Active vs Passive 3D Myth

UPDATE: I revisited this subject because the comments on this post made it clear there was still a lot of confusion on this topic. Read the clearer explanation here.

I’m interested in 3D technology and lately there have been a lot of interesting developments. One of the most promising new 3D technologies for the home is imho the kind of passive 3D technology that LG and others are introducing to their new screens where the screen itself is taking care of delivering the images for the left and right eye to the separate eyes, instead of the glasses. This technology is called passive 3D because the glasses are simple passive polarized lenses, instead of battery powered active shutter glasses.

Active glasses (back) shut out each eye in turn, passive glasses (front) filter the light

But as sometimes seems to happen there is this myth that keeps perpetuating itself that to me seems to be blatantly wrong if you just stop and think about it for a second, but for some reason nobody does that and the myth just gets repeated over and over again. The myth is that:

“Passive 3D offers inferior resolution because the amount of display lines gets halved”

This seems to make sense at first. Only half the pixels is sent to each eye. The other half is reserved for the other eye. So resolution is halved right?

Well yes. Compared to 2D it sure is. But this is always mentioned as a disadvantage of passive 3D when compared to active 3D and I think that is totally misleading. Active 3D *also* halves the resolution, it just does that over a different dimension (time vs space). Active 3D halves the temporal resolution whereas passive 3D halves the spatial resolution. In the end both technologies display 2 frames on the same screen that would show only one frame in 2D mode, so both half the resolution compared to 2D.

Just have a look at the diagram below and I think it will be completely clear that active 3D does not show a single pixel more than passive 3D does. It will also be clear that passive 3D can potentially give a much more stable picture, with less flickering.

Diagram of active vs passive 3D display technologies

For each technology we show 4 frames of 3D animation. The white pixels are showing the image and the black pixels are ‘blocked’ or ‘blacked out’. On the left we see active 3D. Here alternating the left and right eyes are shown a picture and a black frame. On the right we see two ways of doing passive 3D, either alternating horizontal lines or a checkerboard pattern (theoretically columns could be possible too but I haven’t heard of a TV model doing this so I left that out). As you can clearly see with both the passive technologies your left and right eyes combined get exactly the same amount of displayed pixels (as opposed to blocked or blacked out ones) as with active 3D. But the patterns on the right are much more regular than the alternating pattern on the left, which should result in a much more stable, flicker-free picture.

15 responses to “Active vs Passive 3D Myth

  1. -> In the end both technologies display 2 frames on the same screen that would show only one frame in 2D mode,

    Incorrect: In 2D word er maar met 1080p24 24 frames in een seconde weergegeven, waarbij in (actieve) 3D 1080p24 er 48 worden weergegeven, die worden over 2 ogen verdeelt, het enige wat gehalveerd word bij actieve 3d tv’s is de refreshrate, maar daarvoor hebben 3D tv’s over het algemeen (ook afhankelijk van het budget) een hogere, lees dubbele, refreshrate, want dat zijn dus nog 2 verschillende dingen: refreshrate (het aan hz) en het aantal frames per second (fps).

    Refreshrate heeft absoluut niets met resolutie te maken, fps in mijn opinie eigenlijk al niet, maar dat is discutabel en die wordt niet verlaagt, verhoogt eerder zelfs.

  2. Good point about refresh rate vs. framerate. Terminology on this subject can be very confusing.

    I am comparing 3D display technologies here; active vs passive 3D. Ofcourse for a fair comparison I am assuming both the display technology (TV) and the data source (film, game) would have the same characteristics in both cases. So for example a 120Hz active 3D TV showing a 48P film versus a 120Hz passive 3D TV showing that same 48p film. I think the above pictures are acurate in that case.

    If you pick an inferior display technology (say 60 Hz TV) or data source (say 24p film) to compare with, you are comparing apples and oranges.

  3. Ik denk dat Nicolai bedoelt te zeggen dat een individueel frame bij actieve 3d 1 keer voor het linker oog en een keer voor het rechter oog wordt afgespeeld, in dezelfde tijd die voor passieve 3d gebruikt wordt om 1 frame voor links en rechts samen af te spelen. In je figuur zou ik in de linker kolom acht (8) rijen tekenen.

