Hobbit: HFR

From TIGwiki2
Jump to: navigation, search

Hi all

I haven't spotted any discussions on the new Hobbit shown in 48fps HFR...


I went to the cinema last weekend and saw it. I was a bit sceptic in advance and after seeing it I'm still not sold on the idea.

Sure the '3D-effect' was very good in HFR or maybe it was just because I went to a much smaller screen then I usually do. The image was unusually 'bright and sharp' and the 3D-depth was very good... Although with the 3D that good I thought it just made it feel more fake.

But what I really had a problem with was that it felt as the video was playing too fast. It was almost as having a 24fps footage being played at 30fps. But I doubt that the film was playing in 60fps instead of 48. Because the audio still sound good and was in sync.


So, what's your opinion on HFR?

/Carl


(Sent from mobile device) ____________________________

  • Carl Skaff*
  • Colorist*

Stopp Stockholm Office +46 8 50 70 35 00

  • Stockholm | Los Angeles | Linz*

www.stopp.se



I've seen it in both 24 fps flat and 3D and the HFR 3D version.

My first impression was the same as yours: it looked like film running at twice the speed (which it was!) but everything was moving normally, not unlike the speed-up sequences used in motion pictures these days to denote the passage of time.

However, it took me about 15 minutes to get used to the non-24fps judder and once my brain got used to the new frame rate with its own look, I started to enjoy the movie and forget that it was a new technology. For someone who sees dozens of movies a year as part of my work, suspension of disbelief is difficult for me. It worked for me with this film.

It did look more like 'video' to me, but that was more because I was missing the 24 frame judder that I'm used to with 24fps material in an electronic raster (which you DON'T get with 24 fps film, even though projectors play the same frame twice). It was also remarkably like the frame-quintupling 120Hz monitors we use, which allows us to see 24fps moving images without the flicker of running at 24 fps frame rate: clean but with a 'video' (IE judderless) look.

My impression, on seeing it several times in different formats now, is that if folks give 48 fps a chance and get used to the (currently) unique look of motion pictures presented that way, it will open up a level of creativity and perhaps even enjoyment (or at least more effective suspension of disbelief) that 24 fps moving image presentations cannot present because of the lower frame rate. I'm not sure the terms 'better' or 'worse' are appropriate here...just the term 'different'.

James


James Snyder Senior Systems Administrator Library of Congress - National Audio Visual Conservation Center (NAVCC) Motion Picture, Broadcasting & Recorded Sound Division (MBRS) Packard Campus for Audio Visual Conservation http://www.loc.gov/avconservation/packard/




On Mon, 2013-01-07 at 00:56 -0500, James Snyder wrote: > I'm > not sure the terms 'better' or 'worse' are appropriate here...just > the term 'different'.


I agree. I was expecting to hate 48fps, but I tried to give it a chance and I actually enjoyed it. I was also prepared for that sensation of everything being sped up at first -- that's been a very common reaction. That shot of Bilbo walking down the hallway at the beginning looked especially quick to my poor brain. Maybe HFR films of the future will use more slo-mo in their opening sequences?

I did want to get people's impressions of the grade. I liked it overall, but there were a few instances where the sky had been corrected to an intense shade of Florida swimming pool. Much too much green for my taste. Did that bother anyone else?

Owen Williams




If everybody is either hating it or saying "it's not as bad as I thought it would be", my question is whether its really worth the extra expense and overhead to make a film HFR.


Skip Elsheimer A/V Geeks LLC



I will say that HFR makes the 3D much easier to watch, and a much more pleasant overall experience. However the weirdness factor of the motion in undeniable, and i believe it stems from 48hz, being a unique refresh rate. Noting else we're used to refreshes at that rate, except for film running at double time, ergo the impression that it looks sped up. IMO a better target rate for HFR would be 60fps, same as showscan, this gives you truly fluid movement, and it's familiar to us, as it's 60hz refresh rate is familiar to our eyes from NTSC and HD TV. None of this would of course negate the "soap opera" effect which so many people despise.

