Support NeoGAF

[dark10x]

Just checked this out and about cried. I don't have a Rift. The nicest PC I have is a Surface Pro. I've been wanting to get into PC gaming for a while now and build a nice rig but we are trying to build a house in 4 years so I only budget 1 Xbox game a month. I'm constantly trying to stay out of these threads to avoid temptation but the Rift is something I'm really interested in so I check up on it from time to time.

Man, you have no idea how much I miss going to those retro arcades. Every arcade from my childhood has closed down and recently I've been having some serious feels for that era.

Seeing that game and the thought of being able to relive that to some degree in a Rift hit me really hard. It's almost to much to handle.

On another note, how many of you can magic eyes this?

If so, how accurate is that to what you actually see in a Rift?

You cannot magic eye this stuff.

If you haven't used it yourself, it's hard to understand, but it really creates the sense of being in a place with the correct size and scale that you expect in real life. Spending time in that virtual arcade really feels like you're there more than most demos, I've found. I think focusing on a screen within a screen is what does it. Also, it's a dark environment and features some incredible materials work and lighting.

Looking at it on a normal monitor doesn't do it justice at all.

When you try it I think you'll find that it feels as if you were actually there in some ways.

Most other demos I've tried really don't come close, to be honest. It's far and away my favorite thing for the rift.

 

[Exploding8]

Just checked this out and about cried. I don't have a Rift. The nicest PC I have is a Surface Pro. I've been wanting to get into PC gaming for a while now and build a nice rig but we are trying to build a house in 4 years so I only budget 1 Xbox game a month. I'm constantly trying to stay out of these threads to avoid temptation but the Rift is something I'm really interested in so I check up on it from time to time.

Man, you have no idea how much I miss going to those retro arcades. Every arcade from my childhood has closed down and recently I've been having some serious feels for that era.

Seeing that game and the thought of being able to relive that to some degree in a Rift hit me really hard. It's almost to much to handle.

On another note, how many of you can magic eyes this?

If so, how accurate is that to what you actually see in a Rift?

I can do the whole cross eyed thing for those and it's absolutely nothing like having the Rift on. And I say that as someone who hasn't been that blown away just yet by my DK2.

I find it hilarious when people say VR is like 3d for that reason. When this shit really starts getting developed it won't even be close.

 

[Durante]

I suppose you could just render at a lower resolution but I'd take the SDE at native resolution than upscaled, blurry pixels.

No pixel on a HMD is really all that "native" though. And you'll get "blur" anyway because of diffusion optics designed to eliminate SDE I assume

 

[call_kotaku]

I see, I get your point, but I can't read a sheet of paper that I just shake back and forth right now in front of me either, so there's that.


I wonder if technologies like "Motionflow" from those recent Smart TVs would be a good fit for the Rift as a way to reduce GPU load - at least an option to simulate a smooth, non-blurred image at the cost of some artifacting. My Samsung TV does it, and the results are actually not that bad.

The problem with <60fps in VR is that when you move your head (and the HMD) your eyes are rotating in the opposite direction to track the object, which you don't really do on monitors because monitors aren't moving across your view. This exaggerates any motion blurring which makes low persistence and higher refresh really important.

 

[mrklaw]

doesn't morpheus have some kind of breakout box that is doing interpolation stuff? At least that was rumoured.

If they can interpolate to 120Hz (or even 90) that could help. bonus points if they can interpolate and adapt with motion (like timewarp) based on user movement in between those 60 native frames getting sent from the PS4.

 

[bj00rn_]

No it isn't.




Besides, people experiencing CB today are saying SDE is essentially solved by optics (although further fine detail could be derived from increasing resolution).

I actually don't agree with Palmer here. He is of course technically correct, I understand exactly what he's saying, but I don't care about the hair splitting side in this particular case, I only care about what I see. And what I see is a screen door effect created by mainly the low resolution, and the pixel fill from the pentile panel, that's a fact no matter how we argue about technical definitions. And it need to be solved (most likely already is).

doesn't morpheus have some kind of breakout box that is doing interpolation stuff? At least that was rumoured.

If they can interpolate to 120Hz (or even 90) that could help. bonus points if they can interpolate and adapt with motion (like timewarp) based on user movement in between those 60 native frames getting sent from the PS4.

If it's that easy to interpolate without drawbacks (assuming), why isn't the CV1 or earlier kits already doing it (or..?). Sounds too good to be true. Maybe Krejlooc knows something about it.

 

[Betta Lines]

If it's that easy to interpolate without drawbacks (assuming), why isn't the CV1 or earlier kits already doing it (or..?). Sounds too good to be true. Maybe Krejlooc knows something about it.

I thought interpolation requires the refresh rate to be higher than the framerate, and then it is filling in missing info to make it look like it is running at a higher framerate. So in the case of PC, if you had a panel that could do 120Hz, you might as well throw 120fps at it. Sony might go the interpolation route because the PS4 is going to be more comfortable at 60, and then there might be some small benefit of making it look like 120. But throwing a real 120fps at it is surely always going to be better?

 

[Astral/H3X]

Sounds like interpolation is the same thing that Motion flow tvs do... Aka lag. Which is utterly the enemy of VR. In order to interpolate the frames, you need at LEAST two frames of data... So if you have frame A and frame C, in order to interpolate frame B, frame C must already be rendered. You wouldn't be seeing frame B until at least after you should be seeing frame C, and that's not even considering the time it takes the device to figure out what frame B is after it has everything it needs.

