I have said for a long time that the math is far from all that counts... and when arguing things like "1080p per eye at a time" vs "540 to each eye at all times" you have to evaluate what those things mean in terms of what you see and what' going on with them.
Some important concepts:
- The biggest one: 3D imformation is not like 2D because there is different info for each eye for the same point in space. In 2D every pixel represents on precise point in the subject being displayed. In 3D each pixel may represent 2 distinct points. This means a 3D 1080p image has up to (and I argue technically always) twice as much image data as a 1080p 2D image. Because even if both eyes happen to see the exact same thing, it's still unique data that happens to be the same pixel information at that precise moment.
So even though passive 3D does show you 1080p pixels of data at once, a true 3D 1080p image is TWICE that many pixels of information (although realiistically many pixels will happen to be very similar data). This becomes much eaiser to understand when you think head mounted display with 2 1080p displays. Clearly there is twice as many pixels as a 2D 1080p image.
- Another big one: Active only shows to one eye at a time so it's only showing 1080p pixels at any given time, so it doesn't ever show twice as many pixels as active right? Wrong.
A film or TV never displays an actual moving image, but we still percieve a moving image... why? Image persistence.
While yes technically there is never more than 1080p pixels of image adata on the screen at any given moment, image persistence allows for you to effectively experience twice the number of pixels just like phosphor decay time allows a CRT to look like more than 1 pixel being drawn at a time.
So active 1080p is giving your more pixel information data than passive, and while it's not apples to apples, it's also not apples to elephants.
In other words image persistance makes active 1080p functionally 2 1080p frames of data even if it's not quite as nice as it would be to actaully have 2 1080p frames simultaneously.
- Which ties into resolution - spatial vs temporal. May times people argue that because active shows only one frame of 1080p at a time and passive shows one frame of 1080p at a time, if you add the number of pixels per second shown, they are the same number.
This conflates spatial (detail) resolution with temporal (motion) detail. The idea that because active shows 1080p for 1/120 second then 1080p for 1/120 second alternating, while passive shows 1080 for the full 1/60 second. Means the same resolution (add the pixels per 1/120 second and you ge the same number) is like saying 120hz video content is twice the resoultion of 60hz (or that 48hz is twice that of 24 hz).
In terms of pixels per second displayed, yes it is. But the way the term resolution is used most commonly (image detail) does not lend itself to that math. When you play a video game at 30fps and 60fps you do not go "ooh that 60fps experience was much higher resolution!".
Also think about this... if total pixels show per 1/60 second was resolution then I could take a display with 4 only 4 pixels (each 1/4 of the screen) and if I refreshed them fast enough, that would be the same resolution as 1080p refhresed at 60hz. Does that sound like it might possibly be true? Doesn't does it...
- Then there is the idea that resolution is just counting pixels. This is completely flawed... if it were true the terms screen door effect, scan lines and sharpness would never come into play.
Truth is pixel layout is as important if not more than number of pixels in many cases and why passive displays have to account for something very important: The HUGE image artifact of a pixel high black line between every set of pixels.
A display is EXPECTED to have roughly the same pixel spacing in all directions between pixels. When that is not true you care not comparing apples to apples anymore...
Passive displays do not show 540p to each eye, they show 1080p with MASSIVE (50%!!!!) image artifact.
Think about this, if I show you a 320x240 picture on a 640x480 display with line doubling (to stretch out the pixels to fit) it will look ok, a little blocky probably or soft.
Now let's say I take a black fine tipped sharpie and draw a bunch of black lines across your screen over every other row of pixels. That DEFINITELY doesn't look the same does it?
Just counting dots isn't good enough, how they are laid out counts.
- Which leads into horizontal and vertical resolution... active is 1080p and passive is 540p right?
No... if you are going to just through out terms like that you are going to have to expect to be in the same realm and 540p is not 1080p missing every other row... 540p is half as tall AND half as WIDE as 1080p.
Passive does not show each eye 540p or we would have to be cutting out half the pixels in each row also. Again passive shows each eye 1080p with massive (every other line) artifacts.
The fact is each eye only gets 540 lines of vertical image data resolution, but they still get the full 1920 lines of horizontal.
This is a HUGE disparity in the amount of detail that can be resolved from either frame vertically or horizontally. What does this mean for image quality? That's way more complex, but it's important to keep this in mind when thinking about what's happening.
- Lastly there is the whole idea of brain fusion where one eye sees some and adds it to the other eye for a total higher quality picture. This works in theory if both eyes are seeing the same data set, just differnt parts of them.
But remember in 3D your eyes are seeing DIFFERENT data sets so even when you see two 1080p frames they are STILL adding them together to get the 3D Data... remember this is a third dimension now where the information from both frames isn't just to produce a 2D result, it's a 3D result which requires more data.
You aren't just figuring out what you see on the xy plane (which requires only one pixel of data) you are also figuring out on the z plane which requires another pixel of data.
Think about this... let's say instead of breaking the image down into alternating rows of pixles, I just show your left eye the left half of the image and your right eye the right half. Will it fuse that into one image? No... it will be some weird semi flickery painful mess.
Also try this, go into MS paint or something and make your canvas full screen, paint the left half red and the right half blue. Get your eyes close eough so each eye see's only one side of the screen.
You see purple right!
No... you see some messy odd combination of red and blue.
Brain fusion comes from a test which originally used very small text that would not be readable at super low resolution to prove seeing both sets of image data gave you a full image.
The big flaws in this test is one:
1: The image data was 2D - this proves nothing about your brains ability to percieve full 3D data when some is missing
2: This proves you can recognize something via fusion, not recreate missing data. There is a difference between being able to tell what it is and seeing it in the same quality. You can poke a LOT of holes in the page of a book, and I can probably still make out what the words say despite large quantites of the letters being poked out... but it is NOT the same as seeing the page unmutiliated.
Here's a little experiment to pull off: Look out the window and close one eye now in front of your open eye rapidly wave your index finger. You will notice that you do not see a finger in the way, but rather just a slight blurring and darkening. But you still see everything you used to, you can still see the cars driving by and if there is a bird on a fencepost or some other small detail, you can still see it no matter where it is.
Now hold a comb up in front of that eye. Don't move the comb relative to your eye. Notice the static artifact. This means that there are times when you cannot see something. If a bird is on a fencepost but that fencepost is occluded by the comb, you never see it at all!
But you say "but your other eye would see it!" No, it would see ITS OWN VERSION of it... but it would not see what your current eye would be seeing which is where 3D data comes from... two views of the same thing.
Hopefully it's clear which of those represents active vs passive
That is all to show that the math is not the whole story...
BTW as for 120hz flicker and headaches? I am not so sure of that... it may effect some people (the same ones who get headaches from floursecent lights etc) but I would have to imagine if 120hz flicker was that bad, you would HATE seeing film based movies... even with triple black frame insertion you are getting slower flashes than 120hz... also what would the strobing backlight in many of the newer LCD TVs do to you?
Last note: Crosstalk and ghosting - these are not indicative of active vs passive, they are indicative of the display type and quality.
Active has a rep for crosstalk but that tends to be because it's often viewed on LCD displays with poor decay rates or improperly synced glasses.
The fact DLP uses active and has pretty much zero crosstalk is a testiment to the fact it's not the active that's the problem, it's the display tech that is.
Also while passive is a little more sure in it's technology, it's not without flaw either. Of course passive home displays require a sweet spot middle of vertical viewing distance but also I have never found a perfectly sweet spot with mine. On high contrast areas some part of the screen always ghosts.Edited by Devedander - 5/24/13 at 11:24am