All that said, my biggest issues with the Hobbit had little to do with the HFR and much more to do with the ugly clipping and weird chroma peaks. Made the shire sequence look like it was several generation old VHS. And you'd think with the budget they'd be able to do a VFX pass to bring back Gandolf's Hat.

Juan Salvo



I have not watched the Hobbit or any HFR content. However, I am very happy that this is happening right now. The next step is 60fps, which should indeed be better. I have read that James Cameron will use 60fps for a major film.

The reason I am happy that this is happening right now is that we are now seeing "4k" televisions and projectors being marketed for use in home theaters. Most of us would not encounter much advantage from 4K on the screen size typical of a home theater, but there would be considerable advantage from 48fps or 60fps native video on our home theater screens. It is better for the next HD standard to be based on 1080/60P (sourced from Blu-ray media) rather than "4k" at a low frame rate.

Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/




I don't think there's any significant difference between 48Hz, 60Hz, etc. I get the same "sped-up" sensation when I see 120Hz interpolated material at Best Buy. I think 48 is an excellent choice because it's a simple operation to reduce it to 24 for compatibility. If someone wants to give 72Hz a try (3x), go right ahead.

I recently bought a TV and specifically didn't get one that supports 3D. And although I was slack-jawed watching The Master in 70mm, I also wouldn't bother to get a 4K TV (at 46", 1080 is already barely distinguishable from 720). If there were two Hobbit blurays available, 24 and 48, I'd buy the 24 (assuming the 48 worked with my dvd player).

48hz was a fun place to visit, but I don't think I'd want to live there.

Owen




I just saw it today for the first time. I went in expecting it to feel uncomfortable, but I have to say I didn't notice anything out of the ordinary. I put this down to the fact that I have a 3D monitor on my desk that (when I use it in 3D mode) runs at 120 Hz so I must be used to it. My wife says she felt uncomfortable for the first two scenes but then got used to it. Overall I'd say they've created a unique look with the film. For me it seems very "composited", i.e. the foreground looked flat with the background way off in the distance, almost like a diorama. There's a shot right at the beginning with a crowd of people running towards the camera that just looked horrendous, but I think most other shots looked rather good. I also didn't get anything near the eyestrain I'd expect from watching 3D for 3 hours.

Jack




I suppose that for The Hobbit it was fairly easy to go from 48>24 for the 'normal master'. And then from that they just do a 30 (3/2 pulldown) and 25 (speeded) master.

(Although. If it was shot in 48 then they must use that smaller sector on the shutter (maybe that's the wrong translation) and if you just do a 48>24 conversion and remover every other frame then it should 'judder' a lot. Does anyone know if there was any 'frame blending' going on to ad some motionblur to make it smoother? If so... Was that then done as an overall setting. Or as a scene per scene creative way?)


BUT How will Avatar2 handle the other deliverables? If its shot at 60fps... How do you convert that (in a GOOD way) down to 24 or 25?

Although I suppose a LOT of Avatar2 will be CG and then they could just render out a different version?

Anyone know if any more films are coming out soon in HFR? >From what I could google, the next one is won't be until the next Hobit-movie, a year from now. A bit weird that pretty much all 3D theaters in Sweden upgrade (for a lot of $) to support HFR if its just for one movie.

/Carl


(Sent from mobile device) ____________________________

  • Carl Skaff*
  • Colorist*

Stopp Stockholm Office +46 8 50 70 35 00

  • Stockholm | Los Angeles | Linz*

www.stopp.se



On Tue, 8 Jan 2013, Carl Skaff wrote: > A bit weird that pretty much all 3D theaters in Sweden upgrade (for a lot > of $) to support HFR if its just for one movie.

>From what I have read, in most cases this only required a software update (or minor hardware update) and not a complete replacement of projectors. Updating to 60fps may be more severe.

Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/



On Tue, January 8, 2013 3:11 pm, Bob Friesenhahn wrote:

>From what I have read, in most cases this only required a software > update (or minor hardware update) and not a complete replacement of > projectors. Updating to 60fps may be more severe.

