Epic vs Film - the Decisive Battle is looming

[Häakon]

I think that regardless of the tech specs the truth is that digital will replace film eventually due to the simple economics involved. All digital images needed to reach was a position where, broadly speaking, Joe Sixpack cannot differentiate between it and film - It's pretty fair to say that we've arrived at this stage.

...and I think that's pretty much all Tom was saying in his original post.

 

[Stephen Williams]

Another major point here, that I fear might be getting overlooked, is the CONTROL directors and DPs gain over their images with the RAW workflow specifically. It will be a driving factor.

Hi,

Actually we loose all control of the image as it's done in post.

Stephen

 

[James T Mather]

...and I think that's pretty much all Tom was saying in his original post.

the difference in my post is that I am saying it will be driven by economic concerns from a production standpoint rather than aesthetic concerns.

 

[jpp]

Another major point here, that I fear might be getting overlooked, is the CONTROL directors and DPs gain over their images with the RAW workflow specifically. It will be a driving factor.

Like I said earlier, I remember handing in film and crossing my fingers that the lab would do things right. At that point, control was out of my hands, literally.

Sorry to insist on a not very subtle point, but here we go again. What you say is absolutely true. Desktop control over images is wonderful. Although, like cinematography, it's best left to specialists.

But this wealth of powerful technology, and the fact that individuals of modest means can now own a virtual studio, has very little to do, I'd argue, with actually making movies. If worries about handling film or trusting the lab or cutting on a flatbed are really impediments, then the movie will probably never get made on any format, because there are million other reasons, Red or no Red, not to make features. And most of those reasons are grand: making a movie is the dumbest undertaking going if viewed as a business venture. Even having high hopes for a successful aesthetic realization is pretty dumb, unless you've got a lot of money.

I'm not sure I'm saying quite what I want to say here, but.... There's a falsity in all this digital enthusiasm. Seems like an evasion of the real difficulty, or a taking of comfort in all the wrong things. In its worst form, it can be just another kind of consumerism. Or maybe I'm just plain wrong here. Hope so, anyway.

 

[David Mullen ASC]

As a cinematographer, I can lose control over the image just as easily if it were digital or shot on film -- if I'm not there for the final color-correction. It doesn't matter how perfectly I delivered the image to post, how closely it matched my vision... it can still be screwed up in the D.I. process, whether it was shot on the RED or on film.

Control over the image on a typical feature film is much more of a political issue than it is a technology issue.

There are aspects of digital cinematography that I like in terms of controlling the image... but once it enters the post phase, it doesn't matter much how I shot it because it can be changed by other people. What really matters more is if I'm there at the finish line to complete what I started. Even when I shoot film and the dailies aren't accurate, as long as I know what is on the negative and I'm there for the final color-correction in the D.I., the final image will be how I want it to be. So ultimately, with digital (including film transferred to digital), the ultimate control is in post, and the issue is who controls that post.

At least with a photochemical finish through contact printing, I could lock in the look on the film negative a little more easily. Traditional RGB color timing couldn't make radical changes to contrast, saturation, or framing, not like today with digital post. So do I have more control today or less control? It all depends on if I'm the one doing the final color-correction. If so, then today I have more control, but I have that control whether or not it was shot digital or on film, but if I'm not doing the final color-correction, then I have less control today than ever, whether I had shot digital or on film.

There is a lot I like about shooting digitally... but to be blunt, everything about film sucks... except the final quality! So it partly depends on whether the process is more important to you than the results, because the results with film can be beautiful, whereas the process of shooting digitally is more fun (for me.)

It's sort of like shooting in 35mm anamorphic -- everything about it is a pain in the ass, except that the results are great.

 

[Tony Post]

I totally enjoyed that last post from David. The wisdom is just oozing. This is the program we all need to get with, especially post people. The new tech gives us the chance to reclaim the workflow, make it work for the film makers first, Directors, DOPs, people who have the vision. Thats what us post pros want too.
Film is the coolest thing ever, but roll over Beethoven, if we can make movie making easier, cheaper, more modern, more now, and still retain the discipline that our forebears have taught us......
The goal has to be a better experience for everyone, shooters, posters, clients, and most importantly the viewing public.:weight_lift:

 

[Bing Bailey]

I know the DOP is the guardian of the image , but isn't the director the final authority on if that image serves his intent and if it does not can he have it changed so that it does ? I think in the past colour timing was more like voodoo and the dop got to instruct the lab because he really was the only one besides the lab tech and colour timer who knew anything about the process.

