Our Sponsor Sideshow Collectibles Send us News
Lord of the Rings Tolkien
Search Tolkien
Lord of The RingsTheOneRing.net - Forged By And For Fans Of JRR Tolkien
Lord of The Rings Serving Middle-Earth Since The First Age

Lord of the Rings Movie News - J.R.R. Tolkien
Do you enjoy the 100% volunteer, not for profit services of TheOneRing.net?
Consider a donation!

  Main Index   Search Posts   Who's Online   Log in
The One Ring Forums: Tolkien Topics: Movie Discussion: The Hobbit:
How come PJ never saw any problems with 48 fps?
First page Previous page 1 2 3 Next page Last page  View All

Artemis Roach
Bree


Dec 11 2012, 1:30am

Post #51 of 57 (120 views)
Shortcut
Anime [In reply to] Can't Post

Dropping the frame rate won't necessarily remove blur. Think of an animated gif which can be 2 to 10 frames a second. The motion is there - it isn't as smooth as 24 frames per second - but the motion is there, and there's no blur unless it is in the original image. (Same thing with a flip-book.) Indeed, blurred images can actually help the perception of motion in an animated gif.


In Reply To
If they need to add blurring, then why aren't they just dropping the frame rates back down instead?



Escapist
Gondor


Dec 11 2012, 2:08am

Post #52 of 57 (119 views)
Shortcut
It sounds like this may be related to what a person is "used to" [In reply to] Can't Post

Are they used to 48fps from watching soaps or 24fps from watching absolutely boatloads of movies? Or neither?

I know that different people's eyes adjust differently in other ways such as how well they focus on distant or close object and how they perceive color sometimes.

I guess that the decision makers got it right by having a variety of viewing options.

But I do have another question ... if we automatically create blur when things move too fast in real life, why wouldn't we perceive/create blur if the things we are looking at in real life are things on a screen? I know that the blur we see in reality isn't "really there". What about the screen capture stops a person from creating the blur as usual? Is it dependent on how we have trained our eyes to see? Is that the whole point of this comment I am replying to?

Sorry if I am asking stupid questions ... I guess that is my approach to a new idea - think of questions to ask.


Artemis Roach
Bree


Dec 11 2012, 2:14am

Post #53 of 57 (114 views)
Shortcut
They aren't stupid questions. [In reply to] Can't Post

We're discovering in real time the way our brains are processing technology.


JWPlatt
Grey Havens


Dec 11 2012, 2:33am

Post #54 of 57 (112 views)
Shortcut
Mostly [In reply to] Can't Post

I agree with most of this, except that camera motion can alleviate parallax and tracking that would otherwise contribute to the blur of a stationary observer or even an observer who cannot move his eyes as quickly as a camera can track the object.


Phibbus
Rohan


Dec 11 2012, 2:54am

Post #55 of 57 (109 views)
Shortcut
Indeed, no stupid questions [In reply to] Can't Post

Neuroscientists are still working this stuff out, and they haven't gotten much further with tangible measurements than flies' eyes (and they change their minds frequently.)

In Reply To
But I do have another question ... if we automatically create blur when things move too fast in real life, why wouldn't we perceive/create blur if the things we are looking at in real life are things on a screen? I know that the blur we see in reality isn't "really there". What about the screen capture stops a person from creating the blur as usual? Is it dependent on how we have trained our eyes to see? Is that the whole point of this comment I am replying to?

Picture the case of someone's hand moving across your field of vision in the period of one second. After the first 48th of a second, it will have moved a short distance across that field. In real life, during that 48th of a second, the person's hand would have been in an infinite number of intervening positions, and your eye would have received all of them as sensory stimulusŤ. Your brain cannot process all of that information; it captures a certain amount of it as recognizable images of the hand, and the rest it interprets as a blur (which is greatly oversimplifying, but that's the long-and-short of it.)

Now, if you were filming that same hand moving across the the camera's frame at 48fps, the same infinite number of positions that the hand passed through in that first 48th of a second would be reduced to two still images separated by a given amount of space. The infinite information that occurred between those two frames is not present for your eye to receive as stimulus, and your brain will not supply it or trick itself into thinking it is present.

Man is but an ass if he go about to expound this dream.


Lacrimae Rerum
Grey Havens

Dec 11 2012, 10:07am

Post #56 of 57 (83 views)
Shortcut
Forgive me if I'm being dim. [In reply to] Can't Post

This is not my field. How does that differ if the numbers involved are 24 rather than 48. One would imagine it is just a bigger gap between the still hand images, still no blur. And yet that seems at odds with the general duscussion.

LR


Phibbus
Rohan


Dec 12 2012, 4:08pm

Post #57 of 57 (120 views)
Shortcut
Sorry, I'm sure I'm not explaining terribly well [In reply to] Can't Post


In Reply To
This is not my field. How does that differ if the numbers involved are 24 rather than 48. One would imagine it is just a bigger gap between the still hand images, still no blur. And yet that seems at odds with the general duscussion.

First, understand that I'm not arguing for or against 48fps, only challenging the notion that deliberately minimizing motion blur necessarily makes a "better" or more realistic filmic image.

I should also have explained that the example I gave assumes an ideal condition which couldn't actually exist: namely a motion picture camera capable of an infinitely short exposure time, since that is what would be required to actually capture no motion blur at all in the image of the moving hand, whether at 48 or 24 fps. The main point I was trying to make was that, given that situation—which would be sort of a real-life equivalent of the animated GIF which Artemis Roach cited—our brains would not supply the blur.

