Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Epic vs Film - the Decisive Battle is looming

But I mean, it's not the DP or director sitting there at the lab pouring chemicals onto the negative and deciding how long to leave it sitting in the chemicals, right?
 
Tom said:
What if, for example, you accidently underexposed a shot by a large amount, ]
Click to expand...

Hi Tom,

Very simple, the DOP gets fired.

Stephen
 
What Steven said :biggrin:





In the lab it's a technician but it rarely has problems these days, I find, if you go to a good lab. Processing is a pretty mundane job which is highly automated in general.

But yes, you are handing over your neg to someone.

The only things I've seen are dirty neg and scratches or chemicals left on the film which can cause the odd flicker but (1) not for about 10 years and (2) wet gate and a clean usually got rid of them then.
 
In the modern DI workflow what really matters is the process of scanning the negative to master digital files. It is at this stage where skilled operator can "pull" the most info out of the neg by calibrating his range, frame and exposure. This is almost the equivalent of what can You do with RAW. The only difference being that if You get it wrong with film - You have to re-scan the neg... :)
 
Peter Majtan said:
In the modern DI workflow what really matters is the process of scanning the negative to master digital files. It is at this stage where skilled operator can "pull" the most info out of the neg by calibrating his range, frame and exposure. This is almost the equivalent of what can You do with RAW. The only difference being that if You get it wrong with film - You have to re-scan the neg... :)
Click to expand...

Hi,

It's very important that different emulsions have seperate set ups.

Stephen
 
Jeff Coatney said:
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.
Click to expand...

What makes you think they aren't?
 
Tom said:
But I mean, it's not the DP or director sitting there at the lab pouring chemicals onto the negative and deciding how long to leave it sitting in the chemicals, right?
Click to expand...

Nobody "decides" these things based on personal taste. Professional film labs operate on standards. This is very different than in the digital world, where there seem to be very few standards (or too many "standards," depends on how you look at it). Lab work is not creative, it is scientific. Calibration procedures for DI's work because they're based on standards, developed by Kodak a number of years ago, that define digital levels and their resultant negative densities. Scanners are set up based on these standards. Film recorders are set up based on these standards. And printing lights, while a bit different in each lab, are determined (at least for DI work) by what light values produce specific print densities on a LAD image, usually using 1.09, 1.06, and 1.03 as the basic standard (although some DI providers like to make the prints a bit darker, so those aim values will be specified a bit higher than that). What I'm trying to say is that unlike the digital world, where the approach is basically "do what you want as long as it looks good", the film world has standards that all major processes are based on. Developing a negative at just about any reputable lab in the world will generally produce a very similar result. There is very little of the guesswork you seem to think is prevalent.
 
Peter Majtan said:
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... :)
Click to expand...

So that would be setting up the 5K camera to scale the analog signal stream from the 5K chip (A to D converted) into a 4K image? You would then be in the same 4K workflow just with a qualitatively better image?

Bob
 
It would be a 5K RAW digital recording -- the sensor would be a 5K Bayer-filtered sensor, so the RAW signal that is converted from analog to digital would also be 5K.

Then you'd have a choice as to what resolution to convert it to when transforming from RAW to RGB. Again, remember that a 4K RGB signal is 4K per color channel, whereas a RAW file is a single 4K monochrome file with a Bayer pattern that has to be decoded and converted into three files for red, green, and blue. The 5K RAW file would have information from 2500 photosites filtered green, 1250 photosites filtered red, and 1250 blue (not exactly since it may not be exactly 5K) -- from that information, you have to create an equal number of pixels per red, green, and blue. So during that transformation using a "smart" debayering algorithm that attempts to interpret the information to know what the surrounding color information near that photosite should be, you have some flexibility to create new RGB files in different resolutions.

But in terms of measurable line resolution, the RGB image that results from a RAW conversion tends to be around 75% of the original monochrome resolution, so hopefully a 5K Bayer sensor would deliver a measurable 4K line resolution. But that depends on a host of other real-world variables, from the strength of the optical low pass filtering, the quality of the debayering, and the quality of the lens on the camera, etc.

A "dumb" conversion algorithm would just involve scaling pixels -- you'd take the 2500 green-filtered photosites and scale those green pixels to 5K, let's say, but then you'd be scaling the 1250 red and 1250 blue pixels to 5K. But that's not how most debayering is done.
 
I think film won't die without a fight. Sure, Epic may render 35mm film obselete in the filmmaking process, but motion picture film could always evolve. The reason why DSLR overthrew film SLR cameras was because still pictures are more limited in size. Who really wants to look at a 6 story picture? Motion Picture film on the other hand, has the advantage of becoming more grand and amazing. Hence the iMax theater. Film could always get bigger and better along with digital cinema. And until Red develops the Godzilla sensor that can handle 10K on the iRed camera, film will still have a home.
 
The good and bad thing about film is that more quality (essentially more "data") can easily be achieved just by increasing the size of the negative... but at considerable cost, and perhaps physical awkwardness. But the technology to do this has existed for decades. 5-perf 65mm cameras gather dust at some rental houses as we speak.

