Luminous Landscape Forum

Equipment & Techniques => Motion & Video => Topic started by: DaveCurtis on April 15, 2011, 06:53:07 pm

Title: Hobit shot at 48fps
Post by: DaveCurtis on April 15, 2011, 06:53:07 pm
I see Peter Jackson is shooting The Hobit down here in NZ at 48fps rather than the usual 24. He is also using RED cameras. The production has bought dozens of them.
Title: Re: Hobit shot at 48fps
Post by: feppe on April 15, 2011, 08:46:18 pm
I see Peter Jackson is shooting The Hobit down here in NZ at 48fps rather than the usual 24. He is also using RED cameras. The production has bought dozens of them.

He's shooting at 48fps because he's shooting in the latest fad: "3D." I haven't been able to confirm if that means that only the "3D" version of the movie is 48fps, or whether also the 2D version is.

I'm mostly concerned that it'll have the same too-smooth hyperreal video look that utterly, completely and irrecovably ruined the cinematography of both Public Enemies and Miami Vice. I felt so disgusted about the look of the latter I considered demanding my money back - and Michael Mann is in my top 5 directors of all time. Curiously, it worked wonderfully in Collateral which is one of my favorite movies of the previous decade.

In the end I guess 48fps is where we're headed, and in 5-10 years 24fps will start looking outdated - and in 20 it'll be retro cool again. I hope Mr Jackson has the decency to make a good movie, instead of resorting to pure visual bravado, lacking substance and with the misanthropic story like Avatar did.
Title: Re: Hobit shot at 48fps
Post by: Sareesh Sudhakaran on April 16, 2011, 12:56:53 am
Shooting at a higher frame rate is actually a technical requirement for 3D. Cameron himself prefers to shoot at 60p or more. It won't matter if the story is told well, will it?
Title: Re: Hobit shot at 48fps
Post by: feppe on April 16, 2011, 06:55:13 am
Shooting at a higher frame rate is actually a technical requirement for 3D.

It's not a technical requirement; it's possible to shoot 3D at 24fps, but it would have an effective framerate of 12fps, which would look jerky.
Title: Re: Hobit shot at 48fps
Post by: Sareesh Sudhakaran on April 16, 2011, 07:32:30 am
You misunderstood my definition of technical requirement. Anyway, the higher the frame rate, the better the 3D effect. 48fps probably is a compromise between 24 and 60fps, plus the fact that PJ uses a lot of slo-mo. It makes perfect sense in his case.
Title: Re: Hobit shot at 48fps
Post by: feppe on April 16, 2011, 10:35:33 am
You misunderstood my definition of technical requirement.

No, I didn't; the phrase you used incorrectly implies one needs higher than 24fps to get a 3D effect.
Title: Re: Hobit shot at 48fps
Post by: Sareesh Sudhakaran on April 16, 2011, 01:56:39 pm
You are right. I should have been more precise.
Title: Re: Hobit shot at 48fps
Post by: feppe on April 16, 2011, 02:04:48 pm
How are the stereo cameras set up - are they both 48fps, or are 2x24fps combined to yield 48fps 3D picture? If latter, it wouldn't help smooth out slomo - although there's some very convincing software (http://www.youtube.com/watch?v=oj-EMtLIdPk) available to do that in post (think that's Twixor), technique pioneered by John Gaeta in The Matrix -, and the 2D version of the film would be in traditional 24fps.
Title: Re: Hobit shot at 48fps
Post by: JeffKohn on April 16, 2011, 03:10:04 pm
I agree with feppe about the effect of higher frame rates as far as regular 2D projection goes. If the 2D version is at 48fps the result would be similar to what you get from some of the newer 120hz flat panel TV's which interpolate to higher frame rates - something I don't care for at all. Some people like that hyper-real 3D look, but I personally would rather not have movies look like soap operas or news broadcasts.
Title: Re: Hobit shot at 48fps
Post by: Peter McLennan on April 16, 2011, 04:41:39 pm
Higher acquisition speeds than the traditional 24 fps used by film cameras have decided advantages.  I shot aerials in Showscan format many years ago and the 60 fps camera rate made for some remarkably clear imagery. However, crystal clear rendition of motion may not be consistent with the film-maker's creative vision.  Motion blur can be attractive to the eye, as any computer animator knows.

James Cameron knows what he's doing.  He's considering using 60 fps for future 3D productions.  "Good!", I say.
Title: Re: Hobit shot at 48fps
Post by: hjulenissen on April 17, 2011, 07:40:31 am
I agree with feppe about the effect of higher frame rates as far as regular 2D projection goes. If the 2D version is at 48fps the result would be similar to what you get from some of the newer 120hz flat panel TV's which interpolate to higher frame rates - something I don't care for at all.
I think that the "guesswork" of so-called 120Hz tvs cannot be directly compared to the real information recorded by increased capture rates.