    Of neem dit gedachte-experiment: Neem een 3d film van een schilderij/foto op. Speel beide af met 1000 frames/ seconde, overkill dus.
    Actieve 3 d : Je linker oog ziet het schilderij HD/1080p
    Passief 3 d: Je linker oog ziet het schilderij voor de helft ( hoeveel frames je ook draait, het zal de helft blijven).

    De echte vraag is nu: Wat gebeurt er in je hersenen als je linker oog half HD ziet en je rechter oog de andere helft van de HD? Waarschijnlijk vullen je hersenen het ontbrekende gedeelte aan, waardoor de passieve 3d toch even goed uit ziet als de actieve 3d.
    ( Ik heb al moeite om het verschil te zien tussen 720 en 1080 ).

    Ik ben benieuwd wat je ervan vindt, ik vind het ook moeilijk om te begrijpen wat nu het verschil is tussen de 3ds, maar ik dacht dat dit wel een leuk gedachte experiment is. Ik hoor graag wat je ervan vindt.

  4. In the real world:

    Active 3D uses double frame rate on the TV and as such – you are not losing any resolution.

    Passive 3D uses half the resolution.

    Your scenario is just a theory and does not apply to ANYONE wanting to make a decision between Active nor Passive 3D. How on earth is it misleading to tell people that with Active 3D they get the full resolution and frame rate ? With passive 3D they don’t get the full resolution.

    So for theory crafting you are right. But with your reasoning you can also say that Active 3D is the worst technology ever because the frame rate should be 12 (not 24). Because it’s half what it should be. Good thing it actually doesn’t work like that 🙂

  5. @marcus: That is exactly how I imagine it. And at 240 fps, I think reality is pretty close to what you are saying.

    @Jonas: I don’t get why you assert that Active 3D uses double frame rate on the TV… Double compared to what? Surely passive 3D TV’s can also display at that same frame rate? No matter how you cut it, with active 3D each eye is looking at a black frame half the time. With passive 3D each eye is seeing half the pixels black. It doesn’t matter much in total pixels observed.

    Have a look at this promo video from LG which explains it (it is biased pro passive 3D I admit, but I still think it offers a decent explanation):

    The set LG is talking about here offers 240Hz refresh rate. I don’t see why they would throw half the frames of an 48p source away so I’m assuming that they would just show each frame 5 times before proceeding to the next frame.

    Also this promo video talks about an extra black frame for *both* eyes inserted between each normal frame to prevent crosstalk, but I don’t know if that is actually realistic. If it is it makes the case for active 3D even worse.

  6. @marcus: I missed that you are saying that there should be 8 frames in the left image….

    I strongly disagree with that. How is a 240hz active 3D TV going to be able to show twice as many frames in the same time as a 240hz passive 3D TV? Don’t compare apples and oranges!

  7. As i understand: Movies are filmed at 24fps. Peter Jackson is now filming the Hobbit at 48fps . In the past there were some experiments with 60fps movies, but they didn’t catch one ( celluloid is expensive at that rate).
    Problem with 24fps is that in fast action scenes the frames will “blurr” and if you pann the camera from left to right, you get a “strobe or judder ” effect. We have all seen movies in the cinema with these artifacts.
    What if you would multiply each frame 10 times and with some clever algoritme adjust each of those 10 frames to look increasingly more like the following frame? In that case blurr and strobe would diminish. So ideally , a movie playing at 240 Hz (24 original frames * 10 duplicates ) would look better. But if you have a blurred frame at 24fps and you multiply this 10 times, it wont get any sharper. Garbage in, garbage out. That is why mr. Jackson shoots at 48fps, that is why i have turned off the 240HZ feature on my TV set ( it makes a movie looks artificial, i prefer the blurring over the plastic). And that is why a 3D movie shot in 720 at 48fps looks better than shot in HD with 24 fps ( active or passive) . So forget the 240Hz marketing ( unless you can make 10 sharp pictures out of 1 blurred one) gimmick and lets talk about real 24fps.
    Active 3D has to double the framerate to “alternate serve” each eye a frame, so that each eye will see 24 fps.
    Passive 3D serves both eyes at the same time, so it doesnt have to double the framerate. But, because it is passive, it halves the resolution. That is the point @Jonas is making.
    So i would draw in your figure 8 frames in the left column and 4 in the middle and right.
    Or better, 48 left and 24 mid and right.

