Eastman Kodak: Strong increasing demand for movie film

[Luckless]

You'd better have a definitive source for this claim. Reality is that the QC is likely tougher for still than motion picture - at 24/25 FPS a small flaw is long gone before you notice it.

That is actually an interesting point. Does anyone have any insight on if motion film is given a wider margin than their still film lines for quality control?

Personally I would have guessed they were using the same gear and standards for all their film, but I also wouldn't have expected much in the way of flaws from mid-roll sections once those coating rigs get up to speed.

 

[MattKing]

I would assume that the two different types of emulsions and substrate + anti-halation combinations and sprocket differences are so different that the quality control specifications would be particular to the films themselves.
If you were to compare the two types of Ektachrome stock - still and motion picture - thespecifications would be much more comparable.

 

[Lachlan Young]

That is actually an interesting point. Does anyone have any insight on if motion film is given a wider margin than their still film lines for quality control?

Personally I would have guessed they were using the same gear and standards for all their film, but I also wouldn't have expected much in the way of flaws from mid-roll sections once those coating rigs get up to speed.

Purely anecdotally, I've seen comments about 5222 having occasional tiny flaws from people who have used it for still uses - but it might also be due to end users not cutting meticulously and loading cleanly into cassettes. There are likely specific acceptance criteria for imperfection sizes of various sorts during coating (and probably more importantly, converting) - I don't know if they are different between still and motion.

 

[cmacd123]

Since a defect would flash on a 40 ft screen and distract the audience, Plus that many shots represent big money to stage for movies, and if they blow up or burn down a set, their is no second chance, I would expect that defects would be taken MUCH more seriously on movie stock.

the material cost for Movie film might also be higher as it needs the REM-JET back coating. with a slight relief as 400ft at a time are put in a black bag and a can, rather than spooling 5 and a hlf feet at a time into multi part cassettes/cans and boxes.

 

[Kino]

At work I use about a Million feet of Kodak 2234, 2366 and 2302 a year and can't remember a stock or coating defect in the last two decades.

Kodak motion picture quality control is second to none.

 

[138S]

Reality is that the QC is likely tougher for still than motion picture - at 24/25 FPS a small flaw is long gone before you notice it.

Tell this to Dan Mindel:



https://theasc.com/news/mindel-named-2019-kodak-cinematographer-in-residence-at-ucla

Recently he cinematographed a $275 million budget production with VISION3, if a little pin is found in a frame then an entire truckload of film can be returned to Kodak.

VISION3 is a top notch product with top notch QC, in fact Portra uses technologies developed for VISION3.

Also important, most of the QC investment is Off-Line type, rather than inspection type. A great effort is dedicated in the process design to ensure quality in every detail.

VISION3 is slightly more expensive to manufacture than Portra because it includes the remjet.

 

[Lachlan Young]

VISION3 is slightly more expensive to manufacture than Portra because it includes the remjet.

Unlikely. Remjet is a very simple anti-hal/ lubrication solution that's at least as old as Kodachrome & much easier to implement than an integral layer that has to clear during processing. If anything it reduces both manufacturing and design costs.

Quite apart from the fact (as pointed out earlier) cinema film needs both simpler packaging and less of it (400/ 1000ft vs 4-5ft) - the most expensive parts of 135 and 120 are the cores/ paper/ canisters/ packaging - as per Ilford and others on here over the years. If you think about it 400ft of 5207 may require as little as a couple of percent of the packaging as 80 rolls of 135 Portra 400.

The main thing I was aiming at is that cinema and still films are going to have slightly different aim quality markers because of the different requirements for the end result. For all we know, Vision 3 may actually be more profitable than Portra because it might be cheaper to manufacture in very big quantities.

 

[138S]

Unlikely. Remjet is a very simple anti-hal/ lubrication solution that's at least as old as Kodachrome & much easier to implement than an integral layer that has to clear during processing. If anything it reduces both manufacturing and design costs.