While the upgrade to support 48fps stereo was simpler for Sony media block and projector, it wasn't necessarily true for TI based projectors. The only upgrade path was for series 2 projectors and required an integrated media block (a decryption and JPEG2000 decoder card located in the projector itself). Lucky if you've got a series 2 projector, not so lucky if you have a projector more than about a year old and have to replace the projector.

Simon

-- Simon Burley RPS Film Imaging Ltd Mobile: 07702 732 655



I asked the projectionist (actually he was the popcorn-guy) and he said they had to update with new boards in a hurry.

/carl

____________________________

  • Carl Skaff

Colorist* Stopp Stockholm Office +46 8 50 70 35 00

  • Stockholm | Los Angeles | Linz*

www.stopp.se



On 1/6/13 2:58 AM, "Carl Skaff" <carl at stopp.se> wrote:

> So, what's your opinion on HFR? >------------------------------<snip>------------------------------<

There's almost 900 messages hotly debating the look of 48fps in THE HOBBIT over on the RedUser Forum:

http://www.reduser.net/forum/showthread.php?90770-The-Hobbit-at-48fps

Me personally, I was expecting it to be hideous, but it didn't *kill* me. It definitely takes some getting used to, and a few shots throughout the movie bothered me. I also felt like some of the wigs and makeup seemed more obvious with the added clarity. My gut feeling was that Jackson was able to solve quite a bit of 3D's issues with motion artifacts, but he did so at the expense of creating new visual issues that some find distracting.

But it's still essentially a pretty good movie, and Andrew Lesnie does beautiful work as a DP (as did the Park Road Post color-correction team). And the movie looks like it's well on its way to making $1 billion; it just hit $777 million this weekend, worldwide...

--Marc Wielage post supervisor/colorist


Does HFR look odd because it does not simulate our visualcortex.

I went to see The Hobbit, 2K HFR yesterday. It all lookedpretty good to me,

except some object motions showed multiple sharp edges andsome of the

pans looked weird, again with multiple edges on some objecthighlights

For many years it has been realised that our visual cortex“blurs” fast motion.

The Film industry picked upon 24fps, with 180 degreecamera shutters that

gave a frame exposure of 1/48th of a second.Some in the visual perception

industry will say that this causes motion blur onthe frame that largely matches

our brains equivalent of motion blur.

With regular 3D the images are still shot at 24FPS, with twocameras, “grabbing” the left & right image simultaneously. They are thendigitally projected,

usually, as sequential left & right images, repeatedtwice or thrice for each image, ie, a 1/24 of a second camera captured image isrepeated, for each eye two

or three times on the screen, at projected rates of96 or 144 frames per 3D pair.

L,R,L,R, or L,R, L,R,L,R. But as the initialcamera rate was 1/24th of a second

with a 180 degree shutter, theimage blur represents 1/48th of a second..

Shooting 3D HFR, or indeed 2D HFR the camera frame rate is,at minimum,

increased to 48 frames per second. Assuming an effective 180 degreeshutter

is retained then the frame capture period is reduced to 1/96thof a second.

On static shots this is fine, but where motion is involved theimage captured will

be much sharper with less motion blur due to the shorterexposure.

Moving objects will look extra sharp. But the visual cortex is notused to seeing

moving objects this sharp, for no matter how long, or short theprojector periods

are for each frame or repeats of the frame, effectively thecortex is presented

with a 1/48th of a second of period in time, butwith the motion for only 1/96th

of that second.

Is this what we see in HFR as objects in motion look toosharp and the

visual cortex becomes confused..

The subject can be further subdivided into object motionwithin a shot

and panned scenes.

Objects that the eye tracks within a scene should remainsharp, those that we

do not follow should remain motion blurred. Scene pans arealways an unreal

experience to our visual system. We never see motion blur whenwe move our

eyes or turn our heads to pan across a scene as we move our eyes invery

high speed “jerks”, known as saccades. (Apparently the eye is the fastest

moving device in our body.) During these movements the brain turns our vision

off so we are blind to this motion in the real world while we move our visionfrom

point to point. Try it, If you move your eyes or your head and concentrateon

what you see you will note that the image almost moves in discrete steps,

either as jerky eye motion or as a weird feeling that you are not seeing

everything pass in front of you as you move your head; at no time does

the viewlook blurred.

