelessly or tethered with Quest 2. That was it.

Nobody is going to be developing exclusive VR games with 90 TFlops in mind. And the TFlops you are using are not comparable, since they are literally getting inflated with every architecture improvement in many cases (NVIDIA).

Saying they "only have 10 TF" to work with is hilarious when Quest 3 standalone will not even have 10% of that. And Sony's foveated rendering solution allows for significant resource optiomization. 10 TF for VR development as a baseline is HUGE

Furthermore, it's not like Sony can't release a PS5 Pro or PS6 and have it compatible with PSVR2.
 
Last edited:

midnightAI

Member
What does age have to do with it? You posted an article from a time when wirelessly streaming VR games was thought to cause vomit-inducing latency. Meanwhile, millions of people are doing it now, including myself. So, it's really difficult to take that article seriously considering so much has changed on the VR streaming front. Not saying it would be possible or not, just saying I don't consider the article valid anymore. I never said the article was bullshit, you misread my post. I think the path sony went was fine, they only have 10TF to work with when we're moving towards 90+ in the PS5's lifetime. So you're looking at 10x the horsepower. And I'm not complaining at all, I'm just having a discussion. I think when people like something, negative comments might come off as complaining. It's not in my case. I just mentioned I can play AAA games wirelessly or tethered with Quest 2. That was it.
The paper isnt talking about whether its wireless or wired, its about the latency required to effectively use foveated rendering.
Did you read the paper (or my post) at all? because I am confused by your responses, this isnt about one way latency (PC to headset), we know that part is fine for the most part. This is about a round trip, headset to PS5 then back to the headset, it needs to do that as the eyes need to be tracked to know what to render. If that is more than 80ms then there is an issue because that is larger than the saccade (which is biology) which means that what you see could be blurred for a short amount of time until the frames are back to what you are looking at.
That is why the PS5 is wired to the PSVR2, it guarantees a stable connection that is fast enough to make foveated rendering work, wireless cannot guarantee this.
 
Last edited:

midnightAI

Member
Nobody is going to be developing exclusive VR games with 90 TFlops in mind. And the TFlops you are using are not comparable, since they are literally getting inflated with every architecture improvement in many cases (NVIDIA).

Saying they "only have 10 TF" to work with is hilarious when Quest 3 standalone will not even have 10% of that. And Sony's foveated rendering solution allows for significant resource optiomization. 10 TF for VR development as a baseline is HUGE

Furthermore, it's not like Sony can't release a PS5 Pro or PS6 and have it compatible with PSVR2.
Which also then goes full circle....
PSVR 2 is too expensive... is it though? because to match it you need a PC and VR headset that cost way more than PSVR2 (with PS5 which people like to bring up in cost complaints)
 
I'm not so worried they leave the accessory without games. I'm worried they split their resources too much and struggle to get enough games out for either PS5 or VR system.

I only see this being problem if PSVR2 has strong early success beyond expected, because then they will want to push heavier investment quickly to try and keep such momentum going. Same if it was a good but not amazing launch and the bigger games end up being the popular causes, or Horizon.

But if it's modest or less than I expect them to spread out investment.
 

Romulus

Member
The paper isnt talking about whether its wireless or wired, its about the latency required to effectively use foveated rendering.
Did you read the paper (or my post) at all? because I am confused by your responses, this isnt about one way latency (PC to headset), we know that part is fine for the most part. This is about a round trip, headset to PS5 then back to the headset, it needs to do that as the eyes need to be tracked to know what to render. If that is more than 80ms then there is an issue because that is larger than the saccade (which is biology) which means that what you see could be blurred for a short amount of time until the frames are back to what you are looking at.
That is why the PS5 is wired to the PSVR2, it guarantees a stable connection that is fast enough to make foveated rendering work, wireless cannot guarantee this.

I read it, and my point still stands. The conversation is about wired or wireless. You're making the point that it's impossible based on that old article because of latency. Technology is evolving so quickly in terms of latency, wifi I'm not going to take any article from 2017 with a grain of salt. Someone saying it's impossible in 2017 means little to me.
 

midnightAI

Member
I read it, and my point still stands. The conversation is about wired or wireless. You're making the point that it's impossible based on that old article because of latency. Technology is evolving so quickly in terms of latency, wifi I'm not going to take any article from 2017 with a grain of salt. Someone saying it's impossible in 2017 means little to me.
I am actually, going to do something here which I rarely do.... I give up, you obviously dont understand why low, knowable and controllable latency is needed for foveated rendering, either that or you are being deliberately obtuse/ignorant. The age of the article is irrelevant unless our understanding of biology has changed (which it hasnt)
 
Last edited:

Romulus

Member
Nobody is going to be developing exclusive VR games with 90 TFlops in mind. And the TFlops you are using are not comparable, since they are literally getting inflated with every architecture improvement in many cases (NVIDIA).