Quite apart from the fact (as pointed out earlier) cinema film needs both simpler packaging and less of it (400/ 1000ft vs 4-5ft) - the most expensive parts of 135 and 120 are the cores/ paper/ canisters/ packaging - as per Ilford and others on here over the years. If you think about it 400ft of 5207 may require as little as a couple of percent of the packaging as 80 rolls of 135 Portra 400.

The main thing I was aiming at is that cinema and still films are going to have slightly different aim quality markers because of the different requirements for the end result. For all we know, Vision 3 may actually be more profitable than Portra because it might be cheaper to manufacture in very big quantities.

You can buy 200ft rolls of VISION, say 40 rolls, at some $2.4 per 1m.

Film price is not based in ex-factory cost, but in marketing strategy, this is an oligopoly and general competition rules do not apply.

Kodak could sell Portra in 100 or 200ft cans at same price than VISION, but they won't do it.

Of course they are free to do what they want, but VISION3 retail price says a lot.

 

[Lachlan Young]

You can buy 200ft rolls of VISION, say 40 rolls, at some $2.4 per 1m.

Film price is not based in ex-factory cost, but in marketing strategy, this is an oligopoly and general competition rules do not apply.

Kodak could sell Portra in 100 or 200ft cans at same price than VISION, but they won't do it.

Of course they are free to do what they want, but VISION3 retail price says a lot.

If someone could make a better product cheaper, they would. That Kodak had an effective monopoly until well into the 80's on cinema camera film seems to be an inconvenient reality that your agenda wants to gloss over. Quite apart from the fact that ordering 1600 rolls of 120 or 135 Portra from Kodak probably would get you a very decent discount. The cinema camera film competitors either failed because they either over diversified while the market was changing (Fuji's at times incoherent panoply of choice) or didn't have products that had enough overall strengths to compete (Agfa) outwith specific aesthetics. Nobody owes you cheap film. If Portra matters to you, find ways to afford it. I'd rather have Portra 800 than not have Portra 800.

 

[semi-ambivalent]

Purely anecdotally, I've seen comments about 5222 having occasional tiny flaws from people who have used it for still uses - but it might also be due to end users not cutting meticulously and loading cleanly into cassettes. There are likely specific acceptance criteria for imperfection sizes of various sorts during coating (and probably more importantly, converting) - I don't know if they are different between still and motion.

Stopped lurking to say I use 5222 almost exclusively and I do find 'little chunks' from time to time. Just like Tom Abrahamsson said I would. In the grain focuser they're easily discerned from things on the surfaces of the film but I like to spot so it's not bad up to 11x14-ish. Denver is dry and static attraction is always a problem. I recently shot Union Pacific 4014 in Colorado on its way to Cheyenne and of the three frames I got off in the few seconds I had two had lighter faint irregular ovoid circles on the prints. They're visible (in the emulsion or the base) as darker patterns. Almost as if a bubble in the coated emulsion had popped but not completely leveled before gelling. I've considered going back to Tri-X since I have a lot of that in the freezer but I like 5222's look, and I have a lot of that too. I just embrace the non-sterility of the process. Nobody else seems to care either. hth

 

[138S]

If Portra matters to you, find ways to afford it.

I have the way to afford it: renouncing to shot LF Portra sheets and using a 120 roll fim back in substitution, and shooting way less color film, 1/2 of what I was shooting.

In the EU 4x5" Portra is $7 per sheet, and $28 per sheet in the 8x10" case... For the standard of living in my location this is simply too much, not only for me, others I know made the same decisions.


...but VISION3 short film ends are $0.56/FT , and Vision 50D is superior in many ways to Portra, so it's about exploiting its particular spectral response to get good results with it. It's better to develop an "Acquired taste" for what one can afford !!

(http://mononoawarefilm.com/film-stock)

 

[Lachlan Young]

and Vision 50D is superior in many ways to Portra, so it's about exploiting its particular spectral response to get good results with it.

Really?
Tell us how.

I can think of a long list of why 5203 isn't superior as a still imaging film. And yes, I've used it. For starters all Portras are drastically sharper in areas that are perceptually important in still imaging.

