AVS › AVS Forum › Display Devices › Flat Panel General & New FP Tech › Are panels that are wider than 16:9 a sustainable trend?
New Posts  All Forums:Forum Nav:

Are panels that are wider than 16:9 a sustainable trend? - Page 2

post #31 of 39
1/14/13 at 9:01pm
Quote:
Originally Posted by Chronoptimist View Post

It really comes down to whether you watch TV or Film content. With the exception of maybe one or two shows on Blu-ray, I don't watch TV at all.


I guess you're not much of a sports fan. biggrin.gif


Quote:
Originally Posted by Chronoptimist View Post

The same thing could be said about 16:9 displays and HD rather than 4K.



As far as today's market is concerned, I would have to disagree with that analogy. However, when 4k becomes mainstream, it's absolutely right. wink.gif



Ian
Edited by mailiang - 1/14/13 at 9:27pm
Reply
Reply

AVS Top Picks

post #32 of 39
1/14/13 at 10:01pm
Thread Starter 
Quote:
Originally Posted by Joe Bloggs View Post

What if they entire frame or 90% of it is black?
The frame data couldn't possibly just be flipping by without demarcation of any kind. That doesn't make any sense. If it were doing that, (chuckling), you'd have a vertical hold option like in the old days and but even back then it could lock on the "vertical blanking". Further, not since the RS-232 days have I worried about "no data" being confused with 0's, or whatever analogy works in HDMI.
Reply
Reply
post #33 of 39
1/14/13 at 10:12pm
Quote:
Originally Posted by tgm1024 View Post

The frame data couldn't possibly just be flipping by without demarcation of any kind. That doesn't make any sense. If it were doing that, (chuckling), you'd have a vertical hold option like in the old days and but even back then it could lock on the "vertical blanking". Further, not since the RS-232 days have I worried about "no data" being confused with 0's, or whatever analogy works in HDMI.
I meant just looking for "black" (or RGB 16,16,16) in a 1.78:1 encoded frame (or other frame) doesn't tell you the intended aspect ratio of content (though I suppose if you looked at enough frames you could tell or have a good idea, but even then there could be problems eg. switching aspect ratio, lots of black in the frame, or even when the content makers want the black bars visible, eg. to have things go past the black bars etc.). And even if it did I don't think a normal (16:9) TV takes much notice of it. eg. if you increase the brightness, it will increase the brightness of the black bars, when surely those should always be totally black. Also, being RGB 16,16,16, there will be video below that level, so even normally I don't think they're the blackest thing in a video.

eg. a totally black frame from a 2.35:1 video, encoded inside 1.78:1 video format (such as Blu-ray's 1920x1080) would be the same as a totally black frame from a 1.78:1 video encoded inside a 1.78:1 video. In both cases, all visible pixels would be black. You couldn't tell by that 1 frame alone if the video was supposed to be 2.35:1 or 1.78:1 (assuming the black for the active picture area is the same one use for the black bars).

And TVs that do zoom aren't always a good thing - that's the reason why people like the BBC zoom clips from old shows to 16:9 instead of showing them at their original 4:3 ratio - so that TVs don't keep switching.

So I'm basically saying, if you are going to encode everything in one frame aspect ratio (eg. 3840x2160 or 1920x1080) but use black bars/pillar box bars encoded into the picture, it would help, assuming they did it correctly, if there was a bit of extra information encoded that contained info about the intended aspect ratio of the content.
Edited by Joe Bloggs - 1/14/13 at 10:40pm
Reply
Reply
post #34 of 39
1/14/13 at 10:41pm
Thread Starter 
Quote:
Originally Posted by Joe Bloggs View Post

I meant just looking for "black" (or RGB 16,16,16) in a 1.78:1 encoded frame (or other frame) doesn't tell you the intended aspect ratio of content (though I suppose if you looked at enough frames you could tell or have a good idea, but even then there could be problems eg. switching aspect ratio, lots of black in the frame etc.). And even if it did I don't think a normal (16:9) TV takes much notice of it. eg. if you increase the brightness, it will increase the brightness of the black bars, when surely those should always be totally black. Also, being RGB 16,16,16, there will be video below that level, so even normally I don't think they're the blackest thing in a video.

eg. a totally black frame from a 2.35:1 video, encoded inside 1.78:1 video format would be the same as a totally black frame from a 1.78:1 video encoded inside a 1.78:1 video. In both cases, all visible pixels would be black. You couldn't tell by that 1 frame alone if the video was supposed to be 2.35:1 or 1.78:1 (assuming the black for the active picture area is the same one use for the black bars).

