Digital vs Film as it stands right now

[Matthew W. Phillips]

Most of the neutral observers like David Mullen have stated that 35mm cine film is about 4- or 5K equivelent.

I'm sure David Mullen would also tell you that 1) The resolving power is related also with the way the footage was shot 2) It depends on the stock you use. 3) Stocks are getting better all the time just like digital is making advancements.

[Tom Lowe]

Matthew, are you a professional film scanner? Because the guy who wrote that post I quoted is. Do you want to argue the facts with him? Go debate him on his knowledge of scanning. Just sign up there and challenge him on the facts - if you are sure.

 

Are you saying Mullen and basically every other pro in the business are totally wrong and somehow film actually resolves at 8K and everyone else is crazy and has been ruining their films by working at 4k? Come on. Give me a break.

 

A lot of the guys who are enthusiastic about digital are often vilified as being "koolaid drinkers"... but in fact die hard film guys can be just as blindly religious in their beliefs.

 

If 35mm film can resolve at 8K, why doesn't IMAX scan 35mm features at 8K for their DMRs?

Edited June 20, 2008 by Tom Lowe

[Matthew W. Phillips]

Matthew, are you a professional film scanner? Because the guy who wrote that post I quoted is. Do you want to argue the facts with him? Go debate him on his knowledge of scanning. Just sign up there and challenge him on the facts - if you are sure.

 

Are you saying Mullen and basically every other pro in the business are totally wrong and somehow film actually resolves at 8K and everyone else is crazy and has been ruining their films by working at 4k? Come on. Give me a break.

 

A lot of the guys who are enthusiastic about digital are often vilified as being "koolaid drinkers"... but in fact die hard film guys can be just as blindly religious in their beliefs.

 

If 35mm film can resolve at 8K, why doesn't IMAX scan 35mm features at 8K for their DMRs?

 

Ahh. comeon Tom! You surely wouldn't resort to logical fallacies, would you? It surely sounds like you are using Appeal to Authority. I'm not saying anyone is crazy. A certain amount of this is subjective too, you must realize. Determining whether X image looks improved over Y image is going to be different depending on who you ask. Some people can't tell the difference between Bose speakers and Klipsch speakers so does that mean that they are the same? I think you would at least find that few people think there is any HARM in scanning higher, and many may feel that it has merit at least for having a higher INITIAL scan rate to down sample from.

[Tom Lowe]

Lol, well i don't think it's an appeal to authority when I mention that the entire film industry has decided that 4K is about the max scanning resolution of 35mm cine film! That's like saying I am making an appeal to authority because I mention that 90% or more of astronomers think that at least some other solar systems have orbiting planets... it's just the way it is!

 

I only mentioned David's name specifically because he is well known and respected here as a fairly neutral guy on the digital/chemical film debate.

 

I'm not against overscanning, but I am against claiming 35mm cine film is "8K" or anything close to 8K... 65mm is probably about 8K.

[Jason Anderson]

I know that your question was of a technical nature: that exactly is the problem, because these two technological aspects do not provide any substantial insight into a comparison between video and cine-film - which was your underlying question.

 

 

I humbly except your criticism of my view point. Thank you for your interest in this topic.

 

Your position is that cine-film and digital cannot or should not be compared on any level? I understand they both have aesthetic purposes that are very different, but they can be compared on some level.

[Michael Lehnert]

Tom, I do quite follow your argument, and also agree with David's quote you use, albeit in a somewhat decontextualised way and derived argumentation on your behalf. I would certainly not describe David as a neutral observer, though, as I believe he is more passionate than anyone else on this board.

 

I shall stick to my argument - that Matthew was kind enough to second me with, too, so that I am not all alone here scribbling my nonsense - that just as it was in the past, advances in scanning resolving power will move the perception of adequacy for cine-film into the higher numbers of xKs. It was already higher with Cineon, and will be once technology creeps back where it was with that system in 1993 already.