 

[Kino]

Really?
Tell us how.

I can think of a long list of why 5203 isn't superior as a still imaging film. And yes, I've used it. For starters all Portras are drastically sharper in areas that are perceptually important in still imaging.

This is in reply/reference to what post?

 

[Lachlan Young]

This is in reply/reference to what post?

#136 - the claim that Kodak 5203 is somehow 'superior' to the Portra films for still imaging, nothing to do with motion picture capture!

 

[138S]

Really?
Tell "us" how.

(Are you several people? why do you say "us"??)

Well, this is the 50 the graph:



If you see the "0", it is 5 stops far from speed point, so as an starting point they suggest overexposing it by two stops... but there is more... even when overexposing the 50D by 2 stops you still can overpose 8 additional stops and still conserving texture, not a joke, see the graph.

Regarding colors, it is designed to be copied on film, not on RA-4, so regular color inversion based on RA-4 nature may not do a perfect job, so it requires (and deserves) a 3D LUT Creator job.

I like a lot the V3 spectral signature, but I found I need 3D LUT job to suit my taste.

so.... regarding your question about how to exploit the V3 spectral footprint, here there is the answer: A 3D LUT to overcome RA-4 vs Copy film inversion missmatch.

I can think of a long list of why 5203 isn't superior as a still imaging film. And yes, I've used it. For starters all Portras are drastically sharper in areas that are perceptually important in still imaging.

The 50D is damn sharp, also see the granularity vs exposure graph in the datasheet. Single thing you have to do with V3 is not underexposing it.

If you don't get perfect sharpness then overexpose it more, this stuff likes light, don't worry, you are not to toast it easily.

 

[cmacd123]

market was changing (Fuji's at times incoherent panoply of choice) or didn't have products that had enough overall strengths to compete (Agfa) outwith specific aesthetics.

for as long as they could Fuji was in an all out war with Kodak for market share in the movie market. that is quite likely why one can buy "Vision 3" as both vision and vision 2 were outflanked by various Eterna products. Only when Fuji backed off when no one was managing to be profitable on Movie stock, (when the print film market dried up overnight) did Kodak become a monopoly.

Agfa mostly marketed in Europe, and they probably were even more dependent of Print film sales. I do remember the Agfa Movie negative and much prefered the more laid back colour balance.

 

[Kino]

Yes, when the studios were still making release prints, Kodak could have given the color negative stock away for free and still made a enormous profit. The volume of sales of release print stock was a HUGE cash cow for Kodak. When that shrank to a tiny percentage of its former glory with digital distribution and acquisition, it really hurt Kodak's bottom line.

 

[Lachlan Young]

If you see the "0", it is 5 stops far from speed point, so as an starting point they suggest overexposing it by two stops... but there is more... even when overexposing the 50D by 2 stops you still can overpose 8 additional stops and still conserving texture, not a joke, see the graph.

It would help if you looked more carefully at it. The '0' represents the midtone - 3 stops up from the start of the straight line and 4 stops below the shoulder. 7 stops of recorded range on the straight line. As it should be.

Regarding colors, it is designed to be copied on film, not on RA-4, so regular color inversion based on RA-4 nature may not do a perfect job, so it requires (and deserves) a 3D LUT Creator job.

I like a lot the V3 spectral signature, but I found I need 3D LUT job to suit my taste.

so.... regarding your question about how to exploit the V3 spectral footprint, here there is the answer: A 3D LUT to overcome RA-4 vs Copy film inversion missmatch.

I know which Vision 3 films do have a bit of the Portra look to them (if you have some actual familiarity with the cinema films) - and I think you are grossly overestimating how much 50D is used vs certain of the other films that do have similar 30 & 50% MTF responses (which affects colour rendering) to Portra 160 and 800 - and that decent colour timing can fool your eyes quite significantly. The cinema films don't have the same 100%+ large object edge response as Portra however - In still usage the cinema films won't be as visibly acute/ sharp in areas known to be essential for the appearance of 'sharp' still imagery. A LUT won't really improve any of this - sure you can sort of mangle colours to match, but the tone and spectral curves get in the way when working with film origination rather than a low contrast linearised sensor output. What a LUT can do very well is make something look like a much better version of itself and/ or a legitimate print - and indeed that's essentially how they are used to model the behaviour of output media for cinema usage.

