Page 1 of 2 12 LastLast
Results 1 to 10 of 17

Thread: technical question regarding retro reflective screen Gain

  1. #1
    Join Date
    Jun 2019
    Posts
    3

    Default technical question regarding retro reflective screen Gain

    I was wondering how much Gain you would get if the video projection beam was coming out towards a retro reflective screen nearly exactly from your eye position.

    In my own home theatre the projector is on a tripod rather than mounted on a wall and the screen is a 120'' retroreflective one. I've always noticed when I crouch to grab something from the floor (the beer), the image gets so bright when my eye is amost in the same angle of incidence as the projection lens that the light dazzles me.

    But the question of just how much Gain you get at that angle come to my head when I saw this:

    https://www.youtube.com/watch?v=XPOgLt_73cY

    To summarize the video it's a head-worn, very compact dual projector which directs the projection beam with special optics (a reflective polarizer and QWP) to the screen and back to your eyes. So I'd guess you could expect very bright images even if the source LED wasn't very bright in this. Is 5.0 gain a reasonable estimate?

  2. #2
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    I hate that they use the term "gain" to describe how bright a screen is. They should be talking about reflectivity. Grrrr.

    To further complicate things, the "gain" that most screen manufacturers quote is actually a comparison to a standard coating, like titanium dioxide, and not a measurement of the actual percentage of light that is reflected. (Thus it's a measurement of how far off from the reference titanium dioxide coating your test screen is.) So you can end up with gain numbers greater than one, which would imply that the screen is actually generating light - obviously violating the laws of physics!

    With all that said, high-end screens (like the glass-bead-coated ones you see in movie theaters) can reach a gain of 2.5 compared to titanium dioxide, at least according to Wikipedia. (Note that this is measured perpendicular to the surface.) Based on that, a gain of 5 is probably not realistic.

    Are you certain about the operation method? Because in the pictures the glasses appear to be nothing more than passive polarization filters - possibly with some minimal head-tracking hardware. I definitely don't see any embedded projectors in the pictures though. Also, the glasses are supposed to be wireless, so if there are embedded projectors, how are they powered? (No room for batteries either.) And don't forget that they'd also have to stream *two* video feeds via a wireless data link and decode it on the fly. Seems like a lot of hardware to fit into the admittedly stylish (but very small) frames.

    Then too, if the projectors were in the glasses, you'd be constantly adjusting the geometric correction to get the "board" to perfectly overlay with the video image from the glasses. That's going to take some serious computation, and this is supposedly all running off a cell phone? Doubt it...

    I'm guessing that it's actually designed to use an overhead projector running at 120 hz with a polarizing filter that swaps in and out every other frame to give you left-eye and right-eye fields. (Or a DLP-based solution to flip the polarity - result is the same.) Basically the same technology behind the passive 3D televisions from 10 years ago that you can't find anywhere anymore.

    And this is assuming that it all actually exists and the video isn't just a teaser, of course. (Suspiciously, they want you to sign up to be notified when these will be available and all the news sites have the same teaser video with no further information on the system, so it could just be vaporware at this point.)

    But at the very least, passive 3D displays that react to head-tracking for a single user are a solved problem, so I don't see anything about this product that is impossible. (Heck, they did it 12 years ago using a 2D display, and we demo'ed this at FLEM in Orlando that year...)

    Adam

  3. #3
    Join Date
    Jun 2019
    Posts
    3

    Default

    The glasses are PHMDs (projective head mounted display) also known as HMPDs (head-mounted projective display).
    These have existed since the 80s, but shrinking of reflective microdisplays into the sub-inch sizes and development of high power monochrome LEDs suitable for pico video projection has made it possible to make them much more affordably and in a much smaller form factor.

    There are two reflective LCoS panels in this for each eye of the user. Someone had made a tear down video of one. DLP is too expensive in comparison. Additonally an LCD shutter layer is used on the glasses to reduce crosstalk between eyes.
    The device was known before as CastAR and there were many 3rd party reviews of it. https://www.youtube.com/results?search_query=castar+ar

    Since the beam is projected and reflected from a reflective polarizer towards the screen and then back through a wave plate to pass through the same polarizer, the beam is basically retro-reflecting almost exactly perfectly towards the eyes. This is not possible with a standard projector as the eyes are at an offset relative to the projector lens.