How about object motion. Focus on, say, a point on your desktop,

now while not moving your eyes place your hand on the desktop and slowly

move it across you field of vision. At anything other than a very slow rate

your hand seems to take on almost a blur.

Is it, in HFR, this shooting of images at shorter exposuretimes

than 1/48 of a second, causing or visual system to see detail in motion

that we would not naturally see?

Just a thought!


As usual I have rabbited on for far too long


Happy New Year

Peter Swinson


It has become clear that humans learn how to see. In a natural environment, we learn how to see in a particular way. With old NTSC CRT TV we learned to see a nice image even though the picture was actually pretty terrible. If we read a novel, we envision the scenes as they happen, but in our own particular way. The brain fills in the gaps for the parts which are unclear or missing.

After switching from VHS to DVD, we became accustomed to the additional clarity and found VHS to be objectionable. After being accustomed to HDTV, we find DVD to be objectionable. Similarly, after being exposed to higher frame rates and more clarity, we will eventually find 24fps to be objectionable.

Bob -- Bob Friesenhahn bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/



On 1/10/13 10:34 AM, "Bob Friesenhahn" <bfriesen at simple.dallas.tx.us> wrote:

> It has become clear that humans learn how to see. In > a natural environment, we learn how to see in a particular way. With > old NTSC CRT TV we learned to see a nice image even though the picture > was actually pretty terrible. If we read a novel, we envision the > scenes as they happen, but in our own particular way. >------------------------------<snip>------------------------------<

I've said for years that if you grew up in the 1960s, you remember hearing classic rock hits playing on an AM radio, sounding about like a cellphone conversation today. But our brains filled in all the bass, gave it all the impact it needed, and our *memories* sound fine.

I don't doubt there are a lot of people watching movies and TV shows on their iPhones and iPads today, doing the same thing visually. (Sadly.)

--Marc Wielage post supervisor/colorist



@Peter Swinson...

Not rabbitting on at all. "LIfe of Pi-3D", even at 24 exhibited quite a bit of edge doubling on moving objects, although I'm trying to recollect whether it was more pronounced on moving objects within a scene or camera movement on still objects...

The mention of saccadic eye movement took me back to 1985, attending the Ottawa HDTV colloquium, where one of the early Sony HDVS systems was on display and one presenter made a point of tying HD resolutions, aspect ratio, and frame rate to the physical characteristics of human vision.

The Wikipedia entry on the subject mentions something called Saccadic Masking:

It is a common but false belief that during the saccade, no information is passed through the optic nerve to the brain. Whereas low spatial frequencies (the 'fuzzier' parts) are attenuated, higher spatial frequencies (an image's fine details) which would otherwise be blurred out by the eye movement remain unaffected. This phenomenon, known as saccadic masking or saccadic suppression, is known to occur in the time preceding a saccadic eye movement, implying neurological reasons for the effect, rather than simply the image's motion blur.[12] This phenomenon leads to the so-called "stopped-clock illusion", or chronostasis.

A person may observe the saccadic masking effect by standing in front of a mirror and looking from one eye to the next (and vice versa). The subject will not experience any movement of the eyes nor any evidence that the optic nerve has momentarily ceased transmitting. Due to saccadic masking, the eye/brain system not only hides the eye movements from the individual but also hides the evidence that anything has been hidden. Of course, a second observer watching the experiment will see the subject's eyes moving back and forth. The function's main purpose is to prevent smearing of the image.[4]


Joe Owens Presto!Digital Colourgrade 302-9664 106 Avenue Edmonton, Alberta T5H0N4 +1 780 421-9980 jpo at prestodigital.ca



--Rob Lingelbach 02:38, 12 January 2013 (UTC)