The 50D is damn sharp, also see the granularity vs exposure graph in the datasheet.

You're looking at the wrong chart for sharpness.

for as long as they could Fuji was in an all out war with Kodak for market share in the movie market. that is quite likely why one can buy "Vision 3" as both vision and vision 2 were outflanked by various Eterna products.

The two major things Fuji seemed to do better than Kodak was more emphatic 'chilly' tones in some film stocks & some films (Eterna Vivid) that could get in the same tonal area as the cine packaging of Velvia - which was very popular for certain purposes - but neg film offered much easier workflow paths & intercutting.

Agfa mostly marketed in Europe, and they probably were even more dependent of Print film sales. I do remember the Agfa Movie negative and much prefered the more laid back colour balance.

From what I've seen of it, I wouldn't disagree - sadly subtlety of colour is not much valued!

 

[138S]

It would help if you looked more carefully at it. The '0' represents the midtone - 3 stops up from the start of the straight line and 4 stops below the shoulder. 7 stops of recorded range on the straight line. As it should be.

Lachlan, please see the basics about what 50D is, and later we may advance in the debate. 10 Stops linear, 9.5 at least:



______________


Regarding lp/mm, the same than portra, V3 graph is better done, not surpassing 100%:

 

[Lachlan Young]

Lachlan, please see the basics about what 50D is, and later we may advance in the debate. 10 Stops linear, 9.5 at least:

Your reading of sensitometric curves is rather creative here. 5201 delivers about 6 stops of straight line, 5203 can deliver maybe as much as 8 on to 2383 - thus the 'improved highlights' claim. Your toe and shoulder points are way too deep in the toe and too high on the shoulder.

When you set the exposure for MP negative, you key to the highlights - because you cannot easily burn them in at the printing stage (unlike still negatives). Thus highlight latitude/ retention is important especially for outdoor usage where not having to fill shadows heavily (always fun on a slow stock) is a definite advantage - and 8 stops often seems closer to an average outdoor scene than 7 - and given the conditions you may end up using a 50 speed stock under, a stop or so of fill is a heck of a lot less than 3-4!

Regarding lp/mm, the same than portra, V3 graph is better done, not surpassing 100%:

Those data sheets are correct & overwhelmingly match what I have seen from using both films. For various reasons well covered in the academic literature of film system engineering, colour neg still film is engineered to give significant edge effects on big objects, which is the cause of the over 100% response on the MTF curve - which is a measure of sharpness, not high/ low contrast resolution - for that matter because MP film is principally used in formats smaller than 645 in area, contrast matters more than edge effects in terms of perceived sharpness - indeed it may be the case that if the format is not large enough to exploit the edge effects on large area objects, the enhanced sharpness may merely raise the noise level and reduce the film information capacity. What the MTF curves show are some of the differences between a film engineered for S8-65mm versus one engineered for 135-8x10 or bigger. This difference becomes quite obvious in 120 format where key changeovers between edge effects and contrast seem to happen.

 

[138S]

Your reading of sensitometric curves is rather creative here.

Lachlan, perhaps if I show you the 50D datasheet graph at full size then you would be able to count linear stops: 10, not 7. Its about counting, 1, 2, 3...

https://www.super8.nl/file/7203.pdf

It is quite evident you have shot little 50D (or none), I you were experienced with that film you would have noticed how it works in the highlights, and you would realize how top notch cinematographers use it.

10 stops linear, 15 total, with standard processing. With some pulling you get 18 or 20 total, not a joke.

Buy some and expetiment a bit, $0.5 per feet.