I think whats happened since digital colour correction and post everyone is that much more informed and the process has been demystified somewhat and is more open to opinions of directors and producers. I'm not saying one way is better than another. but dop's kinda had the world to themselves for a very long time via the fact nobody else knew enough to instruct anybody and know things have evolved and thats changed for good and for bad.

DOP is now only one voice of many on the final image. maybe thats terrible but thats the way it is.

 

[David Mullen ASC]

Trouble is, we DP's often get the blame for a bad shot in a movie that's only bad because of the way a director asked for it to be color-corrected. Then later you read these comments: "jeez, the DP was an idiot -- did you see how noisy that scene was? He clearly had no idea what he was doing..."

When the truth may have been that the scene was shot a certain way because that's what the director said he wanted... but in post, the director changed his mind and wanted it another way, and the image couldn't handle the post manipulation and fell apart, but the director didn't really care about the technical quality.

Truth is that not all directors even notice some image problems. And not all directors have a sophisticated visual sensibility. They may end up brightening everything because they have this notion that comedy should always be bright regardless of scene content or the story point, or that there should never be a dark shadow in a movie.

Of course it's the director's right to make decisions that affect the image, but it can have an impact later on quality control standards for distribution... or the reputation of the DP, who may end up losing work because of some poor post decisions made by other people.

The thing is that half the time if the DP is there for the color-correction and points out some bad artifact that arises from a decision the director just made in regards to the timing, the director will say "thanks for pointing that out" and back-off from the mistake. There is a reason why some good directors surround themselves with experts in every department.

And sometimes I've found out that a bad color-correction decision made by the director was done as a lark, that they didn't really care one way or the other and were just goofing around with the image. Not every decision a director makes is a brilliant one, or even one they are passionate about or committed to. Half the decisions they make, they will modify when presented with a reasonable argument by one of their key trusted collaborators.

For example, you can have one scene where the director takes out all the warmth in the faces to make them neutral, causing the shadows and blacks to go blue-ish. You see it later and point out that the scene was supposed to be an orange sunset scene for a story reason and the director says "that sounds great -- I wish you had been there to remind me of that when we timed it!" He didn't have an opinion one way or the other, he was just "correcting it to look normal" as a default.

 

[David Mullen ASC]

I'm not sure the process has demystified at all -- I think some directors are more confused than ever about the color-correction process now that it has gotten so advanced. Sometimes having a lot more options just makes things more confusing for people.

I find myself having to explain a LOT MORE to directors these days about how things work in terms of image creation and color-correction and the path through post.

If the process is so demystified, then why do we have more "image experts" than ever in the process? DIT's, colorists, engineers, DP's, etc. Not every director is interested in the subtleties of RAW versus video encoded, compressed versus uncompressed, RGB versus YUV, CMOS versus CCD, etc.

 

[Radoslav Karapetkov]

Excuse me, I'm not too informed on this...

Doesn't the DP have the right of veto [kinda] over what happens with the shots?

And the right to be present and supervise the overall grading process, in case the director or someone else tries something crazy with the footage?

It is evident that improper grading can ruin the efforts and work of the DP.

Can all this be put in the contract?

I mean,... I thought that the director and DP have joint power over what happens to the vision of the film?

If one of them doesn't agree = nothing happens.

The DP is the main specialist on the creation of vision, and so - should have the right of co-authorship and control over that aspect of the film.

 

[Bob Gruen]

And most of those reasons are grand: making a movie is the dumbest undertaking going if viewed as a business venture. Even having high hopes for a successful aesthetic realization is pretty dumb, unless you've got a lot of money.

I'm not sure if I'm following your line of thought, or if you are being sarcastic, but to not view making a movie as a business venture is beyond foolish. how are you going to get investors to pony up a quarter of a million dollars to make a low budget shoot without treating it as a business venture?

To the original point of Epic vs Film (and the implied Red One) I'd say that there are many attractive aspects on either side, and it is the job of the producer to make the ultimate decision as to how the project is built.

I think that the Red One will emerge as a real player in the low budget realm for one reason: shooting ratio. I can build a 22TB storage server complete with tape backup for under $6,500, which is effectively an open shooting ratio for a low budget, single cam feature. That allows you the ability to give your actors repeated chances to get the scene right which means you get better performances. After all, poor acting is the single biggest problem in low budget independent film, and a low shooting ratio is usually why.

