is the Rift's resolution going to be high enough?

[PatimPatam]

Hey Dycus, posted this back in July on the main oculus thread but didn't get a reply from Palmer..

I understand the main benefits of having a single screen as explained here:
http://www.mtbs3d.com/phpBB/viewtopic.php?f=140&t=15301

However for the consumer version, do you think a 2-screen design similar to the following would be possible in order to improve FOV, resolution and profile?



(approximate sizes/angles)

[zacherynuk]

Hey Dycus, posted this back in July on the main oculus thread but didn't get a reply from Palmer..

I understand the main benefits of having a single screen as explained here:
http://www.mtbs3d.com/phpBB/viewtopic.php?f=140&t=15301

However for the consumer version, do you think a 2-screen design similar to the following would be possible in order to improve FOV, resolution and profile?



(approximate sizes/angles)

Not quite as simple as that: using high magnification lenses, which is a requirement, the screen really has to be presented to the lens flat - otherwise large portions of the screen are out of focus. millimetres matter.

Therefore using two screens which cannot be combined to be under the average IPD require the use of mirrored surfaces, or fresnels. This adds weight, complexity and can degrade image quality.

Further, utilising a single screen, of a standard resolution, allows standard outputs to be used, eg consoles and entry level PC's - multiple screens would introduce the requirement for on-board electronics to split the image or two discrete inputs.

[PatimPatam]

Not quite as simple as that: using high magnification lenses, which is a requirement, the screen really has to be presented to the lens flat - otherwise large portions of the screen are out of focus. millimetres matter.

Thanks for the reply zacherynuk. Silly question maybe: could the lenses not be slightly inclined as well, so they are at the same angle as the screen?

[EdZ]

Not quite as simple as that: using high magnification lenses, which is a requirement, the screen really has to be presented to the lens flat - otherwise large portions of the screen are out of focus. millimetres matter.

I don't think that's an issue: as long as the lenses and displays are parallel, you can tilt them out as far as you want as long as your eye remains within the exit pupil. Too far and you squeeze the exit pupil (and thus the available FoV) unless you use a larger lens, but a few degrees off-normal shouldn't be an issue.

The display connection is the real issue of dual-panel setups. Either you use a custom driver board setup in order to present the panels as a single display to the computer, or you connect both displays and write custom drivers to handle dual-panel stereo. The hardware route is more expensive (either a custom dual-panel driver and input-splitting board, or two single-panel COTS driver boards and a COTS splitter) but easier to use with existing software, and the software-only approach will almost certainly conflict with existing GPU drivers and may have synchronisation issues.

[zacherynuk]

Not quite as simple as that: using high magnification lenses, which is a requirement, the screen really has to be presented to the lens flat - otherwise large portions of the screen are out of focus. millimetres matter.

Thanks for the reply zacherynuk. Silly question maybe: could the lenses not be slightly inclined as well, so they are at the same angle as the screen?

Does this answer you question

[zacherynuk]

Not quite as simple as that: using high magnification lenses, which is a requirement, the screen really has to be presented to the lens flat - otherwise large portions of the screen are out of focus. millimetres matter.

I don't think that's an issue: as long as the lenses and displays are parallel, you can tilt them out as far as you want as long as your eye remains within the exit pupil. Too far and you squeeze the exit pupil (and thus the available FoV) unless you use a larger lens, but a few degrees off-normal shouldn't be an issue.

You can get away with very little, I have found, even with fresnel help; hence people turning to mirrors and polarisation to bring the images together.

[PatimPatam]

Hahah, good one!

Anyway from what I understand you can still move your eyes while wearing the rift, so I was thinking it could work even if your eyes are looking straight but not perfectly aligned with the lenses (maybe you would lose a bit of resolution only as Palmer mentioned a side effect of the distortion is that you have more pixels in the center).

Just curiosity, i admit i have no idea about optics!

[zacherynuk]

Hahah, good one!