    I don't know how much "Gain" this provides and was hoping to get an idea from others. A retro reflective screen has a very high gain inside a cone of a narrow angle, but inside the cone itself the brightness is not uniform based on the angle as well. Who knows how bright you can get if you beam the video almost exactly towards the eyes as in here unlike nearly exactly (a 5-10 degree deviation) like in some movie theater screens.

  4. #4
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by taxayy View Post
    The glasses are PHMDs (projective head mounted display) also known as HMPDs (head-mounted projective display).
    How are you determining this? There is no information on the website, nor in any of the new articles I've found. (If you have links, please share...)

    Given that they are wireless, I'm very curious (and more than a little skeptical) about how they're handling the power delivery, data link, and geometric correction issues in such a compact and seemingly lightweight package. Although I will admit that if it's a real product and it has decent battery life it could be a very attractive unit with uses far beyond simple tabletop gaming applications.

    There are two reflective LCoS panels in this for each eye of the user. Someone had made a tear down video of one.
    How could someone have done a tear-down video on a product that isn't commercially available yet? Or are you referring to the failed CastAR product (which was *not* wireless)?

    DLP is too expensive in comparison.
    Compared to what? We don't have a price on these things. Are you the developer?

    Since the beam is projected and reflected from a reflective polarizer towards the screen and then back through a wave plate to pass through the same polarizer, the beam is basically retro-reflecting almost exactly perfectly towards the eyes.
    I think what you meant to say is that the light viewed *by the wearer* is essentially retro-reflected from the screen, and the polarization is also preserved which is needed to maintain the 3D effect. However, light is also scattered in other directions by the screen. Note that they don't make any claims about brightness compared to viewing angle.

    I suspect that the screen on the table (the game board) is probably something very similar to the venerable glass bead movie screens used in theaters and not an absolute retro-reflector. (Allowing for the fact that a glass bead screen does have some retro-reflective properties, and this can be adjusted by varying the bead size.) And that would lead you back to the published maximum gain spec of ~ 2.0 to 2.5 I mentioned above.

    Note also that the trade-off with glass bead screens is thus: More gain = narrower viewing angle (before you start to notice the brightness change). Given that this product would benefit from the widest possible viewing angle (while still preserving polarization), I suspect the gain is probably a lot lower than the maximum for glass bead screens. So probably something closer to 1.5.

    This is not possible with a standard projector as the eyes are at an offset relative to the projector lens.
    On the contrary, it is possible. Ever been to a 3D movie? Ever notice that the effect works no matter where you sit in the theater, even though your viewing angle is offset from the projection angle?

    I don't know how much "Gain" this provides and was hoping to get an idea from others.
    Again, the upper limit is probably somewhere in the range of 2.0 to 2.5.

    A retro reflective screen has a very high gain inside a cone of a narrow angle
    A true retro-reflector would not allow the wide range of viewing angles displayed in the video. It would also greatly increase the cost of the screen. Until someone can test the actual screen to provide hard data, I'm going to assume they haven't invented a revolutionary new display screen material and are in fact using an industry standard glass bead screen material.

    Who knows how bright you can get if you beam the video almost exactly towards the eyes
    Well, you can calculate it, if you're super curious, although it's not a very useful number. If you use a first-surface mirror, you can get ~95% of the incident light to come back. Slightly more even, if you can spend big bucks on said mirror. As to how that 95% reflectivity corresponds to gain, that depends on what reference standard you're using for gain, but it's safe to say that it would be much higher than the gain from a glass bead screen.