Same goes for film projection -- you want more quality, then use a bigger print size, or shoot and project at higher frame rates.

But the momentum is all on the side of digital.

One aspect I find interesting though is that when you look at individual frames, most tests show digital stills to have more resolution than film stills, yet when you project these images, often film looks better, at least, better than it did as a single frame. This is because the "data" on a piece of film is in different positions on each frame, due to the random nature of grain, so as you project the image fast enough, there is an averaging effect of multiple frames in your eye-brain that seems to increase the perception of detail while reducing grain. Digital camera images don't quite benefit as much from this effect due to the fixed nature of pixels.

But I think film is ultimately doomed to being phased out of existence, except for perhaps a boutique market. But we're only haggling over the time it will take to happen. I think two years is too short of a time for such a major flipflop in Hollywood production. Maybe five to seven years.
 
David, if you had to, what year would you predict for the benchmark I mentioned earlier - 51% of major studio pictures shooting digital? My prediction is Dec 31, 2010.

I'm not sure exactly how you'd measure this. Maybe IMDB techinical specs, but then you also run into the problem of deciding exactly what qualifies as "major studio picture." And sometimes tech info is not listed on IMDB right away.
 
David Mullen ASC said:
One aspect I find interesting though is that when you look at individual frames, most tests show digital stills to have more resolution than film stills, yet when you project these images, often film looks better, at least, better than it did as a single frame. This is because the "data" on a piece of film is in different positions on each frame, due to the random nature of grain, so as you project the image fast enough, there is an averaging effect of multiple frames in your eye-brain that seems to increase the perception of detail while reducing grain. Digital camera images don't quite benefit as much from this effect due to the fixed nature of pixels.
Click to expand...

I think I've experienced this myself and I would agree. The human brain has an amazing capacity to combine detail over hundreds of milliseconds.
 
Remembering of course:

That you can't get the rated resolution from any sensor of any kind unless you also allow for high levels of aliasing artifacts. RED, nor any other camera manufacturer is immune from this. At RED, we sensibly start with a very high resolution sensor so that we can use proper optical low pass filtering, and still ensure lots of resolution and detail in the finished result. This is not always the case with other cameras.

Graeme
 
Graeme

Would it be possible to build a camera without an OLPF if the sensor clearly out resolved all the glass that could be used on it? Since lens softness is not the same as a OLPF what would the 'out resolving factor' have to be?

just curious

Michael Lindsay

PS I'm personally glad you went with a fairly strong OLPF..
 
Yes, if you can ensure the MTF of the lens is such that the sensor is not provoked into aliasing, then an extra OLPF would not be needed. All that matters is that detail higher than the sensor can cope with be properly attenuated enough that nasties don't occur.

Graeme
 
Unnatural assumptions

Unnatural assumptions

The world currently assumes everything digital is better when in fact it is not. Digital has well known artefacts including current specs like MPEG4 AVC / H.264 / JPEG 2000 etc.
I take many digital still photograpghs and quite frankly digital B&W photography is miles off of chemical so while DSLRs may well be perfect for news, paps etc for quality exhibition many exhibitors still use B&W film developed traditionally.Secondly a new generation are going back to spinning discs rather than CDs or i-tunes because they preffer the analogue sound.
What may kill film for theatrical is economics if both Kodak & Fuji fall below a certain threashold it will be unviable to simply supply for high end features so digital would win by default.
The whole digital debate and particularly from evangelists on the reduser pages misses the point about choice and using a medium in a certain manner to tell the story they are the same people who assume (wrongly) you can fix everything in post. Think of film & digital as oils & water colors are you going to tell artists to stop using oils?
David Mullen said the film workflow up to the finished product was messy, read the article from Gavin Finny in British Cinematographer issue 28, digital is messy Red has its code, Sonys its, Panasonic its, Phantom its, no common platform until DPX files whereas 35mm film will go in any 35mm camera. Anamorphic on HD is barely represented (Arri D21) and anyway looks nothing like anamorphic on film. Still digital has allowed all our hard work to be watched on a 2" screen on an i-pod where all that 4K detail makes a mockery of all the boasting people regularly do. Is digital the future yes, is it better or superior to film? depends what your definition of better or superior are. Unfortunately boffins dont understand this they simply get locked into technical chest puffing instead the losers are thoses only asking for reasonable choices which we should embrace and accept as normal.
 
There are, of course, plenty of fine art photographers shooting digital, be it on high end DSLRs or medium format backs, and they get superb results for black and white or colour. DSLRs are not just for news etc, and are not just used for that.

I firmly believe that there will be increasing choice in camera technology be it digital or chemical based. I still listen to vinyl, on tube amplifiers (no transistors) because I like that. People will still project film because they like that. A friend of mine has one of the largest private collections of 16mm film that I know of. But although he watches film all the time, he still projects from digital sources also.

When it comes to aesthetics, there is no "better", just "preferable". All we're doing at RED is producing the best digital acquisition we know how to.

Graeme
 
You must log in or register to reply here.
Back
Top