Anyway from what I understand you can still move your eyes while wearing the rift, so I was thinking it could work even if your eyes are looking straight but not perfectly aligned with the lenses (maybe you would lose a bit of resolution only as Palmer mentioned a side effect of the distortion is that you have more pixels in the center).

Just curiosity, i admit i have no idea about optics!

I'd say print off some screen sized pictures and glue them to some screen sized bits of modelling board.
Buy a heap of lenses, mirrors, prisms and give it a go! Just cheap science kits etc.

I have spent about £500-1000 on optical whatnots and I am still astounded that ideas I have just don't want to work. 7cm concave mirrors are my current 'thing' - an absolute bugger to line up and hold in position, but when you do it's very good for twin screen viewing; I am trying to adapt for single larger screen viewing, but difficult to keep weight down.

[rfurlan]

Nvidia's 3DTV Play is just a piece of software that supports the HDMI 1.4 3D specs. Using the 3DTV Drivers you should be able to play content on any HDMI device that supports 3D regardless of certification. I am not sure why it isn't working for you

That would make sense, but it actually doesn't work like that. Nvidia has to whitelist "certified" displays in their driver (filtered using the EDID). So unless your particular device has been approved by Nvidia, it won't work (barring any hacking). I have asked Andrew Fear about this point-blank, and he claimed it was so they could ensure a quality experience and that they don't charge vendors for this "service". Honestly it sounds fishy to me, but that's the official word.

This makes me feel like asking for a refund, not what I thought I was paying for :/

[mayaman]

First of all, lets put things in perspective. Up until as recently as a year ago (even less) all you could buy was 30-35 degree sub-HD headsets that pretty much sucked. Its just recently we have broke into HD HMDs with the HMZ and ST1080. So even a 1280x800 display split per eye is not that shabby. And the FOV is absolutely massive compared to previous efforts. So this is a HUGE leap forward for consumer VR. Lets not forget whats going on here. Is it the be-all-end-all? I sure hope not. If everything goes well we will have a real consumer product coming 2013-14 with hopefully better specs. So I think that time will come. But lets be happy with what we are getting here, it really is an amazing step forward.

Secondly, I don't think higher resolution make things more immersive or realistic even. Ever since HD hit the scene everyone is obsessed with resolution, but that doesn't make an experience compelling. VR is really about the experience, and I think a lot of that has to do with immersion, not visual fidelity. I mean, back when VR started they were rendering with flat-shaded polygons with polygon-budgets in the 100's. But they still investigated interesting concepts because they had immersive interactions, for example with head-tracked HMDs, data-gloves, etc. Its the rich interaction that makes a VR experience compelling. And you don't need a super-computer or 4K displays to do that. Beyond specialized hardware, and think there are real gains to be made with better character animation, more intensive physics simulations, and computer AI. All of this is stuff we can do now, people are just not spending enough time and money to do it.

Just listen to this guy please.

[rfurlan]

...Secondly, I don't think higher resolution make things more immersive or realistic even...

Just listen to this guy please.

It isn't as simple as that. Higher resolutions don't necessarily make anything more realistic, but lower resolutions make realism harder to achieve depending on the size of the virtual space. You must have enough pixels to encode the information necessary to convey the experience at hand. Large virtual spaces will not feel realistic at 640x800 per eye with a wide FOV because there will not be enough pixels to encode everything on the scene and too much information will be lost. The key concept here being information loss - larger spaces with more "stuff" in it will require more pixels to transmit the information to the user. Smaller spaces will be fine though.

Playing Crysis (one) at 800x600 would illustrate this issue nicely - give it a try

In contrast, Doom 3 at 800x600 isn't bad at all.

[greenknight]

...Secondly, I don't think higher resolution make things more immersive or realistic even...

Just listen to this guy please.