After all, David also argues in your quote that he - for a more untangible than scientifically explainable reason - "prefers the higher scan ppi as the prints on paper look more natural when the grain is visible and important for the expression".

 

So, as grain shape as basially a "crystaline content container" at the heart of shaping the image that is visually depicted by film and hence viewable for a human, why would more accurate grain shape not matter resp. be irrelevant or be treated as no real info, as those arguing for cine-film resolving power cappings do in effect propose - and quite arbitrary based on current digital technology maximum scanning levels and not on a mathematically-deducted intrinsic cine-film capability? After all, if a digital taste scanner has no grasp of 'salty', how would it be able to scan and declare the difference between sea water and fresh water (a crude example, I agree, but making the point, I think).

Where is the (ultimately artificial) border between regarding grain shape being of info and grain shape being of no further info, laid out? That's exactly my point. After all, some logically argue that the lower the resolution of the scan, the more "beautiful" 'cause "fuzzy" cine-film actually looks. Whereas video-measuring-scale inspired resolution junkies will go to 6-8K to make the realistic point and add a leaway premium, just as well as David did in the quote I used in my first post here, where he was adding two f-stops to cine-film latitude out of gut feeling against video's dynamic range.

 

After all, scientifically regarded, the maximum scanning possibility of cine-film would only be reached ? irrespective of the subjective info claim of "no added relevant info" ? when no further molecular resolution can be added to what is actually at the structural heart of the piece of film. Only then would the scanned copy be "close to identical" to the cine-film in all its structural organic (!) nuance - an argument you have to make around cine-film due to its nature, but cannot make with the non-organic geometry of a Bayer sensor.

 

One could of course argue that such a microscopic scanning would not depict any info a viewer would effectively see on a tv set, but what in respect to a very large cinema screen? And while we talk about the viewing culture: what can the human eye really see, outside the biologically-estimated resolution of eyesight? After all, humans "feel-hear" sound beyond 20k Hz border as well, as anyone who recently attended a live concert will know, but shun if he grew up with mediated music-listening only (through CD or worse, MP3). Once this 20k Hz border was the argument to articifically cap audio resolution for digital compression systems on must-be-commercially-viable products that are now rightly regarded as obsolete! Likewise, the optical capability of every human differs, so what eyesight effectively defines relevant viewable information when showing a film? I am sure young Matthew Buick will see more naturally than I can with my glasses on?

 

Scientifically - which it was this thread is centred around, for some reason - the scanned resolution achievable of cine-film is finite, alright, but far beyond the limited scanning capablities of digital technology, let alone digital cameras. One would first have to define what "relevant visual information" is, and trust me, that criterion was different when NTSC came out, and when HD came out, and it will be different when I have my rants with Geordie on the holodeck, soon... And I can tell you, that won't be filmed with either a 765 or a RED.

[Michael Lehnert]

Your position is that cine-film and digital cannot or should not be compared on any level?

 

No, I am not saying this at all! I am saying that when you set-out to compare two fundametally different means of image acquistition, like a piano and a syntheziser for sound creation, you must have an idea of

a) the universally applicable comparative sets of criteria (which resolution/latitude isn't alone)

B) acknowledge for your own arguments sake that the criteria you use (namely resolution/latitude) are being measured according to one of the two comparative subjects you look at (namely based on numerical results measured in Ks and dynamic range)

c) what other critera might be applicable to the other subject (like texture/tonality) and acknowledge that based on that perspective, how the underlying medium for comparison "B)" would fare (I would love to see someone making scientific measurements of texture and tonality for digital)

d) make a discoursive and reflected conclusion based on all steps a) to c)

e) accept that you will get serious critique when your publicly state to focus on only two critera that you also single-handedly declare as "only of importance" which are two in effect highly disputable aspects tilted theoretically highly in favour of one of the compared subjects (although even those can't reach cine-film, yet, but will, soon - although it will stiil not come close to the film look, which leads us back to cultivating the vide look, but this is a loop and I shall resist from repeating myself yet again)

 

I understand they both have aesthetic purposes that are very different, but they can be compared on some level.