 

[call_kotaku]

doesn't morpheus have some kind of breakout box that is doing interpolation stuff? At least that was rumoured.

If they can interpolate to 120Hz (or even 90) that could help. bonus points if they can interpolate and adapt with motion (like timewarp) based on user movement in between those 60 native frames getting sent from the PS4.

I think if Morpheus can do 1080p at 60+ fps, it would be good enough for what it is. There's no reason to compete with Oculus on specs; as long as its better than mobile VR like GearVR or Cardboard (which shouldn't be difficult for a dedicated gaming console) and it's priced reasonably, it should do well enough to target people who already own PS4s. No point in trying to chase specs meant for something to be used with gaming PCs, especially since the pace of hardware iteration for PC VR on both HMDs and GPUs will outpace Sony's generational cycle.

Basically, if Sony sells the equivalent of a DK2 with PSMove controllers, I think they'd be fine.

 

[Despera]

I'm currently thinking about my future VR setup once the CV1 becomes available... and other than stuff that are a given like the Oculus Rift itself, a powerful gaming PC and the Virtuix Omni (which seems more practical than the Cyberith Virtualizer at the moment), I can't seem to pinpoint a suitable motion control solution to replace the gamepad.

You have Nimble, Perception Neuron, ControlVR, PrioVR, Sixense, Razer Hydra and Leap Motion... and I'm sure I'm forgetting a couple here. There's a lot to keep track of and it feels like a new one is announced every month.

And when I think about it further it seems that having only one of these won't be enough to cover most game types, so perhaps combining two solution into one setup would be the best way to cover more ground. Like a Neuron+Sixense setup for example? Is such a thing even possible?

For me it's kinda important to have accurate motion controls for a more immersive VR experience, so I hope we get more detailed impressions of all the negatives and positives from people like Krejlooc in the months leading up to CV1.

 

[PGamer]

I'm currently thinking about my future VR setup once the CV1 becomes available... and other than stuff that are a given like the Oculus Rift itself, a powerful gaming PC and the Virtuix Omni (which seems more practical than the Cyberith Virtualizer at the moment), I can't seem to pinpoint a suitable motion control solution to replace the gamepad.

You have Nimble, Perception Neuron, ControlVR, PrioVR, Sixense, Razer Hydra and Leap Motion... and I'm sure I'm forgetting a couple here. There's a lot to keep track of and it feels like a new one is announced every month.

And when I think about it further it seems that having only one of these won't be enough to cover most game types, so perhaps combining two solution into one setup would be the best way to cover more ground. Like a Neuron+Sixense setup for example? Is such a thing even possible?

For me it's kinda important to have accurate motion controls for a more immersive VR experience, so I hope we get more detailed impressions of all the negatives and positives from people like Krejlooc in the months leading up to CV1.

FYI Oculus recently bought Nimble VR so they cancelled their Kickstarter and presumably won't be releasing their original product. The big question right now as far as input goes seems to be what is Oculus going to do for an official input solution and when will it actually release (with CV1 or a later model). Unless you need something right away I would probably hold off for a while until we get a better picture of how things shape up.

 

[Despera]

FYI Oculus recently bought Nimble VR so they cancelled their Kickstarter and presumably won't be releasing their original product. The big question right now as far as input goes seems to be what is Oculus going to do for an official input solution and when will it actually release (with CV1 or a later model). Unless you need something right away I would probably hold off for a while until we get a better picture of how things shape up.

Well I hope they officially announce whatever they're working on before the CV1 releases. Otherwise I'll properly make do with the most versatile third party solution available at that point.

 

[Big Takeover]

Well I hope they officially announce whatever they're working on before the CV1 releases. Otherwise I'll properly make do with the most versatile third party solution available at that point.

http://youtu.be/t-IdUI0qs7w?t=14m10s

 

[Blizzard]

I had to google for a bit to finally realize that SDE stands Screen Door Effect.

The screen door effect is one of the things that is most noticeable and bothersome in my opinion, especially for anything not close to you, though it is a bit better in the DK2.

 

[vermadas]

Sounds like interpolation is the same thing that Motion flow tvs do... Aka lag. Which is utterly the enemy of VR. In order to interpolate the frames, you need at LEAST two frames of data... So if you have frame A and frame C, in order to interpolate frame B, frame C must already be rendered. You wouldn't be seeing frame B until at least after you should be seeing frame C, and that's not even considering the time it takes the device to figure out what frame B is after it has everything it needs.

I suggest you learn about timewarp in case you haven't yet. Here's a good video:

http://youtu.be/WvtEXMlQQtI

Timewarp is somewhat the same concept as "Motion Flow", but you don't need the next rendered frame to interpolate. The video does a great idea of explaining why this isn't a perfect solution, and I remember that someone at Oculus, perhaps Carmack, calling it "framerate insurance". So it's not like devs can aim for locked 60 FPS and use timewarp to fill in every third frame for them, but it should make things a bit easier.

 

[Mechanical Snowman]

I had to google for a bit to finally realize that SDE stands Screen Door Effect.

The screen door effect is one of the things that is most noticeable and bothersome in my opinion, especially for anything not close to you, though it is a bit better in the DK2.