It isn't as simple as that. Higher resolutions don't necessarily make anything more realistic, but lower resolutions make realism harder to achieve depending on the size of the virtual space. You must have enough pixels to encode the information necessary to convey the experience at hand. Large virtual spaces will not feel realistic at 640x800 per eye with a wide FOV because there will not be enough pixels to encode everything on the scene and too much information will be lost. The key concept here being information loss - larger spaces with more "stuff" in it will require more pixels to transmit the information to the user. Smaller spaces will be fine though.

Playing Crysis (one) at 800x600 would illustrate this issue nicely - give it a try

In contrast, Doom 3 at 800x600 isn't bad at all.

Has anybody tried skyrim to confirm this idea? It does make some intuitive sense though.

But, again, the oculus is a great deal. There isn't anyone who can say otherwise as it stands in a class of its own.

[MaterialDefender]

...The key concept here being information loss - larger spaces with more "stuff" in it will require more pixels to transmit the information to the user. Smaller spaces will be fine though.

Playing Crysis (one) at 800x600 would illustrate this issue nicely - give it a try

I don't want to deny that this effect exists, but I don't think it's a really big problem for the following reason: If you simply play a modern game like Crysis at a low resolution it does look awful. No doubt about it. But the main problem is not the information loss, the main problems are various sampling and aliasing issues, that lead to much flickering and a generally unsteady image. With 2x2 or better 4x4 supersampling (should be possible with high end graphics cards for many games at the Rifts resolution) these problems can be reduced significantly. You get an image as steady and good looking as a 1920x1080 image, just less sharp. Similar to a photo that is downsized to a lower resolution. I'm quite sure that the human brain will be able to adapt to this reduced sharpness pretty quickly, as long as you don't look into a world that is made of aliasing flicker and shader crawling.

[rfurlan]

...The key concept here being information loss - larger spaces with more "stuff" in it will require more pixels to transmit the information to the user. Smaller spaces will be fine though.

Playing Crysis (one) at 800x600 would illustrate this issue nicely - give it a try

I don't want to deny that this effect exists, but I don't think it's a really big problem for the following reason: If you simply play a modern game like Crysis at a low resolution it does look awful. No doubt about it. But the main problem is not the information loss, the main problems are various sampling and aliasing issues, that lead to much flickering and a generally unsteady image. With 2x2 or better 4x4 supersampling (should be possible with high end graphics cards for many games at the Rifts resolution) these problems can be reduced significantly. You get an image as steady and good looking as a 1920x1080 image, just less sharp. Similar to a photo that is downsized to a lower reason. I'm quite sure that the human brain will be able to adapt to this reduced sharpness pretty quickly, as long as you don't look into a world that is made of aliasing flicker and shader crawling.

I agree that the problem can be reduced to some extent with super-sampling but ultimately there is a limit of how much information you can pack in NxN pixels. If your view frustum captures significantly more information than you can express with the pixels available to you, information loss will become a significant driver of fidelity loss. Crysis is a good example of this because at times it asks you to survey and then assault large enemy installations. Even if you completely solved the flickering issue (through magic, lets say) you would still be dealing with massive information loss when you have only 20 or so pixels to render an enemy sniper or a mission objective.

That said, I am confident that resolution will not be a problem for applications designed around the Rift's specs

[2EyeGuy]

I think the (not finalised, may be higher) 640x800 resolution will be too low for some games, unless there is a zoom function. Unfortunately, without 6DOF you can't zoom the human way that Lister suggested to Kryten ("just move your head closer to the object), so it will need a button.
I tried 640x480 on the VR920, and it was good for slow-paced first-person style games that I like, but it was annoying not being able to read cockpit displays in racing games and flight simulators. The extra vertical resolution of the Rift probably won't help so much with reading writing, because that's normally horizontal.
Hawken has a custom cockpit deliberately designed for the Rift though, so that should prevent those problems in that game.

Anyway, I think a zoom option would help a lot.