 

Sure, as long as you don't get sloppy as you did in your original post. Intellectual laziness has killed millions of people and destroyed many good things in favour of crap.

 

Let's see: when I watch a projected cine-film, what I notice is colour depth. This is the first thing that gets to me. By that I don't mean colour reproduction in bit-values, I mean the three-dimensionality of the colour layers of the film! When you watch a film, you really start to delve into the colour depth. Technicolor was great, even ECN and EXR had it (no, I am not on Acid when going to the cinema! - wow, that Acid remark just dated me, I fear). Now based on that criterion (which I measure on a non-numerical scale of "flat" to "abyssal"), how could a RED with the latest firmware upgrade possibly fare in comparison to "abyssal" Kodak 5201? Elaborate while I am off to a full English breakfast with friends.

 

Cheers,

 

-Michael

[Matthew W. Phillips]

Lol, well i don't think it's an appeal to authority when I mention that the entire film industry has decided that 4K is about the max scanning resolution of 35mm cine film! That's like saying I am making an appeal to authority because I mention that 90% or more of astronomers think that at least some other solar systems have orbiting planets... it's just the way it is!

 

I don't care if Thomas Edison told you, it's Appeal to Authority to state that you are right just on the basis of some iconic reference without actually proving your case. So far what I gather is "I, Tom, am right because so and so agree with me." That is not reasoning...anyways, onward.

 

I agree with Michael that you are basing the effective resolution of film based off of when grain becomes visible which is not really accurate. Accuracy would be based off of when no further relevant information is available, also like Michael said. Can you prove that @ 4k, 35mm frame data becomes completely irrelevant? If so, how did you prove this? That is sortof akin to saying that digital cameras can reach a maximum resolution because the real life object information becomes seemingly redundant. The only limits are not film or sensors but human's own ability to discern between significant data. Remember, it doesn't matter who says what...in the end all that matters is what the image looks like.

[Michael Lehnert]

Lol, well i don't think it's an appeal to authority when I mention that the entire film industry has decided that 4K is about the max scanning resolution of 35mm cine film!

 

I think you overstate your case with this formulation, Tom, because you are too closely involved with the entire film industry.

 

More professional would be the following formulation:

4K is regarded as the current and short- to medium term industry standard for 35mm because it is the threshold that is technologically achievable, technically reasonable, and economically viable, and thus financially payable by all stakeholders involved in producing film projects of various scale for output on both cinema screens and television sets (incl. future-proofing for HD and HD+).

 

I am sure you will agree that industry standards change at an exciting pace, especially when they are electronics-based. As I said: we've been there, we'll go there again.

 

Also: never argue with astronomy or rather astrophysics. No science had to fundamentally change the worldview on the universe we live in more often over the past 400 years than this profession (and I am a passionate amateur of astrophysics).

[Matthew W. Phillips]

Also: never argue with astronomy or rather astrophysics. No science had to fundamentally change the worldview on the universe we live in more often over the past 400 years than this profession (and I am a passionate amateur of astrophysics).

 

Damnit Michael, I was going to say something along those lines!

 

Might I also add that there was a time when ALL of the scientists thought the Earth was flat. So much for authority, eh?

[DJ Joofa]

Can you prove that @ 4k, 35mm frame data becomes completely irrelevant? If so, how did you prove this?

 

No, at 4k it does not become irrelevant, however, please refer to the following SMPTE journal paper to find out that there is little benefit in going over 3K scans of typical film negatives:

 

B. Hunt, et. al. "High resolution electronic imaging system for motion picture film", SMPTE Journal, March 1991.

 

Stephen Williams (and some others on this forum) objected to me when I mentioned this before saying that study is over a decade old and film stocks have had an opportunity to improve. Fair enough. However, I went back to Glenn Kennel's book from 2007 (BTW, Kennel was also a co-author on the above mentioned paper), and he says:

 

"Original negative films improved substantially in the last 15 years, now exhibiting higher speed and lower grain. However, the MTF has not improved much, and typical camera lenses are the same, and today's CCD film scanner and laser film recorders have similar MTF characteristics as the original Cineon scanners and recorders."