I agree. Thankfully it sounds like they've nailed it by the CB demos. The only thing I'm concerned about now is the resolution, if it's still 1080p then they've really limited the Rift's potential. Their silence on this makes me nervous, especially Nate's constant "we're not talking about the resolution because it's just not important - we don't want people to focus on non important details." It is important.

 

[Big Takeover]

I agree. Thankfully it sounds like they've nailed it by the CB demos. The only thing I'm concerned about now is the resolution, if it's still 1080p then they've really limited the Rift's potential. Their silence on this makes me nervous, especially Nate's constant "we're not talking about the resolution because it's just not important - we don't want people to focus on non important details." It is important.

The resolution of an internal dev kit isn't important is what he's saying. It's not like Crescent Bay is preproduction, or anything. What you'll be buying will be higher than 1080p. That's pretty much a given.

 

[Astral/H3X]

I suggest you learn about timewarp in case you haven't yet. Here's a good video:

http://youtu.be/WvtEXMlQQtI

Timewarp is somewhat the same concept as "Motion Flow", but you don't need the next rendered frame to interpolate. The video does a great idea of explaining why this isn't a perfect solution, and I remember that someone at Oculus, perhaps Carmack, calling it "framerate insurance". So it's not like devs can aim for locked 60 FPS and use timewarp to fill in every third frame for them, but it should make things a bit easier.

I've heard about Timewarp and how it works, and it's ENTIRELY different to interpolation. Interpolation is taking two existing frames, and figuring out the median of them. This requires you to have both frames in question, on top of processing time. Timewarp acts on a single frame, and simply "moves" it to adjust for latency. There's no interpolation of anything. It isn't comparing two frames. I guess you could say it's interpolating the head position data, but it's more double sampling and updating it. IMO Timewarp's biggest feature is as I saw on a Gear VR demonstration, where if the Note 4 didn't have a whole new frame ready, Timewarp would grab the last one it had and "move it into position" to make it less apparent. That is completely different from what Motionflow does.

EDIT: To be clear, they were asking why they couldn't use interpolation to get a "60fps" scene when it was rendering at 30fps. Interpolation would NOT help at all, since the main reason for high refresh rates is to combat latency, which that method of faking higher framerates would actually make worse.

 

[somesang]

My DK2 is half an hour away and won't be delivered til Monday. Gonna be a long weekend. Guess I'll spend some time in Unity and get my bearings. Trying not to get too hyped. Glad a release date for CV1 wasn't announced but also sad.

 

[Mechanical Snowman]

The resolution of an internal dev kit isn't important is what he's saying. It's not like Crescent Bay is preproduction, or anything. What you'll be buying will be higher than 1080p. That's pretty much a given.

I hope you're right, it's just weird behavior IMO from a company that is usually quite transparent, so it makes me think the worst.

In any case, I still think it's important right now to know what the resolution of CB is, considering all that devs have got to work under the assumption of is that "Everything in CV1 will be as good as or at least better than CB". If you are developing your content optimized for 2560x1440 @ 90hz, and the final resolution of the HMD is lower, you've been making performance optimizations and sacrificing graphics that weren't necessary, and if it's vice versa, you've been making performance optimizations that aren't sufficient.

Maybe it's not an issue. I haven't released any content yet so I can't quite call myself a developer, but I just think that why wouldn't we need to know the hardware specifications, at least as close as possible, to what the final unit's specs will be? I'd be interested in hearing an opinion from developers who might have released stuff on other platforms, or even for DK2.

 

[mrklaw]

I've heard about Timewarp and how it works, and it's ENTIRELY different to interpolation. Interpolation is taking two existing frames, and figuring out the median of them. This requires you to have both frames in question, on top of processing time. Timewarp acts on a single frame, and simply "moves" it to adjust for latency. There's no interpolation of anything. It isn't comparing two frames. I guess you could say it's interpolating the head position data, but it's more double sampling and updating it. IMO Timewarp's biggest feature is as I saw on a Gear VR demonstration, where if the Note 4 didn't have a whole new frame ready, Timewarp would grab the last one it had and "move it into position" to make it less apparent. That is completely different from what Motionflow does.

EDIT: To be clear, they were asking why they couldn't use interpolation to get a "60fps" scene when it was rendering at 30fps. Interpolation would NOT help at all, since the main reason for high refresh rates is to combat latency, which that method of faking higher framerates would actually make worse.

Perhaps I shouldn't have used interpolation as a term. If it is taking an existing frame and warping it based on user movement, while waiting for the next real frame to come down the pipe, then it could emulate a higher perceived frame rate while only needing 60fps from the source device.

 

[Seanspeed]

I agree. Thankfully it sounds like they've nailed it by the CB demos. The only thing I'm concerned about now is the resolution, if it's still 1080p then they've really limited the Rift's potential. Their silence on this makes me nervous, especially Nate's constant "we're not talking about the resolution because it's just not important - we don't want people to focus on non important details." It is important.

When they say it isn't important, they mean that the overall result is the important thing, not what the resolution is specifically.

 

[call_kotaku]

Also, keep in mind that whatever Samsung (presumably) screen they are using might be an unannounced display from their next generation of phones, so there might be NDA issues for them announcing exact specs. If I remember correctly they couldn't even say which company manufactured the OLED screen for Crystal Cove/DK2, even though there honestly was only one possibility (until people took it apart and saw it was literally the front of a Note 3).