[MaterialDefender]

I agree that the problem can be reduced to some extent with super-sampling but ultimately there is a limit of how much information you can pack in NxN pixels. If your view frustum captures significantly more information than you can express with the pixels available to you, information loss will become a significant driver of fidelity loss. Crysis is a good example of this because at times it asks you to survey and then assault large enemy installations. Even if you completely solved the flickering issue (through magic, lets say) you would still be dealing with massive information loss when you have only 20 or so pixels to render an enemy sniper or a mission objective.

That said, I am confident that resolution will not be a problem for applications designed around the Rift's specs

Maybe I'm a bit less pessimistic about this because I'm slightly nearsighted, so I know the effect of information loss pretty well completely without VR equipment of any kind. And can't remember a single moment in my life where I felt my immersion level drop in the real world just by taking off my glasses.

But the sharper, the better, that much is clear.

[MrGreen]

But the sharper, the better, that much is clear.

[Fredz]

I don't want to deny that this effect exists, but I don't think it's a really big problem for the following reason: If you simply play a modern game like Crysis at a low resolution it does look awful. No doubt about it. But the main problem is not the information loss, the main problems are various sampling and aliasing issues, that lead to much flickering and a generally unsteady image.

Aliasing will certainly be an issue with the Rift, but I think what rfurlan said will be a much bigger problem for outdoor games.

When Crysis is played on a HDTV it has an approx. 30° HFOV when using recommended viewing distances, generally with 1280 to 1920 pixels horizontally. The Rift has a 90° HFOV with 640 pixels horizontally.

So for the same central FOV of 30°, you would have only 213 pixels available on the Rift (640/3). Probably a little bit more since it's not uniform, but quite difficult to calculate without knowing the optics.

With this low resolution I think objects or people at a distance will simply not be rendered because they'll account for less than a pixel. No kind of anti-aliasing or anisotropic filtering will help in this case.

[cybereality]

To be honest, I played Crysis on the VR920 (at 1024x768 downscaled to 640x480) and it still looked great. It was totally playable.

[MaterialDefender]

With this low resolution I think objects or people at a distance will simply not be rendered because they'll account for less than a pixel. No kind of anti-aliasing or anisotropic filtering will help in this case.

That's only (partly) true for multisampling and even more so all kinds of modern Shader(pseudo-)AA modes. Good old brute force supersampling does exactly that since the image is rendered at a higher resolution and then downsampled (not exactly, but it's the same effect), which allows for details to be visible (though blurred) that would be smaller than 1px if the image was rendered at the low resolution natively. Not infinitely smaller though, of course. The attached image demonstrates that. The left 512x512 image has a black pixel every 4 pixels, the right image is downsampled (photoshop bicubic) to 256x256 pixels. The pixel grid is still visible in the right image although 1 pixel in the left image theoretically is a quarter pixel in the right image. This is more or less the effect of raw ordered grid 2x2 supersampling, and it works perfectly in motion too.

But you're absolutely right about the blurriness of the image. Won't look too great for anyone accustomed to HD resolutions. But hey, people (including me, I'm that old...) had fun playing games at 320x200.

[mayaman]

To be honest, I played Crysis on the VR920 (at 1024x768 downscaled to 640x480) and it still looked great. It was totally playable.

So did I. *shrug*

Alot of over analyzing lol

[zacherynuk]

had fun playing games at 320x200.

MCGA 320x200 with it's extra on screen colour palette (256 from 260k) was a distinct advantage over VGA (16 colour from same palette) was much better for most games than the higher resolution of VGA. Even things like populous, where detail was important, lost a lot of *something* with less colours. (old man blog)

Interestingly, the HV056WX1-100 is only a 260k 6-bit panel...
More interestingly, http://www.vitrolight.com/ (Is now winstar) and the Vitrolight ebay store seem... dead.

[MaterialDefender]

To be honest, I played Crysis on the VR920 (at 1024x768 downscaled to 640x480) and it still looked great. It was totally playable.

So did I. *shrug*

Alot of over analyzing lol

Be careful what you say.