Sure, sure, but my original comment was in response to this:
Quote:
Originally Posted by CatBus 
Ideally I think the media should be have the film in its native aspect (no black bars encoded at all on any AR), and the player just reads this metadata and adds whatever matting is necessary to match its output resolution.

Having a BluRay movie in it's natively shot format means that there are no black bars (unless actually part of the movie). He was referring to an additional bunch of meta-data (data describing data) to show the actual movie sizes. I was pointing out that there is no meta data needed. All the frame information is already there. He was talking about a situation that didn't exist yet, and if we were to go down that route, there is no need for that meta data at all. The information as is (so long as the black bars weren't added in by the player, or in the movie data itself), is already there anyway. As for what is allowed to be assumed today, I'll have to look into the BD data format specification itself someday to see if there is a header responsible for all frame data that is overly relied on and required to be read only at the beginning. It would make no sense to have the rest just be a stream of data, it would make any recovery from a bounce or glitch be a complete nightmare, but it would be interesting anyway.
Reply
Reply
post #35 of 39
1/14/13 at 10:53pm
And I'm saying that metadata describing the aspect ratio could be useful and work better than not using it - especially if the display device could also make use of it (though I'm not sure it could - it gets the uncompressed picture).
Edited by Joe Bloggs - 1/14/13 at 10:58pm
Reply
Reply
post #36 of 39
1/14/13 at 10:57pm
Thread Starter 
Quote:
Originally Posted by Joe Bloggs View Post

So I'm basically saying, if you are going to encode everything in one frame aspect ratio (eg. 3840x2160 or 1920x1080) but use black bars/pillar box bars encoded into the picture, it would help, assuming they did it correctly, if there was a bit of extra information encoded that contained info about the intended aspect ratio of the content.

(I'm trying to catch up with the edits). Yep, and we're not exactly disagreeing, but I'd still argue that this is the cart before the horse. If we're going to have a world where the black bars are in the data itself along with information to tell us where the original movie is, we'd be far better off just to have the movie as is in some native "ideal" format, and have the receiving device decide what to do. If part way through the film the AR changes from 2.35:1 to 100:1 (pick a number smile.gif), then the frame size is recognized as it is read and the bars are drawn in accordingly by the display device (or player if need be).

It's discussed above that this isn't the case currently, because a movie that starts off X:1 and widens midway to Y:1 (Y > X), is a X:1 movie the entire way through with bars drawn in where the Y:1 scene starts. That's silly IMO.

The only time (as I was pointing out with the Full Screen example of DVD's) that extents need to be supplied that are different from the frame data itself are when you're attempting to throw away data and want someone else's idea (think "pan and scan") to fit a different AR.
Reply
Reply
post #37 of 39
1/14/13 at 11:10pm
Quote:
Originally Posted by Chronoptimist View Post

I'm not sure why that's so surprising. When I get right up close to a 16:9 display, I lose the edges before the top and bottom leaves my field of vision. Probably explains why I never had a problem with 4:3 displays - in fact I am still quite fond of the 4:3 aspect ratio. I would much rather have a 4:3 notebook than the 16:9 or 16:10 displays manufacturers seem to be pushing these days.
An interesting discussion comes up with FOV. Sometimes FOV is confused with PFOV, etc.
1. Is FOV the amount of image you can stare directly at (eyeball tracking area)?
2. Is FOV the amount of image you can see while eyeballs looking at center?
3. Is FOV the amount of image you can scan/glimpse from all peripheral vision?
Remember, you gain some extra peripheral vision if you stare far-leftmost or far-rightmost. There is not much upwards/downwards peripheral vision. You have much more horizontal peripheral vision, even if you can't track your eyeballs to stare directly.

You will observe FOV conditions (1) roughly resembles 4:3, but FOV condition (2) roughly resembles 16:9, and FOV condition (3) is wider than 16:9 -- even resembling 21:9 ... So the answer is not as simple as it seems. A good test is a very wide wall. Walk up to a wall staring at the middle, until the floor/ceiling disappears below the top/bottom of your vision field. Now, you can easily measure approximate aspect ratios of (1), (2) and (3) by marking the leftmost/rightmost part of walls you are able to "see" under conditions (1) vs (2) vs (3). You will notice that they are each progressively wider aspect ratios.