Perhaps I shouldn't have used interpolation as a term. If it is taking an existing frame and warping it based on user movement, while waiting for the next real frame to come down the pipe, then it could emulate a higher perceived frame rate while only needing 60fps from the source device.

If I understand it correctly Time Warp requires very base level access to the actual GPU drivers and bypasses a lot of default behaviors for modern GPUs in terms of buffering. It's why GearVR has very advanced Carmack time warp while the PC still does not, because Samsung let Oculus have higher access but getting that from NVIDIA/AMD/DirectX hasn't happened on PC. I don't doubt that Sony will allow that access on its own console to give Morpheus time warp, but at best it's "framerate insurance" that can eliminate judder from dropping a frame or two, and is not some sort of magic that can double the framerate.

 

[bj00rn_]

Perhaps I shouldn't have used interpolation as a term. If it is taking an existing frame and warping it based on user movement, while waiting for the next real frame to come down the pipe, then it could emulate a higher perceived frame rate while only needing 60fps from the source device.

Does it really though? Is it really plausible that timewarp can do that much to the framerate itself? I thought it was mostly about transients and latency.

 

[Mechanical Snowman]

When they say it isn't important, they mean that the overall result is the important thing, not what the resolution is specifically.

The resolution is one of the largest factors in the overall result.

Also, keep in mind that whatever Samsung (presumably) screen they are using might be an unannounced display from their next generation of phones, so there might be NDA issues for them announcing exact specs. If I remember correctly they couldn't even say which company manufactured the OLED screen for Crystal Cove/DK2, even though there honestly was only one possibility (until people took it apart and saw it was literally the front of a Note 3).

I've seen this theory on Reddit too and it seems plausible. I hope that's the case.

Does it really though? Is it really plausible that timewarp can do that much to the framerate itself? I thought it was mostly about transients and latency.

Timewarp as a concept isn't designed as a method of general interpolation to double framerate, its main use is fixing occasional dropped frames and reducing latency to 0ms. I don't see why in theory it couldn't boost the frame rate artificially, but this isn't what it's designed to do.

 

[Kyoufu]

Have they mentioned if CV1 will have access to the SDK? Or do you need DK2 in order to develop software for it?

 

[Kritz]

I've been dorking around with Unity and the Rift today and ended up with whatever the crap this is: https://www.youtube.com/watch?v=bjLoO_wl1Pw

I don't know if I feel proud or disgusted in myself.

 

[Seanspeed]

The resolution is one of the largest factors in the overall result.

Its only a factor, though. When Crescent Bay first debuted, some people were led to believe it *must* be a 4k display in there. If it looks great, then why does it matter if its a 1440p or a 720p display? If they've got some trickery going on that creates a much improved image over the DK2 without having to bump up resolution too much, then that's great. Don't get me wrong, I still want 1440p, but they're right that the end result is what's important, not just what the spec sheet says.

 

[Mechanical Snowman]

I've been dorking around with Unity and the Rift today and ended up with whatever the crap this is: https://www.youtube.com/watch?v=bjLoO_wl1Pw

I don't know if I feel proud or disgusted in myself.

Get a kickstarter going

Its only a factor, though. When Crescent Bay first debuted, some people were led to believe it *must* be a 4k display in there. If it looks great, then why does it matter if its a 1440p or a 720p display? If they've got some trickery going on that creates a much improved image over the DK2 without having to bump up resolution too much, then that's great. Don't get me wrong, I still want 1440p, but they're right that the end result is what's important, not just what the spec sheet says.

Because this severely limits what you can do in the Rift. Virtual Desktop right now is hit hard by the 1080p being inadequate. Space Engine's UI is barely readable. Elite Dangerous' holographic cockpit UI is hard to read. Or really, anything with text.

I don't want every Rift demo to have text like a children's book to be able to read it, and as great as the Oculus guys seem to be about being able to do more with less, there is no magic solution to this problem: you just need more pixels.

 

[Kritz]

Get a kickstarter going

one game at a time

maybe once the global game jam comes up I'll see if I can flesh out the idea a tiny bit and let people play it.

 

[Big Takeover]

I hope you're right, it's just weird behavior IMO from a company that is usually quite transparent, so it makes me think the worst.

Ye of little faith.

I go completely the other way on that. I think whatever the screen/optics they've got in this (old) prototype, they believe it's nonrepresentative, and really not worth diverting people's attention away from things that are, such as the audio, ergonomics, and improved tracking.

Really though, it's not going to be 1080p. 1440p is the baseline, and whether we get more resolution, more hertz, a different aspect ratio, or anything else above and beyond is where the conversation starts.

Because is was implied that this screen isn't "The One", what it actually is becomes nothing more than a curiosity. I hope it was 1080p, or some other rez to be able to achieve 90Hz. That means they've made advances, and there is still more to come.

 

[Seanspeed]

Because this severely limits what you can do in the Rift. Virtual Desktop right now is hit hard by the 1080p being inadequate. Space Engine's UI is barely readable. Elite Dangerous' holographic cockpit UI is hard to read. Or really, anything with text.

I don't want every Rift demo to have text like a children's book to be able to read it, and as great as the Oculus guys seem to be about being able to do more with less, there is no magic solution to this problem: you just need more pixels.