 

The only limits are not film or sensors but human's own ability to discern between significant data. Remember, it doesn't matter who says what...in the end all that matters is what the image looks like.

 

Human visual acuity is not more than 60 cycles per degrees at the fovea (not to be confused with 60 cycles/mm or 60 lp/mm). In a theater, only one distance from the screen will have the best visual response and even that will not afford a resolution near 4K according to some studies.

Edited June 20, 2008 by DJ Joofa

[Jason Anderson]

No, I am not saying this at all! I am saying that when you set-out to compare two fundametally different means of image acquistition, like a piano and a syntheziser for sound creation, you must have an idea of

a) the universally applicable comparative sets of criteria (which resolution/latitude isn't alone)

 

Please compare film and digital using the universally applicable comparative sets of criteria. Please help me out in elaborating on just what that criteria should be as I have already mentioned. I suggested and was careful to place a question mark, that resolution and Latitude should be included in this list. You feel they should not be included in the list, I beg to know why sir.

 

Jason

[georg lamshöft]

I think it's a sad thing when technology compromises our work, that's why I think that - whenever possible - we should use the best equipment available. I would have loved to see "Juno" or there will be blood" in 65mm...

 

I can only speak for my still photography experience but comparing resolution of film and digital is very difficult. Digital seems to be able to capture high resolution detail with low contrast while film captures even higher resolution (frequencies) in high-contrast situations.

You need very sophisticated scanning technologies and you have to scan at very high resolutions (>>4000ppi) to avoid "grain alaising". I scan Velvia slides at 6000-8000ppi and the results are great, you have to use noise/grain filters and sharpening tools to get the best results. But even then 8000ppi (90MPixels with 35mm!) don't generate 90MPixels of resolution, more like 15-20MPixels - nevertheless it's important to scan with these high resolutions (I think it's called oversampling).

 

There is a very interesting scientific article:

http://www.arri.de/infodown/cam/broch/2008...%20Brochure.pdf

It's from Arri, but they are about engineering not marketing (they don't even claim that their D-21 despite 6MP-Sensor has 2k - just 1080p)... :-)

The article claims that up to 200ASA film stocks you can achieve 4k REAL resolution with 35mm (oversampled from 6k) with technology available for big productions. Bad scanning, marketing hype etc. "forced" many people in the still photography world to shoot digital even when they don't profit from it's advantages...

I hope the film industry is more clever and more about achieving the best possible quality in the final product and not about marketing or cutting costs...

[Michael Lehnert]

Thanks, Georg, for the elaboration and actual contribution. That was very insightful indeed!

 

I suggested and was careful to place a question mark, that resolution and Latitude should be included in this list. You feel they should not be included in the list, I beg to know why sir.

 

If you had actually read and quote-posted my follow-up points from B) on, you would have anwered your question yourself as the methodology I suggested for your problem is clearly outlined for your delectation!

 

I did not suggest that resolution and latitude should NOT be included in the list or utterly omitted! My entire posting in this thread is about the promotion of reflectivity on this poisonous subject that is primarily driven by corporate interests and not only be technological advancement, which becomes particularly obvious when looking at our current debates from a historic perspectives some here actually lived through!

 

I pointed out that your original premise of focusing on these two aspects alone - which is normally done due to videographic bias in criteria setting and comparative assessments - and declaring them to be the two important ones, cannot be upheld if you want to avoid criticism for a flawed approach that is essentially telling you nothing about digital versus film outside the video-criteria-heavy discourses that get even intrinsically eroded in marketing blurbs of "real HD", "true HD" and "HD so good, it's better than HD!".

I suggested alternative cine-film-centred and also artistic criteria and also elaborated on the circumstance that there are cine-biased criteria under which video will always loose out against cine, while the vice-versa situaton is also easily constructed by following marketing brochure stream of thoughts.