The amount of blur that does get captured in the image depends on how fast the object being filmed is moving and how long you actually expose the medium to it (whether film or electronic sensor.) The greater the distance covered by the moving object while the medium is being exposed, the more area of the medium is actually exposed to it, and thus the greater amount of motion blur. Given an object in constant motion, the less time you expose each frame to it, the less motion blur is captured (and the more detail you will be able to see in the object, itself.)

The exposure time (or shutter) in cinema photography is traditionally expressed in degrees, since film cameras normally use a rotating shutter with an adjustable-sized aperture (think in terms of a pie chart.) Since the shutter rotates synchronously with the film transport, the more degrees you open the aperture, the longer each individual film frame will be exposed to the subject while in the gate. Thus, the shutter represents the percentage of time each frame is actually exposed, up to the maximum dictated by the frame rate.

For 24fps film, each frame has a theoretical maximum exposure time of 1/24 second (since that is the full amount of time each frame is actually in the gate.) A 360° shutter (in which the aperture would be fully open with no actual shutter present) would expose each frame for that full 1/24 second. With film, this is again a practical impossibility since some portion of the cycle is occupied by the transport of the film to the next frame, during which the film must not be exposed.

With a 180° shutter, the aperture would be half open, and each frame would be exposed for 50% of the frame cycle. So, for 24fps, the exposure time for each frame would be 1/48 of a second; for 48fps a 180° shutter would yield an exposure time of 1/96 of a second. A 90° shutter would yield 25% exposure times of 1/96 and 1/192 of a second, respectively. A better-illustrated discussion can be found in the Wikipedia article for "Rotary Disc Shutter." It also has a nice chart of blur effects achieved by various degrees of shutter:



So, to get back to the example: Assuming that the hand was being filmed simultaneously by both 24 and 48 fps cameras, if the shutter were infinitely small ("tight") for each, each would capture perfectly crisp still images of the hand with no motion blur. As you say, between the first two frames of film, the distance covered by the hand would be twice as great for the 24fps camera (since the hand at the second frame would be in the same position captured in the third frame of film by the 48fps camera.) The 48fps motion would appear smother when played back, but both would still appear "staggered" (the Wikipedia article uses "stuttered".)

If both cameras were using a 180° shutter, both would capture half the hand's trail of motion (as illustrated by the middle "Normal Motion Blur" example in the Wikipedia diagram.) However, for the 48fps camera, while the same 50% of the motion is being captured, the frequency is twice as great (the alternating dashes representing the exposure/non- in the diagram would be half their length.)

Thus, at the same shutter, the motion blur captured by 48fps would be half that captured by 24fps (which is again oversimplifying, but basically accurate.) The 48fps image will appear crisper under most circumstances.

In order to capture the same amount of motion blur as the 24fps camera at a 180° shutter, the 48fps camera would have to employ a 360° shutter. With digital cameras, unlike film, a 360° shutter is possible. (In fact, shutters greater than 360° are possible, where information buffered during capture of the previous frame is applied to the current one.) With a 360° shutter, the Wikipedia diagram would be a constant line with no interruptions in exposure.

However, certain problems are presented by this situation: First, our brains don't seem to care much for uninterrupted blur trails under many circumstances, and they can yield a streaky effect akin to time-lapse photography of car headlights on a highway. (My own instinct is that such an image does not present sufficient "stop points" for our brains to process an image of the thing, itself, rather than just the blur, however I have not read any scientific confirmation thereof.)

Secondly, with most mosaic CMOS digital sensors, certain problems affecting later color grading are introduced as the shutter widens above 180°. I've already gone on too long, but I've read that Jackson himself discusses these in one of the vlogs (if someone can point out where and in which one, I'd appreciate it, since I have not seen it.)

Thus, with 48fps digital, there is a natural predisposition toward shorter exposure times and less motion blur.

The question then becomes: how much motion blur is the "right" amount? Is it determined by the way we perceive things in real life, or is it a response conditioned by our exposure to the way things have "always been done"? Cinematographers have approached the problem mostly through experimentation, observation, and practical application rather than through any scientific principle or measurement. They have traditionally used wide and tight shutters with amended or clipped blur as colors in their palettes to represent deviations from "normal reality." However, experience tells us that people can enjoy video games at high frame rates with no motion blur at all.

My own suspicion is that our brains use the blur as a visual queue as to how fast an object is actually moving and that it does affect our perception of how "realistic" the motion is, although I won't speculate as to how much of that is conditioning. I have not yet seen the film, but I also suspect the reports of The Hobbit appearing "sped up" at 48fps to some viewers may be due to decreased motion blur (akin to the "Benny Hill effect" proposed by someone in another post, where footage is played back at a faster rate than it was captured.)

Man is but an ass if he go about to expound this dream.

First page Previous page 1 2 3 Next page Last page  View All
 
 

Search for (options) Powered by Gossamer Forum v.1.2.3

home | advertising | contact us | back to top | search news | join list | Content Rating

This site is maintained and updated by fans of The Lord of the Rings, and is in no way affiliated with Tolkien Enterprises or the Tolkien Estate. We in no way claim the artwork displayed to be our own. Copyrights and trademarks for the books, films, articles, and other promotional materials are held by their respective owners and their use is allowed under the fair use clause of the Copyright Law. Design and original photography however are copyright © 1999-2012 TheOneRing.net. Binary hosting provided by Nexcess.net

Do not follow this link, or your host will be blocked from this site. This is a spider trap.