Again, you're ignoring that resolution isn't the only factor here. They've been going out of their way to say that they can make meaningful gains in clarity without necessarily increasing resolution. That's all they meant by that comment. Its not like they're incompetent and don't understand how resolution works...

 

[Astral/H3X]

Get a kickstarter going


Because this severely limits what you can do in the Rift. Virtual Desktop right now is hit hard by the 1080p being inadequate. Space Engine's UI is barely readable. Elite Dangerous' holographic cockpit UI is hard to read. Or really, anything with text.

I don't want every Rift demo to have text like a children's book to be able to read it, and as great as the Oculus guys seem to be about being able to do more with less, there is no magic solution to this problem: you just need more pixels.

I don't necessarily think its that you need MORE pixels, so much as you need BETTER pixels. Changing the UI to green in Elite creates massive gains in the readability of the text, and the primary reason for this is the pentile display. If they can reduce the gap between pixels, and then get a more even size ratio between the different subpixels, that ALONE would make a huge difference in usability, without even changing the resolution.
Now, I do fully expect them to go to a 1440 display for CV1, but what I would ACTUALLY like them to do is have a 4K display, being run at 1080p. If they do that, they can have text be rendered at a higher resolution on top of the 1080p. What this would do is improve perceived resolution on everything (due to lowered SDE, and less artifact caused by pentile), as well as vastly improve the readability of text, without completely destroying the performance.

 

[Mechanical Snowman]

Again, you're ignoring that resolution isn't the only factor here. They've been going out of their way to say that they can make meaningful gains in clarity without necessarily increasing resolution. That's all they meant by that comment. Its not like they're incompetent and don't understand how resolution works...

It's not about incompetence, it's about where their priorities lie and what they think will work best for the majority of the software developed for the Rift. Say 1080p would still be sufficient for 90% of the content developed, and they calculate a cost that isn't viable to up the resolution for that 10% of content that would benefit, they might not consider it to be worth it. And while I'm pulling those percentages from the air, I think they work as an example.

Ye of little faith.

I know I know, it's my biggest problem.

I don't necessarily think its that you need MORE pixels, so much as you need BETTER pixels. Changing the UI to green in Elite creates massive gains in the readability of the text, and the primary reason for this is the pentile display. If they can reduce the gap between pixels, and then get a more even size ratio between the different subpixels, that ALONE would make a huge difference in usability, without even changing the resolution.
Now, I do fully expect them to go to a 1440 display for CV1, but what I would ACTUALLY like them to do is have a 4K display, being run at 1080p. If they do that, they can have text be rendered at a higher resolution on top of the 1080p. What this would do is improve perceived resolution on everything (due to lowered SDE, and less artifact caused by pentile), as well as vastly improve the readability of text, without completely destroying the performance.

Absolutely a better subpixel layout goes a long way, but there's still only so far you can push it when it comes to smaller details like text, or say, stars in a space game. I can use the DK2 lenses on my Xperia Z, which is not pentile, and the pixel fill and subpixel layout is great. It's a huge improvement over DK2's screen, but small text is still pixelated, and there is no way, even after they've achieved some hypothetical perfect pixel fill, to further improve on that other than to up the resolution.

Anyway, I'm thankful that it looks like I'm in the minority with my pessimism. The theory that I actually think is the most likely, is that CB is running a custom screen that is under NDA and that it will be 1440p or some non standard resolution close to that. I just don't buy the "we're not telling you because it's not important" thing, because it's totally out of character for Oculus. The kind of answers that Nate gives when asked about the resolution seem more like forced secrecy converted into some PR appropriate speech than the straight forward truth.

 

[Seanspeed]

The kind of answers that Nate gives when asked about the resolution seem more like forced secrecy converted into some PR appropriate speech than the straight forward truth.

That's without a doubt what it is.

I think you're just *way* overthinking it. Nate does a ton of interviews and the whole 'resolution isn't important' thing is probably not the best way he could have phrased things at that moment in time for that question in that specific interview. He speaks elsewhere about how resolution *is* important so again, its not like they are just unaware of this all works.

 

[Mechanical Snowman]

I think you're just *way* overthinking it. Nate does a ton of interviews and the whole 'resolution isn't important' thing is probably not the best way he could have phrased things at that moment in time for that question in that specific interview. He speaks elsewhere about how resolution *is* important so again, its not like they are just unaware of this all works.

I don't think I can remember an interview where he doesn't downplay the resolution, do you have a link?

It's this interview where most of my concerns come from, and I don't think I'm reading 'way too much' into it to draw the conclusions that I'm concerned about.

Nate: And so whenever we're looking at display technology and optics, what we wanna do is have them compliment each other such that we are, um, optimizing for things that we think are important. With DK2 for example, there was a much sharper, much clearer image, but that actually resulted in significantly more screen door effect, basically, and visibility of sort of the PenTile display. And so what you see in Crescent Bay is that we're doing a lot more to combine the optics and the display to smooth out some of those pixels, or the space between the pixels -
Norm: Optically smoothing out some of those pixels?
Nate: Exactly, exactly, so that it looks, effectively, like a higher resolution display.
Norm: Does that affect the readability for text for example and how people will display text?
Nate: That's a good question *laughs* ... So, you're absolutely right, so in terms of the clarity, the text legibility is one of the challenges. It's not a major challenge because as long as, you know, just like any Rift, if you have large text it's very easy to read. So what we're talking about is when text gets very very small, it does tend to get blurry, or a little crunchy, um, so that's something we're going to improve on. And overall, you know, we talked sort of high level about resolution, we do want to see major resolution improvements, over the next, you know, two, three, four, iterations of the product, you know Abrash talks a lot about 'we do wanna head toward 4K, 8K' ... So, we do wanna see higher resolution displays get into the Rift over the long term, this (Crescent Bay) is still where we are, today.