Saying they "only have 10 TF" to work with is hilarious when Quest 3 standalone will not even have 10% of that. And Sony's foveated rendering solution allows for significant resource optiomization. 10 TF for VR development as a baseline is HUGE

Furthermore, it's not like Sony can't release a PS5 Pro or PS6 and have it compatible with PSVR2.


I was actually wrong, seems like GPUs are already above my numbers, so we'll probably see over 130+, so even if the GPU TFLops are inflated, I was downplaying. Developing high end VR games with 100TF in mind wasn't my point, even if its 20 or 30 and being able to crank the resolution is huge or even RT. We're getting to a point where we could easily see that in VR. Where did you get your quest 3 numbers? Sure its just standalone but I didn't know it was sub 1TF.
 

Romulus

Member
I am actually, going to do something here which I rarely do.... I give up, you obviously dont understand why low, knowable and controllable latency is needed for foveated rendering, either that or you are being deliberately obtuse/ignorant. The age of the article is irrelevant unless our understanding of biology has changed (which it hasnt)

Right. Latency and wifi have improved drastically. it's as simple as that. This was the point I was making in the beginning. You making it about "biology" is a ridiculous angle. You could easily say the same about streaming a high-fidelity game in VR back in 2017. "Doesn't work because biology. Latency is too low for a stable experience." Again, I'm not making 1to1 comparison to your point here, just showing you I can say that about anything that did not work in the past.
 
I was actually wrong, seems like GPUs are already above my numbers, so we'll probably see over 130+, so even if the GPU TFLops are inflated, I was downplaying. Developing high end VR games with 100TF in mind wasn't my point, even if its 20 or 30 and being able to crank the resolution is huge or even RT. We're getting to a point where we could easily see that in VR. Where did you get your quest 3 numbers? Sure its just standalone but I didn't know it was sub 1TF.

yes, we will likely see higher TFlops the more this TFLop inflation game is being played. That's not really the relevant metric when judging power increases. A 130 TF system won't be 13X more powerful than a PS5. 10 TF is plenty powerful, RE8 looks virtually identical to the flat screen version.

Quest is a mobile chip, of course it's sub 1TF. Lol. There's literally no way it would come in under the power footprint/battery issues otherwise. Best case is Quest 3 being equivalent to something like Switch.
 
Last edited:

midnightAI

Member
Right. Latency and wifi have improved drastically. it's as simple as that. This was the point I was making in the beginning. You making it about "biology" is a ridiculous angle. You could easily say the same about streaming a high-fidelity game in VR back in 2017. "Doesn't work because biology. Latency is too low for a stable experience." Again, I'm not making 1to1 comparison to your point here, just showing you I can say that about anything that did not work in the past.
You definitely didnt read the article or understand how any of this works if you thing it is nothing to do with biology (eye saccade, in particular saccade blindness) which is fine.

Anyway, pointless arguing.
 

Romulus

Member
You definitely didnt read the article or understand how any of this works if you thing it is nothing to do with biology (eye saccade, in particular saccade blindness) which is fine.

Anyway, pointless arguing.

lol never said it has nothing to do with biology, there you go putting words in my mouth again. You're using it as a crutch to sound smart. Just like you said I called the article "bullshit." That never happened.

You're intentionally projecting with your obtuse comment I see now. We're constantly fighting against biology on all these challenges, that's not the point. Latency can improve with time, and it has since that article was written and it will continue to do so. Biology!
 

Romulus

Member
Quest is a mobile chip, of course it's sub 1TF. Lol. There's literally no way it would come in under the power footprint/battery issues otherwise. Best case is Quest 3 being equivalent to something like Switch.

Nope.

Qualcomm’s GPU line used in its chipsets is branded Adreno. The original Oculus Quest used the Adreno 540, with 0.6 teraflops performance. Quest 2 shipped with a three generations newer chipset featuring the Adreno 650, delivering twice the performance at 1.2 teraflops.

I bet this makes more sense to you now why Quest 2 can run PS4 games at a higher resolutions.