You really got a universtiy lecture for free on how to learn to ask the right questions to get meaningful anwers, on the mechanics of discourse construction, aesthetics and cine-tech, mate, and I don't know how to satisfy your curiosity further if you don't want to engage with the points I threw at you for yourself.

 

 

"Original negative films improved substantially in the last 15 years, now exhibiting higher speed and lower grain. However, the MTF has not improved much, and typical camera lenses are the same, and today's CCD film scanner and laser film recorders have similar MTF characteristics as the original Cineon scanners and recorders."

 

Wow, this got past a peer-reviewing process and went into publication? :blink:

Don't let Keith Walters read that, he will be all over you again...

 

 

 

P.S.: sorry for all the typos, this is written on an iPhone form the field... I quite proud to have mastered that much text B)

[Walter Graff]

There is more than just latitude and resolution in terms of the perception of quality. And those factors (random grain, weave, physical layers of color, etc) are the reasons that even with the best spec'ed video cameras, people still choose the aesthetic look of film.

[Nate Downes]

No offense, but I think these numbers are off. After 4K or 5K you are mainly just sharpening grain.

From Raw digital you're sharpening, interpolating, and noise reducting, so why shouldn't film be allowed the same treatment? Otherwise, your arguement is "My car is faster than your car which is out of gas and doesn't have any tires on it."

 

Allow each the same or sit down and be quiet.

[Annie Wengenroth]

Okay, I couldn't resist...

 

The Film Vs. Digital Drinking Game:

 

Take a drink...

 

1. Every time somebody mentions resolution and how a "x"k scan "holds up just as well as film". Because, you know, you might as well "prove" that the sky is blue.

 

2. Any mention of the Red, Genesis, or HVX w/ 35mm adapter. If in conjunction with David Lynch or other "experimental" directors/projects, take two drinks.

 

3. Bonus shot if the phrase "apples and oranges" is used more than once.

 

4. If debating this with a first-year film student (no offense, guys, you know I love ya!) who just took Intro to Video, it's time to line up 3 Jager Bombs and knock em back while loudly proclaiming, "16MM WILL NEVER DIE! GOD BLESS THE BOLEX!" Then set a camcorder on fire!

 

5. If debating this with someone who works in a film processing/transfer lab, offer to buy the whole round because hey, in another 5 years, they might not have a job, right? I mean, if this is the way things are going? :P

 

6. And if you're talking to me, take a drink every time I either say the f-word or make some sort of grandiose statement about how when I have kids, I Want Them To Know What Film IS, Dammit!

 

7. For every citation of 1) the latest American Cinematographer article, 2) a brochure you saw at NAB, 3) that guy you talked to in the sales department of Panasonic.

 

8. Bonus rounds if the following artists are mentioned: Monet, Cezanne, Seurat, Jasper Johns.

 

....That's really all I have to say about this... ;)

[John Brawley]

So the two factors are resolution and latitude? It seems digital is very close to the resolution of film, the new Epic 5k should surpass film. So is it safe to say all we are waiting for is the contrast ratio. The RED one claims that its chip will hold 10 stops, does someone know how these 10 stops compare to the latitude of film. Perhaps the Epic will hold more stop than the RED one.

 

 

 

Colour Gamut. The actual range of colour available. It's far far beyond. Latitude of course.

 

The resolution argument is a false one i believe. Yeah i know you can quote MTF charts and figure in theatres. But the fact is that 2k and now 4k exist because of economics, not quality. Audiences, because of american / hollywood cultural dominance in theatres, have become *used* to seeing DI processed films. I saw a beautiful film that had the most amzing cinematography i have seen for a long time at the Sydney Film festival a week ago.

 

http://www.imdb.com/title/tt0841925/

 

35mm anamorphic shot on Lomo lenses. SHot on 35-3's and IIc's. I was weeping at it's beauty.