But FOV condition (1) is applicable to computer monitors, since you do need to be able to stare directly at all four corners of the display. So many people prefer 4:3 displays for computers; which is understandable. A 16:10 display is often a good compromise; while others of us prefer 16:9 displays for the standardization of everything (computer, gaming, videogames, television, etc).
Edited by Mark Rejhon - 1/14/13 at 11:22pm
Reply
Reply
post #38 of 39
1/15/13 at 6:14am
Thread Starter 
Quote:
Originally Posted by Mark Rejhon View Post

An interesting discussion comes up with FOV. Sometimes FOV is confused with PFOV, etc.
1. Is FOV the amount of image you can stare directly at (eyeball tracking area)?
2. Is FOV the amount of image you can see while eyeballs looking at center?
3. Is FOV the amount of image you can scan/glimpse from all peripheral vision?
Remember, you gain some extra peripheral vision if you stare far-leftmost or far-rightmost. There is not much upwards/downwards peripheral vision. You have much more horizontal peripheral vision, even if you can't track your eyeballs to stare directly.

You will observe FOV conditions (1) roughly resembles 4:3, but FOV condition (2) roughly resembles 16:9, and FOV condition (3) is wider than 16:9 -- even resembling 21:9 ... So the answer is not as simple as it seems. A good test is a very wide wall. Walk up to a wall staring at the middle, until the floor/ceiling disappears below the top/bottom of your vision field. Now, you can easily measure approximate aspect ratios of (1), (2) and (3) by marking the leftmost/rightmost part of walls you are able to "see" under conditions (1) vs (2) vs (3). You will notice that they are each progressively wider aspect ratios.

But FOV condition (1) is applicable to computer monitors, since you do need to be able to stare directly at all four corners of the display.
No, I'd actually argue that regardless of the field of view, we have a set of tasks that are important to us. It's all related to the eye of course, but paper, for instance, is most easily handled in a portrait fashion. Probably because it shortens the burden of L/R tracking. And so it follows that content is easiest that way as well. Our websites for instance, follow that same model. So having a taller screen AR matches the vertical bias of the task at hand, not the AR of the eye.

Quote:
So many people prefer 4:3 displays for computers; which is understandable. A 16:10 display is often a good compromise; while others of us prefer 16:9 displays for the standardization of everything (computer, gaming, videogames, television, etc).

Ok. Couple questions. Here's the 3 scenarios you've outlined, along with the AR you suspect it to approximate in bold:
  1. (4:3) The amount of image you can stare directly at (eyeball tracking area)?
  2. (16:9) The amount of image you can see while eyeballs looking at center?
  3. (21:9) The amount of image you can scan/glimpse from all peripheral vision?

First, this is curious: why is your scenario 2 wider than scenario 1? If there's no tracking path, and you're indeed staring at the center, then you're locked at periphery only. Shouldn't that be narrower than 1?

Edit: Or are you saying that given the eye movements, the ability to point the eyes up and down bring us to 4:3? As I read what you're listing out, either 1 and 3 seem the same, 1 and 2 are the same, or 2 and 3 are the same. I guess I can't ferret out how all 3 are unique.

Second, with human vision, strict FOV aside, there is a field weighting of what we can physically perceive (the rods & cones vary in density) and further the neurological apparatus has an additional weighting, presumably from the evolutionary imperatives of dealing with motion. What do those do to the aspect ratios?
Edited by tgm1024 - 1/16/13 at 6:40am
Reply
Reply
post #39 of 39
2/7/13 at 1:04pm
I'll rephrase, with tweaks for clarity.

(4:3) The amount of image you can stare directly at (eyeball tracking area)
That's your ability to stare directly at. Not peripheral, not corner of eye.

(16:9) The amount of image you can indirectly see while eyeballs looking stationary directly ahead at center of vision
That's your accessible peripheral vision without being allowed to move your eyeballs.
That's the vision that you can see indirectly, but can't look at directly.

(21:9) The amount of image you can indirectly scan/glimpse from all peripheral vision
That's your accessible peripheral vision while being allowed to also move your eyeballs.
You will notice you can see more peripheral vision to your left, if you move your eyeballs a little bit left of center of your vision.
You will notice you can see more peripheral vision to your right, if you move your eyeballs a little bit right of center of your vision.

I completely guarantee you that these are different aspect ratios, even if they are not the quoted aspect ratios.
And because of law of physics, I guarantee vision fields 1 < 2 < 3, regardless of actual aspect ratios.
The aspect ratios are just approximate -- and human vision is non-rectangular.
Edited by Mark Rejhon - 2/7/13 at 1:12pm
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home

AVS Top Picks

AVS › AVS Forum › Display Devices › Flat Panel General & New FP Tech › Are panels that are wider than 16:9 a sustainable trend?