    But what's the point? A mirror is essentially the worst case with regard to viewing angle. A perfect mirror only has one sight line to the projector, so it would be even worse than a retro-reflector in that respect. This is why glass bead screens are so common. They are a good compromise between a white painted wall (poor gain but very even brightness) and a mirror (near unity reflection, but only at a single angle so brightness is very uneven). At the peak gain (around 2.5) for a glass bead screen, you're dealing with just 10 to 15 degrees of usable viewing angle before the brightness changes are unacceptable. That's not wide enough for this product, so shooting for a gain higher than 2 is not practical.

    Adam

    PS: How about an introduction and maybe some background on what you're working on? Seems odd to just join a brand new forum and immediately start making technical assumptions, doesn't it? Or have you been here before under another name, perhaps?

  5. #5
    Join Date
    Jun 2019
    Posts
    3

    Default

    Why are you being a dipshit and acting as an interrogator? If you want to have a discussion that's not how you behave. I will respond to your questions but if you keep acting like this it may be the last time.

    Quote Originally Posted by buffo View Post
    How are you determining this? There is no information on the website, nor in any of the new articles I've found. (If you have links, please share...)
    Sure:
    5th google result, https://www.roadtovr.com/tilt-five-a...g-board-games/
    "Tilt Five appears to be the rebirth of CastAR (which itself was initially spun out of Valve) which reportedly shut down in 2017 before it was able to get its AR system to market."

    The article explains how the device works too. if you search for CastAR you can find many articles and video interviews too.
    The CEO of both companies (TiltFive and CastAR) is the same person too, that could have given you a hint as well.

    For an official confirmation there's also a twitter comment I can dig up.
    EDIT: here - https://twitter.com/tiltfive/status/1134510351767097344

    Given that they are wireless,
    I never said they are wireless, I merely said PHMDs don't use an external projector and the projector is on the headset itself.

    I'm very curious (and more than a little skeptical) about how they're handling the power delivery, data link, and geometric correction issues in such a compact and seemingly lightweight package.
    Google for "PHMDs" or "HMPDs" and you'll find dozens of articles and patents explaining how these devices work.
    Or just type "CastAR" on Youtube: https://www.youtube.com/watch?v=3dltQ4-LYMk
    https://youtu.be/zNWP4WOFvRs?t=470


    Although I will admit that if it's a real product and it has decent battery life it could be a very attractive unit with uses far beyond simple tabletop gaming applications.
    Definitely, https://www.youtube.com/watch?v=7BEL3qy7BuY

    How could someone have done a tear-down video on a product that isn't commercially available yet? Or are you referring to the failed CastAR product (which was *not* wireless)?
    Devices which are going to have 3rd party developers making content for them such as gaming consoles often have the "devkits" shipped to the devs before the commercial launch date so there will be content available for them on launch date... https://www.youtube.com/watch?v=zNWP4WOFvRs

    Compared to what? We don't have a price on these things. Are you the developer?
    No, I'm not Jeri Ellsworth. I also want to again stress out that this is not an interrogation. Also, if I were Jeri I'd have many other places to shamelessly promote my product than a small forum.
    DLP chipset prices have always been public, it's on their website. There are 5-6 manufacturers of LCoS panels you can find online and get bulk pricing via email. Not that hard to do.

    I think what you meant to say is that the light viewed *by the wearer* is essentially retro-reflected from the screen, and the polarization is also preserved which is needed to maintain the 3D effect.
    No, the 3D effect on this is preserved with LCD shutters for each eye to reduce crosstalk from passive 3D. The light is polarized because LCoS panels cannot produce unpolarized projection beams. The TiltFive headset compared to the old CastAR design uses tilted (I assume this is where the name comes from) reflective linear polarizers and a quarter wave plate to first reflect the beam shining from above (as seen in the video I linked to) towards a tilted reflective polarizer. The beam then gets reflected towards the screen while being perpendicular to your eyeball, has its polarization rotated with a quarter wave plate, reflects from the screen and back, rotates from the same quarter wave plate and thus passes this time through the same tilted reflective polarizer. Doing so instead of shining the beam from above the eyes provides more Gain and is more efficient. This last part was confirmed in another twitter comment I can dig up.