I am all for equipment that has the capability of increased realism. If someone wants the esthetics of jerky 24fps, that is easy to do in postproduction, but the other way around is harder.

-h
Title: Re: Hobit shot at 48fps
Post by: Sareesh Sudhakaran on April 17, 2011, 08:28:02 am
Here's an article about someone who probably knows what he is talking about: http://www.geek.com/articles/geek-cetera/james-cameron-wants-avatar-2-to-be-60fps-3d-2011041/
Title: Re: Hobit shot at 48fps
Post by: Graeme Nattress on April 17, 2011, 08:47:46 am
Info from Peter Jackson here: http://www.facebook.com/notes/peter-jackson/48-frames-per-second/10150222861171558

Graeme
Title: Re: Hobit shot at 48fps
Post by: feppe on April 17, 2011, 09:04:47 am
Info from Peter Jackson here: http://www.facebook.com/notes/peter-jackson/48-frames-per-second/10150222861171558

Thanks - he puts a clear marketing spin into it, but I'll pass judgment when I see it. It looks like it's 2x48fps, so the 2D version will be available at 48fps as well.

I wonder how Bluray and 2D/3D TV standards work with 48fps and 60fps. It won't go well with the consumers if they are expected to buy a new set in 3-5 years to display 48/60fps.
Title: Re: Hobit shot at 48fps
Post by: Graeme Nattress on April 17, 2011, 09:13:41 am
Absolutely we need to see it, but there's no reason why it shouldn't look great, and also provide a perfect 24fps extraction for legacy applications - indeed, that's the true beauty of it. Although a bit more expensive in WETA render times for all the VFX, it's almost a no-risk experiment.

Graeme
Title: Re: Hobit shot at 48fps
Post by: John.Murray on April 25, 2011, 07:50:28 pm
fascinating!  although i truly wish he'd left out the analogy regarding vinyl and cd's - a properly setup and aligned turntable with a decent moving coil cartridge simply blows away any cd.......
Title: Re: Hobit shot at 48fps
Post by: Graeme Nattress on April 25, 2011, 07:55:35 pm
Agreed on the vinyl, John.

Graeme
Title: Re: Hobit shot at 48fps
Post by: feppe on April 25, 2011, 08:00:19 pm
CDs don't sound like crap because of an inherent shortcoming of the technology, but because of loudness war (http://www.youtube.com/watch?v=3Gmex_4hreQ). You can't press a vinyl with flat audio, ie. lacking headroom (dynamic range) and have it play properly (or so I've heard), but you can do so with CDs.
Title: Re: Hobit shot at 48fps
Post by: Graeme Nattress on April 25, 2011, 08:30:14 pm
Loudness war certainly doesn't help. There's a lot that can be done in the mastering process that can destroy the music. CD though, as a sound replay medium, has come on a long way since it's inception, and when done well, it can be quite superb indeed. The main issues I see with digital audio are not in that it's sampled, but that often such poor filters are used in the sampling and reconstruction process.

Graeme
Title: Re: Hobit shot at 48fps
Post by: John.Murray on April 25, 2011, 11:37:13 pm
Graeme:  Agreed!  In respect to "loudness" or more properly dynamic range; analog media such as an LP is capable of storing relevant musical information well under the noise floor, not to mention intangibles such as timing and pace (http://www.stereophile.com/reference/23/index.html).  Great or poor engineering is exactly that, irrespective of the medium.

Feppe:  Two legendary mastering engineers, Stan Ricker (http://www.enjoythemusic.com/magazine/rickerinterview/ricker6.htm) and Bob Ludwig (http://www.musictap.net/Interviews/LudwigBobInterview.html) would disagree with you.

Finally, I don't mean to hijack this thread or confuse the issue - digital audio is quite a different thing than digital imaging, primarilly because of the issues raised in the Stereophile article.
Title: Re: Hobit shot at 48fps
Post by: hjulenissen on April 26, 2011, 03:50:24 am
...The main issues I see with digital audio are not in that it's sampled, but that often such poor filters are used in the sampling and reconstruction process.
The "problem" is that there are no known, peer-reviewed listening tests that supports this being a problem. If there were, it would be (relatively) easy to optimize the technology.

If it cannot (with certainty) be heard, then things get very complicated.

http://en.wikipedia.org/wiki/Super_Audio_CD#Audible_differences_compared_to_PCM.2FCD
Quote
Now, it is very difficult to use negative results to prove the inaudibility of any given phenomenon or process. There is always the remote possibility that a different system or more finely attuned pair of ears would reveal a difference. But we have gathered enough data, using sufficiently varied and capable systems and listeners, to state that the burden of proof has now shifted. Further claims that careful 16/44.1 encoding audibly degrades high resolution signals must be supported by properly controlled double-blind tests.(http://drewdaniels.com/audible.pdf)
Title: Re: Hobit shot at 48fps
Post by: hjulenissen on April 26, 2011, 03:56:10 am
analog media such as an LP is capable of storing relevant musical information well under the noise floor,
Would you care to explain this? Analog as well as digital media can both contain information well "below the noise floor". But as uncertainty grows and the needed integration time for information recovery grows, the value of that information decreases.