It only looks that good because people who are making games like Crysis do that kind of 'over analyzing' all day long.

[Fredz]

The pixel grid is still visible in the right image although 1 pixel in the left image theoretically is a quarter pixel in the right image.

Thanks for the correction, I didn't know it worked like this.

But hey, people (including me, I'm that old...) had fun playing games at 320x200.

I'm quite old too, my first computer even had an impressive 256x192 resolution.

Alot of over analyzing lol

Surprinsingly that's not the first time I see comments like this. I'll never understand why some people feel the need to criticize other people when they're simply trying to use their brain, even not in an impressive way.

More interestingly, http://www.vitrolight.com/ (Is now winstar) and the Vitrolight ebay store seem... dead.

Good thing that I bought their 5.6" display some days ago...

[MaterialDefender]

I'm quite old too, my first computer even had an impressive 256x192 resolution.

Then you must be enviably young. My first one had half (or quarter, to be precise) that resolution, and was black and white. I think I was 12 or so when I got that thing.

[Laserschwert]

I played games at 160 x 120 with 16 colors on my C-64 and still had fun

[cybereality]

Well I'm not even that old, and I remember playing games on the C64. Though I'm not sure I'd consider that "good graphics".

[Laserschwert]

Sooner or later Crysis will look dated too.

Right now though I still think no game has surpassed it yet, and that's a good run, considering it came out 5 years ago.

[MaterialDefender]

Well I'm not even that old, and I remember playing games on the C64. Though I'm not sure I'd consider that "good graphics".

Certainly not. Not even in nostalgia mode. But from my experience I would say that the human brain is amazingly flexible when it comes to image resolution. I don't have much experience with VR environments, but I think things like latency and 'image steadiness' (meaning no aliasing flickering and similar temporal effects) are much more important.

[cybereality]

Sooner or later Crysis will look dated too.

Sure, but I feel like the gains will be a lot less that what we've seen going from Pong to Crysis.

[Namielus]

I disagree strongly with a lot of comments here, but have respect for the knowledge people put behind their posts.
Maybe I am getting the tone of this discussion wrong, and I can see how this post might offend some people.
I apologize in advance for stepping on any toes, I do not mean any disrespect.

"using your brain" is not always 100% positive, if it is counter productive. And for the dev-kit discussion, this feels very counter productive to me speaking of the dev-kit resolution.

Is the resolution high enough for what? Giggle and play with a developer kit like its our dream toy from childhood?
Or to do serious development for the future consumer version? You really think the resolution is so bad you cant develop content with future higher resolutions in mind? Good enough for reselling in your online store like its a consumer product?
Good enough to impress your friends? Good enough for pro-gamer sessions?

Please get some perspective. Don't be the first tourist to walk on the moon and complain about the lack of restaurants.
Yes, you can perfectly defend the logic behind how it would be nicer to have a gourmet meal watching earth from a far.
But you are on the _moon_
Stupid analogy maybe, but this is the feeling I get from some posts in this thread.

I've never seen such an argument over something this temporary.
A platform is being built as we get to watch it happen, we are potentially watching history happen.

I just can't believe what I'm reading, those of us who will get to own the early version of this
before it becomes juiced up for the consumer market is really lucky.
Energy and intelligence is wasted on something that's going to be improved without a doubt in maybe less than a year from now. Without even complaining, making suggestions or uttering a single word, the resolution will become better.
Maybe even the dev-kit will get better in that regard, but only if its not getting in the way of reaching developers in time.
There is no doubt Palmer is aware of the benefits of higher res, so what are we achieving?
Everybody knows it.
Whats the goal? Force Palmer to delay the dev-kit and put a
higher price on it so it can become more elite and consumer-like?

You can intellectualize and problematize this cleverly to back your points view, on why this average resolution is so horrible.
But lets focus that critical thinking into suggestions for the final consumer version. The dev-kit is good enough for us to play around with, form ideas both for hardware and software/content.