 

And i think it's REALLY important to distinguish between visual perception limits and what the Digital process works with. And that's because, I can't see all those colours or differences between them, or the difference between 2k and 4K but once you start *working* with those images in a post environment, you'll notice the difference pretty quick. I can't percieve the differences UNTIL i want to manipulate them.

 

These stupid pissing matches really do little to further the craft of cinematography.

 

jb

[Tom Lowe]

LMAO, Annie! Nice one.

 

BTW, for people saying, "Well, are you arguing that 4K is the limit because at 4K you can see the grain?" No, actually, at 2K you can see the grain! 4K is probably being charitable to film, unless it's a very good, new stock and well-exposed.

[Nate Downes]

LMAO, Annie! Nice one.

 

BTW, for people saying, "Well, are you arguing that 4K is the limit because at 4K you can see the grain?" No, actually, at 2K you can see the grain! 4K is probably being charitable to film, unless it's a very good, new stock and well-exposed.

And you can see digital noise at VGA resolutions, what's your point?

[Robert Houllahan]

5. If debating this with someone who works in a film processing/transfer lab, offer to buy the whole round because hey, in another 5 years, they might not have a job, right? I mean, if this is the way things are going? :P

 

 

Oh crap, only five more years left? and by then I'll be forty and I'll have to go back to drug trafficking and whacking people for the asian triads. Please keep shooting film!!! Oh the humanity! you don't know what you will be releasing on an unwitting public if all the people in the film labs don't have jobs! They will put it down in the history books as the great Metol Digeratis crime wave of 2013.. I think at least one of our guys is capable of eating fresh babies and from the looks of his avatar that Aussie guy Case is capable of just about anything.....

 

-Rob- Makes a Pagan sacrifice to the great emulsive god every night" Houllahan

 

And the Film vs. Digital debate is just like watching paint dry, yes really that exciting....

[Nate Downes]

And the Film vs. Digital debate is just like watching paint dry, yes really that exciting....

You give insult to paint drying watchers everywhere. Paint Drying is far more exciting!

 

Now Lenolium Curling, that's more comparible....

[Nate Downes]

Colour Gamut.

You insult us B&W fans....

 

j/k!

 

You know, this whole arguement is dumb dumb dumb!

 

I'd pick digital for a production when it makes sence for it. Not because of some resolution perception, or some fantastical numbers, but when it makes sence. I strive for looks, not for technical specs. Is my audience going to be sitting there going "wow, look at the 4k! Look at that color space! Isn't that CPU power amazing? How about that lattitude!"

 

It's about time this silly little "war" was called off, because, frankly, it makes us all look silly. Shoot on digital, shoot on film, hell, shoot on pixelvision for all I care, just shoot!

[Tom Lowe]

IMO, over the last ten years or so, it has mainly been the Digital camp that has been responsible for the flame wars on this subject, because they were constantly overstating what digital could do, and carelessly predicting the immediate demise of film, with no solid evidence to back up their claims. In the last year or two, especially since the rise of RED and Genesis, I have noticed the tide shifting. It is now the chemical film guys who are making a lot of the Koolaid-drenched statements and trying to play down digital cinema, regardless of the actual evidence at hand. 35mm is 8K? How much does this remind us of the outlandish early claims of the Digital koolaid drinkers?

[Nate Downes]

IMO, over the last ten years or so, it has mainly been the Digital camp that has been responsible for the flame wars on this subject, because they were constantly overstating what digital could do, and carelessly predicting the immediate demise of film, with no solid evidence to back up their claims. In the last year or two, especially since the rise of RED and Genesis, I have noticed the tide shifting. It is now the chemical film guys who are making a lot of the Koolaid-drenched statements and trying to play down digital cinema, regardless of the actual evidence at hand. 35mm is 8K? How much does this remind us of the outlandish early claims of the Digital koolaid drinkers?

I don't know, I still see the flames coming from one direction. Tom, you still haven't responded to direct comments at you above, yet you keep spitting out newer comments to try and support your position. So, please, respond to them else be listed as a troll and ignored.