I've bolded some of the obvious clues, but really there's nothing in that whole interview segment that doesn't hint at low resolution for CV1. I can go back and read that transcript as objectively as possible, putting fears or pessimism aside or whatever, and I still come to the same conclusion. As you said a few times, they aren't incompetent, they know how resolution works. They know where the current screen and optics deliver and where they struggle, and Nate's answer implies that the "challenges" won't be addressed for the first consumer versions, but in later ones.

 

[mrklaw]

Oculus have been very open in the past, but now they're preparing to release something for consumers, it is sensible that they want to keep some things close to their chest. As soon as CV1 is released, every tech company out there will be tearing it down to see whats inside - you'll probably find out the exact screen being used within hours. But there is no value in Oculus telling us now, because that gives OEMs 6 months or more lead time in prepping competitors, or tweaking what they already have in the pipeline.

 

[Astral/H3X]

Absolutely a better subpixel layout goes a long way, but there's still only so far you can push it when it comes to smaller details like text, or say, stars in a space game. I can use the DK2 lenses on my Xperia Z, which is not pentile, and the pixel fill and subpixel layout is great. It's a huge improvement over DK2's screen, but small text is still pixelated, and there is no way, even after they've achieved some hypothetical perfect pixel fill, to further improve on that other than to up the resolution.

Anyway, I'm thankful that it looks like I'm in the minority with my pessimism. The theory that I actually think is the most likely, is that CB is running a custom screen that is under NDA and that it will be 1440p or some non standard resolution close to that. I just don't buy the "we're not telling you because it's not important" thing, because it's totally out of character for Oculus. The kind of answers that Nate gives when asked about the resolution seem more like forced secrecy converted into some PR appropriate speech than the straight forward truth.

Oh there's no doubt as to whether or not resolution improvements will be great for it... but I think he's trying to temper expectations from people who might be expecting a 4k display on it. If a WHOLE BUNCH OF PEOPLE believe that the only way to improve these things is to increase the resolution significantly, then when it comes out and it's only a 1440p display, those same people are gonna say that they got shafted, that text will be horrible and the SDE will be in full effect, so on and so forth, without really even trying it to find out if that's even the case. When in fact having smaller gaps between pixels, better arrangement of subpixels, and possibly even reworked lenses could make just as much of an improvement on these factors as the resolution itself, at a fraction of the increase in computational requirements.

 

[zoukka]

Tried some Oculus porn at work...

There's no going back. Porn will be the first type of content that will sell VR to people mark my words.

Also played Alien Isolation and it was very nice though you can get motion sickness in minutes which is a bummer.

 

[Big Takeover]

I don't think I can remember an interview where he doesn't downplay the resolution, do you have a link?

It's this interview where most of my concerns come from, and I don't think I'm reading 'way too much' into it to draw the conclusions that I'm concerned about.

Nate: And so whenever we're looking at display technology and optics, what we wanna do is have them compliment each other such that we are, um, optimizing for things that we think are important. With DK2 for example, there was a much sharper, much clearer image, but that actually resulted in significantly more screen door effect, basically, and visibility of sort of the PenTile display. And so what you see in Crescent Bay is that we're doing a lot more to combine the optics and the display to smooth out some of those pixels, or the space between the pixels -
Norm: Optically smoothing out some of those pixels?
Nate: Exactly, exactly, so that it looks, effectively, like a higher resolution display.
Norm: Does that affect the readability for text for example and how people will display text?
Nate: That's a good question *laughs* ... So, you're absolutely right, so in terms of the clarity, the text legibility is one of the challenges. It's not a major challenge because as long as, you know, just like any Rift, if you have large text it's very easy to read. So what we're talking about is when text gets very very small, it does tend to get blurry, or a little crunchy, um, so that's something we're going to improve on. And overall, you know, we talked sort of high level about resolution, we do want to see major resolution improvements, over the next, you know, two, three, four, iterations of the product, you know Abrash talks a lot about 'we do wanna head toward 4K, 8K' ... So, we do wanna see higher resolution displays get into the Rift over the long term, this (Crescent Bay) is still where we are, today.

I've bolded some of the obvious clues, but really there's nothing in that whole interview segment that doesn't hint at low resolution for CV1. I can go back and read that transcript as objectively as possible, putting fears or pessimism aside or whatever, and I still come to the same conclusion. As you said a few times, they aren't incompetent, they know how resolution works. They know where the current screen and optics deliver and where they struggle, and Nate's answer implies that the "challenges" won't be addressed for the first consumer versions, but in later ones.

If you're more concerned with how good the image will be on CV1, rather than actually worrying about it being 1080p, then I think you have reason to be unsettled.

The bottom line is they're still going to have to employ those same tactics with 1440p since it's still not going to be enough to be crystal clear. There are no 4K phone screens now, or even in the immediate future, so this is what we'll be stuck with.