I am beginning to understand why Palmer had to so explicitly state that its a developer kit not a consumer product.
Not happy? Wait for the consumer version and let others pave the way instead.

And again, even tho I'm ranting and probably saying stupid things, dont take it personally. I respect all of you.
I just really had to vent this.

[mayaman]

Namielus, well said.

[cybereality]

The important thing to remember is that the Rift is a dev-kit. It was never intended as a mass market product (although it is kind of blowing up regardless).

Its supposed to be enough for developers and hobbyists to get excited about, not something that Joe 6-Pack is going to buy his son for Christmas.

[Fredz]

Maybe I am getting the tone of this discussion wrong, and I can see how this post might offend some people.

No offense taken, but I'm not sure you did get what we are saying here and why. And please don't be offended by what I'm saying.

"using your brain" is not always 100% positive, if it is counter productive.

Counter productive to what ? Are we preventing anyone from doing anything by talking about resolution ? Hell no ! That's why I really don't understand mayaman's remark. What we are discussing here has no incidence on his life, so why this need to criticize ? Wouldn't it be simpler to just ignore the thread ?

And for the dev-kit discussion, this feels very counter productive to me speaking of the dev-kit resolution.

To me resolution becomes important if it makes me miss an important aspect of the immersion. I'm still waiting for my Vitrolight 1280x800 5.6" display to experiment with, but on my 1024x600 5.6" display with 5x aspheric lenses I see popping objects in the distance, which is quite annoying and detracts from immersion. Evaluating the good and bad points of the dev kit is not a waste of time in my opinion, it may well have a direct incidence on the way I'm developing for the Rift (stereo driver for now, probably some demos in the future).

Or to do serious development for the future consumer version? You really think the resolution is so bad you cant develop content with future higher resolutions in mind? Good enough for reselling in your online store like its a consumer product?

As I said, knowing the limits of this HMD doesn't prevent me to develop for it, much the contrary in fact. It's just like in any technical field, if you know the limits you can better handle them. We've already had an interesting discussion in the same vein about warping in another thread, and other discussions about the need of positional tracking and ways to do it. To me that's not wasting time at all, and even if it was, it's still my own time, I do what I want with it.

Energy and intelligence is wasted on something that's going to be improved without a doubt in maybe less than a year from now. Without even complaining, making suggestions or uttering a single word, the resolution will become better.

Unfortunately it's quite far from sure that workable and better displays will be available soon. The best thing that could happen would be for the the 6.1" 2560x1600 displays from Toshiba and Sharp to be available soon, but at this time they are not even mass produced, let alone available with LVDS interfaces.

Some time ago I've posted a quite exhaustive list of small size displays, and most of them are either vapourware (announced several years ago and never produced) or can only be connected to SoC, ie. not usable with the Rift. Considering the screen the devkit is using dates back from 2008 and nothing better has come in the last 4 years, I'm not exactly confident a good screen replacement will be available in less than a year from now. And if that's the case, I'd be curious to see at which price.

And by the way I'm not complaining about the current resolution. I'm a technical person, been programming for 25 years now, when I look at a technical object that I want to develop for, I analyze its good and bad points objectively to get the most out of it. I'm not a consumer whining about subpar specs.

[cybereality]

@Fredz: Yes, I also wonder how much higher the resolution could go in a reasonable time-frame. It would be interesting if Oculus could get a custom display made just for the commercial Rift, but that probably costs mega-$$$ so maybe not too likely. However, it seems with smart-phones getting bigger (some almost 5") and tablets getting smaller (now 7" is popular) they are going to have to meet in the middle at some point. So that time will surely come, and the resolution will surely get better. Will it happen in 1 or even 2 years. I don't know. But its going to happen almost surely.

[Fredz]

The problem with displays used in mobile phone and tablets is that they are not usable with the Rift because they don't have adequate interfaces as Palmer said previously. And creating interfaces for such displays is extremely costly as someone else said in one of those threads. John Carmack even offered a bounty for a reverse-engineering of the Sony HMD box, which nobody handled.