BTW, Quest 3 is said to be well over 2 teraflops. So that smokes the switch. Even the old Quest 2 has more juice. Just imagining the switch trying to run Iron Man VR(an exclusive PSVR game) at 72fps at over 1200p is all kinds of hilarious.

https://uploadvr.com/quest-3-gpu-twice-powerful/
 
Last edited:

StreetsofBeige

Gold Member
For any of you trying to guesstimate sales of PSVR2, put it this way.

PSVR1 sold 5M units as of 2020 according to Google articles. I dont know how many last bits of sales happened after that point. Maybe it hit 6M. I dont know.

That was at a max price of $400 (it went on sale too) across 120M PS4s over about 5 years.
 

midnightAI

Member
lol never said it has nothing to do with biology, there you go putting words in my mouth again. You're using it as a crutch to sound smart. Just like you said I called the article "bullshit." That never happened.

You're intentionally projecting with your obtuse comment I see now. We're constantly fighting against biology on all these challenges, that's not the point. Latency can improve with time, and it has since that article was written and it will continue to do so. Biology!
Sorry, misread your comment you was saying bullshit about not being able to stream HL Alyx, thats my bad

But the latency for foveated rendering is about biology, the article does not care about the method we use (the article doesnt even mention wired or wireless at all), the article is about perceived time between one eye movement and another to determine the latency allowed (including time to refocus (saccade blindness)). This is then a round trip, eye to ps5 to eye again in less than 80ms (ideally less than 40), that 'currently' cannot be done wirelessly (at least not for everyone, consistently, like I said, known latency), and that is why the PS5 is wired, its that simple. PC VR (for the main part) does not use dynamic foveated rendering so they only need one way latency so it isnt the same thing.
 

midnightAI

Member
For any of you trying to guesstimate sales of PSVR2, put it this way.

PSVR1 sold 5M units as of 2020 according to Google articles. I dont know how many last bits of sales happened after that point. Maybe it hit 6M. I dont know.

That was at a max price of $400 (it went on sale too) across 120M PS4s over about 5 years.
You didnt add on the price of two move controllers + camera

Also VR back then wasnt anywhere near as popular as it is now (PSVR was one of the first highish end affordable VR headsets)

I'm still guessing around 3million units sold in the first year
 
Last edited:

Fredrik

Member
Whatever anyone says, when one looks at how fast they dropped the Vita, I believe your concerns are valid.

tl;dr:
Think About It GIF by Identity
Goes for any secondary platform, the only one that has managed to skip the B team curse is Steam Deck since it’s the same game no matter if it’s played handheld or stationary.
 

Romulus

Member
Sorry, misread your comment you was saying bullshit about not being able to stream HL Alyx, thats my bad

But the latency for foveated rendering is about biology, the article does not care about the method we use (the article doesnt even mention wired or wireless at all), the article is about perceived time between one eye movement and another to determine the latency allowed (including time to refocus (saccade blindness)). This is then a round trip, eye to ps5 to eye again in less than 80ms (ideally less than 40), that 'currently' cannot be done wirelessly (at least not for everyone, consistently, like I said, known latency), and that is why the PS5 is wired, its that simple. PC VR (for the main part) does not use dynamic foveated rendering so they only need one way latency so it isnt the same thing.

And I definitely understand why that was the best choice for sony presently.
 

StreetsofBeige

Gold Member
You didnt add on the price of two move controllers + camera

Also VR back then wasnt anywhere near as popular as it is now (PSVR was one of the first highish end affordable VR headsets)

I'm still guessing around 3million units sold in the first year
Good point, but I doubt it'll hit 3M in year one. After the first year it's out PS5 user base might reach around 40M? I think it's at 30M now. 3M on 40M PS5s is a conversion rate of 7.5%. So 1 out of every 13 people will buy PSVR2?

PSVR1 sold about 1M units per year.

 
I'm not ready to jump on board this gen, but I hope to see VR reach for the stars one day with high end aaa experiences. I appreciate all you early adopters and supporters who will pave the way to eventual greatness.
 

midnightAI

Member
Good point, but I doubt it'll hit 3M in year one. After the first year it's out PS5 user base might reach around 40M? I think it's at 30M now. 3M on 40M PS5s is a conversion rate of 7.5%. So 1 out of every 13 people will buy PSVR2?

PSVR1 sold about 1M units per year.

Well we'll have to wait and see wont we :messenger_winking:
 
Top Bottom