    Note also that the trade-off with glass bead screens is thus: More gain = narrower viewing angle (before you start to notice the brightness change).
    Narrower viewing angle means if you offset from the projector you get out of the center of the reflected light cone. With this the projector is strapped to your head and so you are always on the center hotspot.

    Ever been to a 3D movie? Ever notice that the effect works no matter where you sit in the theater, even though your viewing angle is offset from the projection angle?
    Depends whether it is active 3d or passive 3d. Active 3d doesn't even need retro-reflective screens.
    But you are talking about preserving polarization, not preserving gain. Retro-reflective screens preserve polarization even on the edge of their light cone. Two separate things.

    On the contrary, it is possible.
    You are not actually reading what you are responding to. Again, the beam is basically retro-reflecting almost exactly perfectly towards the eyes. This is not possible with a standard projector. It is not possible because here, thanks to the reflective polarizer and how the projector lens is oriented relative to your eyes, your eyes are almost dead center of the reflected light cone, not at an offset like in a movie theatre. In a movie theatre the only thing dead center to the light cone is the projector's aperture.

    I suspect that the screen on the table (the game board) is probably something very similar to the venerable glass bead movie screens used in theaters and not an absolute retro-reflector. (Allowing for the fact that a glass bead screen does have some retro-reflective properties, and this can be adjusted by varying the bead size.) And that would lead you back to the published maximum gain spec of ~ 2.0 to 2.5 I mentioned above.
    Yes, that is pretty much what the screen is. But as I explained the beam is reflected with a tilted polarizer and directed back to the eye at almost the exact offset of the eyeball. So your eye is almost dead center of the narrow light cone from the retro reflective projection screen, not at a slight offset like with video projectors. This provides even more perceived brightness from a retro-reflective screen. That's the whole point of this topic. I'm curious just how much it would be. I have not received my own devkit yet.
    Last edited by taxayy; 06-21-2019 at 14:28.

  6. #6
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Exclamation cypher0, Joec, taxayy, piydadorto, and shoujin walk into a bar...

    Quote Originally Posted by taxayy View Post
    Why are you being a dipshit and acting as an interrogator?
    Why are you resorting to defensive behavior and name calling on your 3rd post to the forum? Is that what passes for a discussion in your worldview?

    Nothing in my post was an interrogation. That's you projecting. And still no introduction! Yet you presume to instruct me on social discourse.?.

    I will respond to your questions but if you keep acting like this it may be the last time.
    You can't be serious. This sounds like middle school courtyard drama! (Oh no! Whatever shall we do if another socially-inept sock-puppet who spent 5 minutes googling something to make him an expert gets upset when people counter his claims and ask for more information? Imagine how horrible it would be if they decided to leave the forum without first gracing us with a reply!)

    Seriously: If you've got it all figured out already then you don't need anyone here on the forum, do you? So why are you here then?

    Adam

  7. #7
    Join Date
    Jun 2009
    Location
    Mesa, AZ
    Posts
    1,277

    Default

    Laser-based picoprojectors could keep the image focused on randomly placed screens in the environment, giving this technique a shot. Tiny LCoS displays, going out of favor because of OLED, make it easier to get nice images in them. I work with those from Compound Photonics, having the smallest pixels. PM me for more information about CP displays if you're interested.

    I recall Jim Ferguson proposing retroreflecting screens with head mounted projectors in the '80s. The screen gain would be very high, maybe too high for good contrast because it has to also be highly polarization preserving. Especially if 3D content is desired. Good luck with that. This is probably one of those ideas that look great on paper but fall short in practical application. May be good for special uses though. Like secure viewing.

  8. #8
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Quote Originally Posted by Eidetic View Post
    Laser-based picoprojectors could keep the image focused on randomly placed screens in the environment, giving this technique a shot.
    Someone posted a while back about a small inflatable dome that you could use in your living room (seriously, like 2.5 meters - maybe smaller?), and if I remember correctly the use case involved a head-mounted pair of MEMS-projectors fed by a laser source. Coupled with the head-tracking 3D glasses (can't remember now if they were active or passive), it gave you a 3D view wherever you looked on the dome. And yeah, perfect focus everywhere. The headset was much less confining than the over-the-face googles like the Rift or Vive and offered a wider field of view to boot. Alas, brightness was an issue.