 

[Lachlan Young]

Lachlan, perhaps if I show you the 50D datasheet graph at full size then you would be able to count linear stops: 10, not 7. Its about counting, 1, 2, 3...

https://www.super8.nl/file/7203.pdf


View attachment 238777


It is quite evident you have shot little 50D (or none), I you were experienced with that film you would have noticed how it works in the highlights, and you would realize how top notch cinematographers use it.

10 stops linear, 15 total, with standard processing. With some pulling you get 18 or 20 total, not a joke.

Buy some and expetiment a bit, $0.5 per feet.

Kodak actually claim 11 stops of absolute range for 5203 (and the rest of V3) - but only in terms of helping shadow detail & holding specular highlights outwith the core 5-stop range. '0' represents an 18% grey card. 3% black card places 2 2/3 below that & white card 2 1/3 above '0'. The latitude is sufficient to let you over or underexpose a stop without hitting the toe or shoulder - or letting you handle high contrast situations without needing to fill shadows hard to allow you to hang on to a detailed highlight that you don't want falling into the shoulder.

If you look at Portra 400 (for example), the straight line alone is 10 stops at least. That's a significant difference in what can be held with good separation as well as latitude - not merely as hints of highlight detail rising onto the shoulder or shadows disappearing into the toe. My own experience mirrors this: 50D doesn't have the same extreme exposure range as Portra - and it wasn't designed to either. Both are beautiful in their own ways, but it is an enormous waste of time and creative energy trying to force one to be the other. Accept them for what they are and learn to use them creatively rather than pretending you can make them into what they are not.

 

[Cholentpot]

My family tossed all the negatives. I'm still torn up about it.

As for Vision 3 being superior to Portra, I shoot it in 35mm and it's far grainier than consumer stuff.

 

[zischga]

I can confirm that the Vision 3 50D is damn sharp and has a lot of latitude, I did not have a single under/over exposed frame yet, I usually have it pushed to ISO 100 by my lab.
In my opinion the colors are very accurate and neutral, I recently shot portraits with my Leica R6 + Summicron 50 and scanned it with a Coolscan V,
the grain is very fine, I think people wouldn't expect this to be shot on film, even in full resolution.

My conclusion: I don't know exactly how it compares to Portra, but it is a damn nice film and with short ends I can shoot Vision3 for about 2€ - 2,5€ a roll, which is a steal compared to Portra.

Here is an example I shot with a Summicron 50 + scanned with the Coolscan:
https://free-picload.com/images/2020/01/24/67c0be0669d6729232ec0f43026ed97c.jpg

@Cholentpot do you develop 50D in C41?
My local lab technican told me that developing cine film in C41 leads to stronger grain and shifted colors.
When developed properly in ECN-2 chemistry it is maybe not superior to Portra, but equal.
Definitely better than the consumer stuff.

 

[138S]

If you look at Portra 400 (for example), the straight line alone is 10 stops at least.

No... Portra 400 linearity is two or three stops less than VISION3 (with standard developments), Portra 400 is under 8 stops in linearity, VISION3 is 10. This is well known.

My family tossed all the negatives. I'm still torn up about it.

As for Vision 3 being superior to Portra, I shoot it in 35mm and it's far grainier than consumer stuff.

Yes, more grainy for the same ISO, but 50 is fine grained, a particularity of VISION3 is that blue channel is more grainy than red and green, compared to pictorial film, a bit optimized for skin tones, I guess.

But when you shot VISION3 50D in MF (Cinestill) it blows away Portra in several fields, in particular when subject is bathed with strong light.

The VISION3 capability to depict face/body volumes from textures in the glares is simply amazing. An exemplary usage of highlight depiction from VISION3 can be found in Django Unchained (2012, 500T/200T) by Tarantino, work made Robert Richardson (cinematographer) is simply extraordinay, an extraordinary master class on photography, and also a master class about how to exploit film medium strengths. This 2020 that kind of work cannot be done by any kind of digital capture, it is beyond what sensors/softwares are able by a LARGE margin.

https://en.wikipedia.org/wiki/Robert_Richardson_(cinematographer)