Finally (and I'm sure this will draw a lot of fire), I think Epic may be a flawed concept when looking at a current filmout workflow. My understanding is that there are two dominate resolutions of film printers in the industry, 2K and 4K. If you shot 5K a computer process would have to unevenly scale your 5K image down to 4K to print it, which would actually degrade the image as the math is uneven. (as an example, even scaling would mean that 4 pixels would be scaled down to 1). If there are laser printers that go beyond 4K then it's a non-issue, but until then you are probably better off shooting with the Red One as you will not have to scale the image when printing.

Bob

 

[James T Mather]

I'm not sure if I'm following your line of thought, or if you are being sarcastic, but to not view making a movie as a business venture is beyond foolish.

Bob

I think what jpp is saying is that most indy movies don't make money - as opposed say to, say, investing in property or selling building supplies. I'd have to agree with that idea. It's a risky investment and prone to more failures than successes.

 

[M Most]

Can all this be put in the contract?

Yes, if you're Vittorio Storaro. If you're not, it's a bit more difficult.

I mean,... I thought that the director and DP have joint power over what happens to the vision of the film?

If one of them doesn't agree = nothing happens.

The DP is the main specialist on the creation of vision, and so - should have the right of co-authorship and control over that aspect of the film.

Theory (and often common belief): What you said.

Reality: The amount of "power" a director of photography has is at the discretion of the director and the producers. Being a specialist doesn't grant you power. Everyone on a film set is a specialist. The ultimate creative decision making power on every aspect of a feature lies in the hands of the director, unless and until the producer or studio takes it away. Now, in television, it's a different story - the director is usually a gun for hire and has little to do with the final product once his cut is turned in, for which he or she gets 4 days with the editor. The power in television lies with the executive producer, who is often also the head writer/show creator. It's often said that features are a directors' medium, while television is a writer/producer's medium, and that's usually how the "power" is delineated.

 

[Jeff Coatney]

What we are in the midst of is the transition from Petro-Chemical Celluloid to Digital - Silicon based Imagers. So, in that respect, celluloid is doomed and it is very likely to happen within two years in the distribution/exhibition sector. It is already dead in the post-production sector, and celluloid is facing serious, rapid erosion in the production (acqusition) sector.

Those things that we most prize and mostly associate with "Film" are aesthetic characteristics. Those are well protected because we all care deeply about them, regardless of how we feel about the digital vs. celluloid debate.

Celluloid has a growing list of enemies that include the price of petroleum, transportation, environmental re-mediation, layer upon layer of needless labor-intensive and power-intensive processes to both make the rawstock and then to deliver a moving image.

Digital imagery is so throughly, so ruthlessly efficient at delivering a high-quality moving image, when compared to celluloid, it staggers the imagination. To produce the same amount of footage, a few ounces of silicon and gold replaces tons of mined minerals and chemicals that require vast resources and are not recoverable. There is no contest. Digital has already won, celluloid's life-span has reached its end. The final death-rattle may be a few years out, but there is no going back. The march of digital continues unabated and it cannot be stopped.

Dynamic range, rolling shutters and optical viewfinders may be important, but they will be solved very shortly and their current absence is not enough to render digital cameras unsuitable for full-scale studio production, a fact that has been proven over and over again.

The DP's that find themselves at the top of their career trajectory will best serve the industry, the audience and themselves by embracing the technology and influencing its development rather than fighting it. It's time to unplug celluloid's life-support system and let it go peacefully into history with the dignity and grace it deserves.

 

[Peter Majtan]

Finally (and I'm sure this will draw a lot of fire), I think Epic may be a flawed concept when looking at a current filmout workflow. My understanding is that there are two dominate resolutions of film printers in the industry, 2K and 4K. If you shot 5K a computer process would have to unevenly scale your 5K image down to 4K to print it, which would actually degrade the image as the math is uneven. (as an example, even scaling would mean that 4 pixels would be scaled down to 1). If there are laser printers that go beyond 4K then it's a non-issue, but until then you are probably better off shooting with the Red One as you will not have to scale the image when printing.

Bob

Scaling down from 5K-RAW to 4K-RGB will IMPROVE the image quality, not the other way round. You are not scaling down pixels, but an analogue sensor signal... I think that was the main reason for Epic to be 5K - to deliver pristine 4K images - same goes for Scarlet being 3K in order to deliver pristine 2K deliverables - and this was confirmed by RED-TEAM...

 

[David Mullen ASC]

Yes, I was about to post the same thing -- since you're talking about a conversion or transformation from a RAW Bayer-patterned monochrome image into RGB, you are doing so much interpolation and whatnot that it's not a simple mathematical scaling process, not like RGB to RGB, so having a few more pixels to convert RAW Bayer to RGB can't hurt.