    My point still would be: half the resolution viewed with both eyes, i think the brain will fill in the other lines for some amount. So dont get hang up over the half resolution thing.
    And turn off the 240Hz gimmick on your TVset . And turn up the brightness, because that is a real bummer with 3d. Saw TinTin yesterday and for the first time really liked the 3d , although i had passive glasses;) Thank you mr Spielberg!

  8. I think you are confusing two things. On the one hand there is the refresh rate of the TV, which could be 50/60hz (for very old sets, depending on europa or us / japan regions), 100/120hz or more recently 200/240 hz and the frame rate of the film, usually 24fps, but this seems to be moving at 48 or even 60fps. But still most films are 24fps.

    On the other hand there is interpolation. When the TV has a refresh rate of 240 hz it will generally just show each frame from a 24p film 10 times in a row. Now theoretically it could be better (or plastic) depending on your taste, to interpolate the missing images in between the movie frames.

    When you say you turned 240hz off, I think you actually mean that you turned the interpolation off. I do not know a single television set, definitely not a modern, 3D-capable one, that has a refresh rate below 60 hz. Most are up there in the 120 – 240 hz range.

    Now here is my question: What difference does it make whether the film was shot at 24p or at 48p for the whole active v.s passive 3D story? Or whether the TV sets are running at 120 or 240 hz? Nothing.

    When we look at a 24p 3D film, it is actually 2*24P. There are 2 times 24 frames per second, one for the left and one for the right eye. Whether these are shown at 120 or at 240 hz really doesn’t matter much as in both cases the original movie frames are just repeated multiple times (either 5 or 10 times) before the set switches over to the next movie frame.

    Now why would a passive 3D TV be so stupid to throw away half of the movie frames? They are there on the disc arent’t they? It wouldn’t be so stupid of course. And why would it not be just as possible to make a 240 hz passive 3D set as it is to make a 240 hz active 3D set? It wouldn’t be any more difficult.

    In the end, given the same source material and the same refresh rate, the only difference is in how the screen is used to display the images to each eye. So for a 24fps film, a 240 hz TV set will show ten equal frames in a row before switching to the next. The picture in the blog post just shows the first 4 of these ten. I still see no reason why the active set would show 8 in the same time the passive set is showing 4. It would need to have a twice as high refresh rate, 480 hz instead of 240hz. That, in my book, is comparing apples and oranges.

  9. Pingback: Active/Passive 3D Myth Revisited | Stijn de Witt's Blog

  10. You’re right, the HZ thing is confusing things.
    Lets talk about frames and go to a movie in your local cinema. The speed of the projector is set at 24 frames per second. If you put on your passive REAL3d glasses, you can enjoy the 3d movie as intended. You see 24 frames each second , each frame at half the resolution in your left eye and another half resolution picture in your right eye. Your brain will combine these different and overlapping half resolution pictures to give you a 3d mental image.
    How much resolution would that mental image be? Hard to tell, because the pixels in the right eye are different pixels then in your left eye: they are seperated the width of a line apart (the stripes in your middel column). So it could be that half+ half=half resolution , half +half=full resolution our somewhere in-between or perhaps even half+half = less resolution*. It is like your apples and oranges, hard to compare. I would ask a neurologist (?) if they have a clue. *

    Now lets put on some active glasses. To get 24 frames in the left eye, we have to crank up the speed of the projector till it turns at 48 frames per second. Then we have to intertwine 24 black frames in the filmroll each second otherwise the film would be over in half the time it would normally run and we could not buy some popcorn in the break.
    Now each eye sees each frame at full resolution and a complete black frame as well as bonus. Poor brain, instead of 2 different, half-resolutioned pictures, it gets 4 pictures, 2 (darkened) full-HD and 2 pitch black . Can it make a full HD picture out of these 4 frames? Perhaps, perhaps not, again we would have to ask the neurologist.