The question is really where in the development cycle do these Crescent Bay screen/optics lie? If these are the best screens (1440p) with the best optics, essentially the ones that will be in the consumer Rift, and they're still downplaying, then that's not a good sign. DK2.5 seems to be the general consensus.

If they're not, and I personally think they're not, then what he said doesn't change anything. It's still the route they need to go down, no matter what resolution this prototype runs at.

 

[Astral/H3X]

Tried some Oculus [/b]porn at work...

There's no going back. Porn will be the first type of content that will sell VR to people mark my words.

Ah, good ol stealth use of it.

 

[Chubzdoomer]

For those of you with Oculus dev kits: once everything is sorted out and the screen is improved as much as it can be, do you think this really will be the next big thing in gaming? After having played games in virtual reality, for example, do you find that it's nearly impossible (or at least very difficult) to go back to playing games on a regular monitor or TV due to the complete lack of immersion in comparison?

I keep thinking of how amazing games like Alien: Isolation, Elite Dangerous, Half-Life 2, and even Euro Truck Simulator 2 must be in virtual reality, and how immersed you must feel in their detailed, atmospheric worlds. Those games are already immersive enough on a flat screen, so I'm sure they're absolutely mind-blowing in virtual reality. If it's anything like I imagine, I know for a fact that when I try the Oculus for the very first time, it's going to blow my socks halfway across the room!

 

[Tain]

For those of you with Oculus dev kits: once everything is sorted out and the screen is improved as much as it can be, do you think this really will be the next big thing in gaming? After having played games in virtual reality, for example, do you find that it's nearly impossible (or at least very difficult) to go back to playing games on a regular monitor or TV due to the complete lack of immersion in comparison?

I keep thinking of how amazing games like Alien: Isolation, Elite Dangerous, Half-Life 2, and even Euro Truck Simulator 2 must be in virtual reality, and how immersed you must feel in their detailed, atmospheric worlds. Those games are already immersive enough on a flat screen, so I'm sure they're absolutely mind-blowing in virtual reality. If it's anything like I imagine, I know for a fact that when I try the Oculus for the very first time, it's going to blow my socks halfway across the room!

Having played all of Quake 2, most of Half-Life 2, a good chunk of Alien Isolation (I have trouble proceeding due to how horrifying it is), a few hours of Elite, and a chunk of credits of Suwapyon 2, as well as tons of demos and a lot of time testing my game that I'm working on...

It will be big. It's just too immersive to not be. If I can play a game on my DK2 (as in, if the game has passable support), it's absolutely the best way to play it.

 

[DunderHassan]

For those of you with Oculus dev kits: once everything is sorted out and the screen is improved as much as it can be, do you think this really will be the next big thing in gaming? After having played games in virtual reality, for example, do you find that it's nearly impossible (or at least very difficult) to go back to playing games on a regular monitor or TV due to the complete lack of immersion in comparison?

Probably yes, why play on a 20"-30" monitor when you can play pretty much any game on a virtual 100"+ screen?
Why play in "2D" when you can be inside the game world?
I think it's highly unlikely that CV1 will be enough to replace montiors though and I barely play any games on DK2 due to the horrible screen door effect (the OR support in most games are kinda shitty too though), can't even bother to watch movies in it.

Speaking of shitty OR support, how do you guys get Assetto Corsa to work with OR?
Tried to follow a few guides, but it just crashes on me every time (some create texture error).
Have amd 7970 gpu.

 

[Yaboosh]

Having played all of Quake 2, most of Half-Life 2, a good chunk of Alien Isolation (I have trouble proceeding due to how horrifying it is), a few hours of Elite, and a chunk of credits of Suwapyon 2, as well as tons of demos and a lot of time testing my game that I'm working on...

It will be big. It's just too immersive to not be. If I can play a game on my DK2 (as in, if the game has passable support), it's absolutely the best way to play it.

But third person games don't really work do they?

 

[Kritz]

I don't think VR will replace traditional games, instead it's likely to generate new types of experiences that are going to be entirely incompatible with what we're playing now. I think there will always be a place for the monitor and television when it comes to games, and I'd sooner think of VR as the birth of a medium of its own, rather than a tool to consume existing game media.

I'm just excited at the possibility that 'experience' games are coming into a genre of their own, with hardware that perfectly compliments them. I'm just so tired of shootin' dudes.

 

[Tain]

But third person games don't really work do they?

There isn't a lot of content out there, but from what I've played, they work fine. There's nothing incompatible about it. When you play a third-person game you're still in that world, just as a floating spectator rather than the character. Suwapyon 2 is largely a vertical-scrolling 2D shooter placed on a 3D playfield, and the bird's-eye view works excellently (the positional tracking makes the game feel like it's happening on some magic scrolling table in front of you). Skyroads, the old DOS forward-scrolling platformer thing works excellently as well. And on the Gear VR, HeroBound (a roughly Zelda-esque game with somewhat fixed cameras made by Oculus themselves) is super polished and comfortable from my brief time with it.

edit: actually, I should try the UE4 third person template project in VR, just to see what a more traditional third person game is like.

 

[revolverjgw]

For those of you with Oculus dev kits: once everything is sorted out and the screen is improved as much as it can be, do you think this really will be the next big thing in gaming? After having played games in virtual reality, for example, do you find that it's nearly impossible (or at least very difficult) to go back to playing games on a regular monitor or TV due to the complete lack of immersion in comparison?