The 5.6" Vitrolight could be used due to its LVDS interface because it was produced for PCs (UMPCs). But now that this market is practically dead their won't be such kinds of displays in the future anymore.

Also mobile phone displays are not available in adequate sizes, you would need two of them to be useful and that completely defeats the concept of the Rift, with a lot of added complexity, driving two displays in a synchronized way. We've already had these discussions a lot of time in the last months.

Tablet displays could be a better solution, but for now there is nothing available with a better resolution than the Rift. I'm not sure why constructors would go for more than 1280x800 on 7" displays either (215ppi). The next move on this market clearly seem to be oriented to lower prices (rumor of Nexus 7 at $99), not better resolutions. I may be wrong though.

I know of only two announced displays that could be used in the Rift (more than 5.5" diagonal) and with a better resolution. The first one is the Toshiba/Sharp I talked about, the second is the Hitachi 1600x1200 6.6" display, but this one has been announced to years ago, with no sign of it being available at any point. Doesn't inspire confidence for the 6.1" displays either.

And I don't see the Rift creating enough momentum to provoke mass production of such displays. With 6,000 devkit baked in a month, we are quite far from the sales of mobile phones (5,000,000 in a weekend for the last iPhone). Not sure there can be any economy of scale in this niche.

Anyway I'm not much versed in this field, all I know is from what I've been reading on the Web for several months, and I may very well be completely wrong.

[2EyeGuy]

It's unfortunate that we're dependent on old PDA-size displays between mobile phones and tablets. But what we're getting from these size displays is an amazing improvement over all other HMDs. So I'm still happy.

[crespo80]

And I don't see the Rift creating enough momentum to provoke mass production of such displays. With 6,000 devkit baked in a month, we are quite far from the sales of mobile phones (5,000,000 in a weekend for the last iPhone). Not sure there can be any economy of scale in this niche.

I want to remind that one of the biggest VR flops of all tme, the Nintendo Virtual Boy, with a not even remotely comparable hardware and a mere 20 compatible games, sold nearly 1 million units in one year of commercialization! And it was a flop!
The rift is actually selling a dev kit to developers or enthusiast early adopters with only mouth-of-word advertising, and it's over 10k units sold (counting pre-order kits too): anyone can see the commercial potential in this, even the big names in the game industry, we are not talking about a niche anymore, we are talking about the next level of entertainment!

[Dilip]

It can easily be forseen that if sales of RIFT 1.0(Developer Kits) Can reach to 15~20K Units and after actaul RIFT suppoted titles start coming up in market may be OCULUS TEAM can approch manufacurer like SHARP or TOSHIBA with their CUSTOM REQUIREMENTS for commercial version RIFT 2.0 & seeing future opportunities Manufacurers may too willingly produce the same.

[wuliheron]

Unfortunately it's quite far from sure that workable and better displays will be available soon. The best thing that could happen would be for the the 6.1" 2560x1600 displays from Toshiba and Sharp to be available soon, but at this time they are not even mass produced, let alone available with LVDS interfaces.

Some time ago I've posted a quite exhaustive list of small size displays, and most of them are either vapourware (announced several years ago and never produced) or can only be connected to SoC, ie. not usable with the Rift. Considering the screen the devkit is using dates back from 2008 and nothing better has come in the last 4 years, I'm not exactly confident a good screen replacement will be available in less than a year from now. And if that's the case, I'd be curious to see at which price.

Yeah, there are supply line issues. Sharp is in debt and went to Foxconn with hat in hand looking to strike a deal for the new technology. Foxconn demanded their soul in return so Sharp took out a loan from the bank instead. It's all up to the bean counters now and the real question isn't whether these will be produced, but just how many will be produced how fast if some huge manufacturer like Foxconn doesn't acquire the licence.