Even without dithering (you should always dither), the CD format allows for a dynamic range that no producers seems to care about using, and no human seems to be able to hear the difference...
Quote
Feppe:  Two legendary mastering engineers, Stan Ricker (http://www.enjoythemusic.com/magazine/rickerinterview/ricker6.htm) and Bob Ludwig (http://www.musictap.net/Interviews/LudwigBobInterview.html) would disagree with you.
Legendary mastering engineers tends to be great at making records, but not always at understanding the technology.

-h
Title: Re: Hobit shot at 48fps
Post by: John.Murray on April 29, 2011, 01:44:55 am
Sorry, I disagree, musical content near and below *objectional* noise in the digital domain has no comparision.  That's exactly why dither was introduced, it sounded like crap until we actually introduce noise.....  We are talking strictly 16bit depth 44.1khz - CD encoding, right?  If you're talking 24 or even 32 bit sampling depth, at 192Khz (which I now suspect Peter Jackson was *really* referring to) that's a different story....

Music is uniquely emotional, tough to pass through a peer review, how do you do that with a fine wine, or a lover?  As I've said before, audio has a rhythm and pace aspect that has no equivalent in the imaging domain (at least not yet... ) 

To give you a specific example; I have a well worn pressing of Miles Davis Kind of Blue (1957) on vinyl as well as a "remastered" version on CD (interestingly a couple of cuts on the vinyl pressing was made from a master from a machine running about a quarter tone slow).  I've asked more than one musician to actually reproduce the bass line from memory on "So What": using the CD required multiple playbacks - the bass line is simply "vague",  from vinyl, people could reproduce the bass line immediately.  Scientific?  - nope! , but to a musical ear - compelling.

I have to disagree with your assessment of Stan Ricker and Bob Ludwig as well, isn't the purpose of engineering to apply a given technology to achieve the above described result?  
Title: Re: Hobit shot at 48fps
Post by: hjulenissen on April 29, 2011, 07:40:23 am
Sorry, I disagree, musical content near and below *objectional* noise in the digital domain has no comparision.  That's exactly why dither was introduced, it sounded like crap until we actually introduce noise.....  We are talking strictly 16bit depth 44.1khz - CD encoding, right?  If you're talking 24 or even 32 bit sampling depth, at 192Khz (which I now suspect Peter Jackson was *really* referring to) that's a different story....
I am not following you here. Digital audio should always be dithered when reducing the number of bits.
Quote
Music is uniquely emotional, tough to pass through a peer review, how do you do that with a fine wine,
Serious tasters of fine wine do their testing of wine blind. If you want to publish any academic papers about audibility, you have to use scientific measures like blind-testing. The nut-cases in audiophile magazines are just that - nut-cases.
Quote
To give you a specific example; I have a well worn pressing of Miles Davis Kind of Blue (1957) on vinyl as well as a "remastered" version on CD (interestingly a couple of cuts on the vinyl pressing was made from a master from a machine running about a quarter tone slow).  I've asked more than one musician to actually reproduce the bass line from memory on "So What": using the CD required multiple playbacks - the bass line is simply "vague",  from vinyl, people could reproduce the bass line immediately.  Scientific?  - nope! , but to a musical ear - compelling.
There are several sources of confusion in your comparision. A CD made from one master and a vinyl made from another master may well sound differently, but there is no way knowing if this difference is due to differences in the master or differences in the medium.

People knowing that they listen to a rare, expensive vinyl pressing may think that they prefer it to a plain iTunes mp3, even if they are in fact listening to the same audio. The mind is a powerful thing.

Quote
I have to disagree with your assessment of Stan Ricker and Bob Ludwig as well, isn't the purpose of engineering to apply a given technology to achieve the above described result?  
Sound engineers are usually not engineers. They are not basing their career on rational, scientific explanations of how and why their equipment do what it does. It is probably possible to be an amazing photographer with out knowing the first thing about photons or CCDs or MTFs. It is primarily about delivering the end-result that appeals to people.

It might well be that some or most acclaimed audio engineers also have a solid grasp of the technical side of the equipment, but that is not a prerequisit to be an audio engineer. Therefore, I am sceptical about claims like "Ah, Bob Ludwig says that this $9000 power cable makes the CD sound better, so it must be true", just like I am sceptical about claims ala "this or that top violin player claims that their violin sounds more realistic when played back on a SACD system, and sinvce neither of us plays violin better than her, you should not argue with those claims"

-h
Title: Re: Hobit shot at 48fps
Post by: feppe on May 01, 2011, 02:46:38 pm
Oh dear, should have known better...