I keep thinking of how amazing games like Alien: Isolation, Elite Dangerous, Half-Life 2, and even Euro Truck Simulator 2 must be in virtual reality, and how immersed you must feel in their detailed, atmospheric worlds. Those games are already immersive enough on a flat screen, so I'm sure they're absolutely mind-blowing in virtual reality. If it's anything like I imagine, I know for a fact that when I try the Oculus for the very first time, it's going to blow my socks halfway across the room!

Coolest thing I've experienced since the NES in 1986, though I don't necessarily prefer games on the DK2 over a monitor yet... the resolution just isn't there yet (and Alien Isolation has an unfixable world scale problem that ruins it for me). In fact my first impression of the DK1 wasn't how immersive it was, it was how hard it was to see anything. But good VR experiences let you forget about that pretty easily, and HL2 on the DK1 felt like a totally new game to me and one of the most important moments I've had in gaming. And certainly once resolution gets up to a certain standard (with no/minimal screendoor), this will be the only way I want to play. Elite Dangerous is the one full game that's already at that point... if I play it on my monitor, half the appeal is lost and it borders on depressing. You lose the totally intuitive headtracking and the mindblowing sense of scale.

There are non-VR games that can be played with third party programs and they're flawed but extremely exciting when I think about what it means for games like these down the line . Skyrim runs like crap on Vorpx for me but it's obvious that if it ran well and was in a higher res, it would be a revelatory experience, there are moments of intoxicating immersion. Walking through a snowstorm in the woods with snow blowing in my face and the feeling of physically being surrounded by trees... just being there is fun. "Walking simulators" will be more viable than ever. I'm primarily interested in traditional gaming experiences like Skyrim, but there's going to be a lot more interesting applications for VR, virtual tourism and the like.

I'm not extremely confident about it taking off immediately, there are some hurdles in the way. I don't know many people who have powerful desktops anymore, and you're gonna need one to run these games at the 1440p minimum at 90hz. I really don't how know this thing is going to fit into the market. But people do love this stuff... it's quite amazing watching them literally duck for cover when an asteroid bumps into the hull in Elite, you realize there was no precedent for anything remotely resembling an experience like this even a year and a half ago. Show them that same scene on a monitor and it wouldn't have any real impact on them, it's just another space video game. It really is a potential game changer but some logistical problems need to be overcome, and I'm not informed enough to speculate on whether that will happen any time soon. I'll be following it closely and hoping they market it correctly and that people are ready. All I know is what it means for me... I'll shell out any amount of money to get the most powerful rig for VR because it's that good, I'm prepared to pour most of my disposable income into it. Just keep in mind the existing weaknesses when you try it on for the first time.

 

[InaudibleWhispa]

For those of you with Oculus dev kits: once everything is sorted out and the screen is improved as much as it can be, do you think this really will be the next big thing in gaming? After having played games in virtual reality, for example, do you find that it's nearly impossible (or at least very difficult) to go back to playing games on a regular monitor or TV due to the complete lack of immersion in comparison?

I keep thinking of how amazing games like Alien: Isolation, Elite Dangerous, Half-Life 2, and even Euro Truck Simulator 2 must be in virtual reality, and how immersed you must feel in their detailed, atmospheric worlds. Those games are already immersive enough on a flat screen, so I'm sure they're absolutely mind-blowing in virtual reality. If it's anything like I imagine, I know for a fact that when I try the Oculus for the very first time, it's going to blow my socks halfway across the room!

For some genres, such as cockpit games (driving and games such as Elite), it'll become the standard. I haven't played Elite outside of VR yet, and the only thing standing in its way of being perfect is resolution, which is obviously a temporary hardware problem. Games like Half-Life 2 are incredibly immersive in VR, but standard FPS's ported to VR are going to be hit and miss depending on the work put into them and how their pacing is. Games built for VR will be much stronger.

The cool thing about VR is that a lot of the current demos and experiences would be utterly mundane and trivial on a monitor. The experience is that immersive and mind-blowing, that it can elevate genres and experiences we'd otherwise dismiss and in that regard the potential is huge.

But third person games don't really work do they?

Yeah, they work really well. Deeper is a really cool VR experience. You feel like you're looking at a little model or toy that you can control. There's a platformer called Lucky's Tale which is supposed to be a lot of fun too. A traditional platformer that incorporates being able to look around and under things with your head.

 

[mrklaw]

There isn't a lot of content out there, but from what I've played, they work fine. There's nothing incompatible about it. When you play a third-person game you're still in that world, just as a floating spectator rather than the character. Suwapyon 2 is largely a vertical-scrolling 2D shooter placed on a 3D playfield, and the bird's-eye view works excellently (the positional tracking makes the game feel like it's happening on some magic scrolling table in front of you). Skyroads, the old DOS forward-scrolling platformer thing works excellently as well. And on the Gear VR, HeroBound (a roughly Zelda-esque game with somewhat fixed cameras made by Oculus themselves) is super polished and comfortable from my brief time with it.

edit: actually, I should try the UE4 third person template project in VR, just to see what a more traditional third person game is like.

do third person games generate nausea? I'm curious because your 'head' in the game is moving without your body moving, which causes issues in FPS games. Does the '3rd person' part keep you removed enough from things that your brain doesn't consider you to be 'in' the world, and therefore you don't become sick?