 

[Tom Lowe]

David, I've been thinking about this RAW vs film color correcting issue. I realize you haven't really sat through the Manure post completely yet, and your opinions may change for better or worse after that is done, but generally speaking, here is my question:

If you shoot 35mm, the film is processed and scanned, and then you color correct with a 2K or 4K DI, aren't you basically working with a scanned film image that is more similar to a "burned in" JPEG file than to a RAW image?

Would it be fair to say that you would have more control over color, exposure, etc, working in a pure digital RAW workflow?

I know with stills, I have far, far more control over a RAW image than a scanned 35mm neg.

 

[David Mullen ASC]

No, just because it's a RAW image doesn't mean it contains more information to work with. Besides, a 2K or 4K scan isn't a compressed JPEG format, it's generally an uncompressed DPX file.

RAW isn't a superior way of color-correcting, it's just a way of storing a Bayer-pattern image. The only thing "good" about it is that it is basically right off of the sensor and its not compressed into a JPEG.

Besides, you have to convert RAW into color, i.e. RGB, just to view it and color-correct it. So there's no reason why a RAW recording is better than an RGB recording that doesn't need conversion to RGB.

You're comparing apples to oranges. Comparing a RAW digital camera recording to an RGB scan of film isn't the same thing as comparing a RAW recording from a digital camera to a JPEG recording from a digital camera.

The film negative holds more dynamic range, which gives you more color-correction options. If the RAW recording from the digital camera contained the same dynamic range as the film scan, then you'd be about equal in terms of flexibility, ignoring other factors like noise vs. grain, exposure problems, etc.

But RAW versus JPEG is not the same as RAW versus uncompressed RGB. RAW is just the result of not processing a Bayer image into color yet. Eventually you have to convert it to RGB, just that if you recorded RAW, you have all the information that the Bayer sensor could deliver. But you are still limited by what the camera could deliver.

And I'd point out that the RED camera records compressed RAW, which is almost a contradiction -- you aren't recording color JPEGS... but you are recording a RAW Bayer image with JPEG-2000-like compression.

There really isn't such a thing as as "pure RAW workflow" -- at some point, a RAW recording has to be converted out of RAW just to be viewable and to color-correct it. RAW is just the first step in the process, it's not the overall process. It doesn't stay in RAW other than the original files. You aren't creating a RAW project, you are creating a color project.

 

[Tom Lowe]

I didn't mean to make a direct comparison (I realize color film scans aren't compressed JPEGs), only a general one, in the sense that once you have a JPEG image, it seems like your ability to manipulate that image - in terms of color, exposure, etc - is limited, compared to what you can get away with straight off a RAW file - .CR2, for example. I'm wondering if the same is generally true for a developed color negative?

With film, you are really talking about two steps. First you have to chemically process the exposed negative (and hope that this is done to your liking) and then comes the scanning and eventual CCing. What if, for example, you accidently underexposed a shot by a large amount, and the lab did not bring up the exposure enough? Then you are stuck with a negative with limited information to work with, right? Or am I wrong about that? What the lab does with the negative makes a difference, right?

My point being, that with RAW, you are the lab. You control every single step along the way.

Anyway, thanks for always indulging me and my half-baked theories, David.

 

[James T Mather]

Unless something goes wrong or push/pull processing is requested, the lab's processing shouldn't make a difference to the neg density. That said, in eighteen years I've only seen the lab screw it once (eighteen years ago).

At the printing stage is where density of the final print is controlled (assuming a healthy neg) - generally higher figures result in a denser image - lower numbers in a thinner image. If you massively underexpose by accident then the next step is to see if it is contained within the latitude of the neg - if not then reshoot. Unless you tell them to force the film in this case (before processing) then there is nothing you can do about it - discovery after the fact is too late. The same could be said of an underexposed raw image though. The beginners rule of thumb on film, as I was taught it is expose for the shadows and print for the highlights (this wouldn't be true for digital since the highlight latitude on digital is not as wide as film). This, however, was before Darius Khondji and Harris Savides started a vogue for moody underexposed images.

In short the upside to digital possibly is that you can see it on a monitor in realtime - the only thing that could fool the eye is shooting in different conditions: looking at the monitor in bright sunlight vs. indoors. It's why a meter for narrative work is handy as you can note your ambient vs. keylight etc as hard numbers against your camera stop - so if you have to recreate a setup much later you can do so with a certain confidence in matching your lighting.