    So, my point: active vs passive : both are a myth because in both cases it is not clear how the brain copes with the information you provide it with. Feed it two different half HD images or 4 images, 2 of them being black? Not so easy to tell, would i guess.

    * To make my point that the left eye is seeing other pixels then the right eye:
    Lets take the checker box glasses ( the right column in your figure) . Lets take a picture of a checker/chess board. The left eye would see a complete white board because the black tiles would all be filtered out by the left glass. The right eye would see a black board, because all the white tiles would be filtered out. So i think that in this case the brain would see an big grey 3d mess where once was a fine chessboard.

    ** a simple experiment would be to walk a day around the house with the passive 3d glasses on . Can you see as good 3d with the glasses on as off. That would be an indication as how your brain can combine two slightly different images.

  11. “So, my point: active vs passive : both are a myth because in both cases it is not clear how the brain copes with the information you provide it with. ”

    I totally agree with you there. This is the reason I wrote this article. I see the whole ‘halved resolution’ argument returning time and time again as a negative of passive 3D but I think both techniques half the resolution in some dimension and indeed you would need a neurologist to determine which is better.

    “Lets take a picture of a checker/chess board. The left eye would see a complete white board [..] the brain would see an big grey 3d mess where once was a fine chessboard.”

    I can’t agree with this one, especially the last sentence: “where once was a fine chessboard.”, assuming you are talking about 2D there. If you would take a normal 2D Full HD image and paint a checkerboard pattern like you describe, you wouldn’t see a checkerboard. You would see a large grey mess. The pixels are (at normal viewing distance) much too small to still see the pattern. You will see a grey area in 2D as well as in 3D.

    Now for the first part, the left eye seeing a completely white board, that would be disconcerting even if the mixed 3D image would be a grey mess. But there is a technique to make sure that the left eye gets to see all the pixels, the white ones as well as the black. It’s called 3D interlacing and I describe how it works in my follow-up post:

    Active vs Passive 3D Myth Revisited

  12. I can see that the author is somewhat confused here. There really is no ‘myth’ to dispel. Keep in mind that “HD” is 720P or above. When it comes to HD3D we have two popular formats: 720P-60Hz 3D and 1080P-24Hz 3D. Both formats are delivering TWO images (L+R) per every 1Hz over HDMI.

    Now lets marry the 720P-60Hz 3D to both an ACTIVE and a PASSIVE 720P HDTV.

    The ACTIVE TV will have to operate at LEAST 120Hz (many may operate at higher frequency but lets not confuse the issue and disregard this) to display the TWO 720P frames per eye @ 60Hz.

    The PASSIVE TV will not have to operate at 120Hz because 1/2 the Right and Left eye images are combined and displayed at 60Hz. This is far cheaper (and physically possible) then attempting to make a 120Hz LCD display. These displays will REMOVE 1/2 the vertical resolution (480P) to deliver RIGHT EYE (even numbered scan lines) and LEFT EYE (odd numbered scan lines). This is a PHYSICAL OPTICAL restriction of the 720P passive HDTV as only 1/2 the scan lines (480P) can be viewed in the LEFT eye and the other 1/2 of the scan lines can be viewed in the RIGHT eye.

    A passive 3DTV would REQUIRE a native resolution of 1440P to be able to display the full 720P content for both eyes.

    However, RESOLUTION is NOT the #1 drawback of passive systems. Viewing angle and ‘ghosting’ are.

    Unlike Plasma and projection systems, LCD displays suffer from limited viewing angles. The contrast and color accuracy will be different depending on the viewing angle. While great strides have been made in LCD technology the viewing angle(s) gets LIMITED FURTHER when polarizing filters are placed in front of the TV screen – a requirement for passive 3D viewing. If your eyes are level with and at a perfect right angle to the screen AND placed approx ‘mid point’ screen height – the 3D ‘might’ look fairly decent. However move away from THAT ‘sweet spot’ and you’ll have ghosting.

    Therefore, accuracy of color, contrast, black level AND ghosting are the significant drawbacks of LCD or passive systems and NOT resolution.

  13. Pingback: Active/Passive 3D Myth Revisited | Stijn de Witt's Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s