    Thinking about the parent post some more, for this specific application (augmented reality for table-top gaming) I don't see a true stereoscopic 3D display being viable for more than a single user. If you have 4 people sitting around the table each user needs a render of the map from his or her own perspective relative to their position at the table. But you can't overlay those projections on the playing surface / screen as they will all blur together. Even if you used an overhead projector the 3D effect will still only work for people sitting on one side of the table.

    Given that simulated 3D overhead augmented reality is already a thing (and relatively affordable), the fact that it hasn't been more widely adopted for tabletop gaming suggests to me that a true stereoscopic 3D display for tabletop gaming might not be the killer app I initially thought it might be. (Assuming this actually makes it to market, that is.)

    The screen gain would be very high, maybe too high for good contrast because it has to also be highly polarization preserving. Especially if 3D content is desired.
    You can see this in the 3D cinema. Everything looks dim and gray, at least compared to the 2D version of the same film. Price you pay for passive 3D... Sigh. (That's with a screen gain of 2 to 2.5)

    This is probably one of those ideas that look great on paper but fall short in practical application.
    Agreed.

    May be good for special uses though. Like secure viewing.
    For a single user, I agree. (Or for multiple users if they "take turns" viewing the map. But that seems like a kluge.)

    Adam

  9. #9
    Join Date
    May 2007
    Location
    1 hr from everything in SoCal
    Posts
    2,753

    Default

    Unfortunately with "3D" there is really very little movement forward. It seems the industry is stuck at stereoscopic displays. That's the only game in town right now and has been for a long time. The only development is just perfecting was has already been the industry standard for decades. Until something new comes along, VR is going to remain a novelty. It's fun to play with for a time. Hell, I even own an Oculus but the novelty wears off after a bit.

    Every time I see these projects come up, I always wonder about the small niche groups that that they will reach. Something needs to change or VR is always going to remain where its at now... just shy of mainstream with no real committed patrons.

  10. #10
    Join Date
    Aug 2008
    Location
    UK
    Posts
    5,704

    Default

    Quote Originally Posted by buffo View Post

    Given that simulated 3D overhead augmented reality is already a thing (and relatively affordable), the fact that it hasn't been more widely adopted for tabletop gaming suggests to me that a true stereoscopic 3D display for tabletop gaming might not be the killer app I initially thought it might be. (Assuming this actually makes it to market, that is.)

    Adam
    That's the bit I'm struggling with and wonder if it's why the original company folded. Why do you need a mat and a projector when you can simply project the whole game into a pair of VR glasses such as Pimax or Occulus?

    Maybe this is an attempt to produce a cheaper version of VR by using AR and a cheap projector and mat instead of more expensive VR goggles. However, for the reasons Buffo has pointed out, I can't see this succeeding. The viewing angle is going to very much affect the brightness as huge amounts of light will be lost. If you bounce light off a mirror at 45 degress, most of it exits the mirror at 45 degrees in the opposite direction and is lost. Yes a glass bead mat might increase the amount reflected back by virtue of the 3D profile of the beads. However, I'd speculate a large proportion of the light will still reflect off in other directions and be lost, resulting in a dim image that fluctuates in brightness according to the angle of incidence. Therein, also lies a further problem - imagine 4 players sat around a table projecting the game. The light lost from the projection by the other players, especially the guy sat opposite is surely going to be reflected into your vision and cause some kind of flare or ghosting, unless they have done something trick with the polarisation timings so other players projections are either on the opposite polarity or fall into the brief switching period between polarities where no light is let through. That said, surely that would just introduce lag, even if it could be achieved.

    Aligning with my 1st comment, it seems to me this is an attempt to reinvent the wheel, possibly in the aim of bringing a lower cost solution. However, the issue with reinventing the wheel, is you need to ensure it still comes out round.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •