Choosing a film stock, considering the contrast curves and color sensitivity

[Sam David Zhang]

Hey everyone,

I am currently doing a research paper for my school, and I am tasked with answering the question: “Reading the film stock curve and selecting a stock based on their color & contrast representation.”

I’ve done some preliminary research into the subject, reading Kodak’s Sensitometry Workbook along with a handful of film manuals at the library, and have interpreted the essay prompt as “How can the characteristic curve and spectral-sensitivity chart inform a decision on how a film stock will perform?”

I decently understand the concepts of contrast/gamma, how to feel out latitude of a stock from a quick glance at its D logE cuve, and the differences in sensitometry between reversal and negative stocks.

However, I wanted to reach out to the community and ask any cinematographer who has worked a lot with film – or is currently working with film on a show – how reading the spec sheet has informed your decision on a film stock. What did you do with the information? How has it led to your testing phase? Is it even important in the end versus real life testing?

In addition, I had one technical question that deals with the RGB curves in the D logE chart when it comes to normal process vs push/pull process.

As an example, let’s take a look at Kodak 200T listed at this pdf link below:

https://www.kodak.com/content/products-brochures/Film/VISION-200T-Sellsheet_US_4PG-EN.pdf

In the contrast curve (named “sensitometric curve”), it appears that Blue and Green curves are slightly steeper when compared to the Red curve. I want to confirm that this means the gamma of the individual channels for blue and green are higher, meaning there is “higher contrast in those colors.” Practically, this means that I can expect more Blues and Greens as exposure go up into the highlights, versus the shadows? (is this right?)

Furthermore, say I push this stock by 1-stop. If I charted the resulting sensitometric curve of the developed negative, the toe would remain the same… but the shoulder would be lifted by that 1-stop compensation. Does this 1-stop push get evenly spread across the RGB channels? If so, does that means the steeper gammas of Blue/Green compared with the Red channel becomes even more exaggerated? In theory, I would expect some more blue/green in the highlights?

Please let me know if I have this concept wrong. Also, would love to hear any of your personal experiences or thoughts on this topic as a whole.

 

Thank you

Sam

[Tyler Purcell]

I don't think it really matters anymore, with the DI process you can correct the stock to fit any color science you wish really. 

Back in the photochemical finish days, these characteristics really meant something. Today it's really about grain structure and sensitivity as the Vision 3 stocks are all so similar. Since Vision 3 ushered in the days of DI, it's really the first stock designed from the ground up to work with film scanners and a digital finish. It doesn't have the same feel as the older EXR or even Vision 1 & 2 stocks, where were mostly made for photochemical finishes. 

Pushing one stop, really doesn't do much. You can recover 1 stop easily in the scanner. Two stops, different story, you can actually effect the sensitivity quite a bit with a two stop push, all be it with a lot more money for the processing. It will actually lift the blacks up a tiny bit and is a trick that's been used on several major films to make a more unique look to the finished product. Pushing does also add some contrast, where pulling generally reduces contrast. That added contrast is a hindrance for photochemical finishes, but it's fine for digital finishes. Yea, the channels generally increase all together. 

David and a few other guys know a lot about the photochemical process and why the different stocks have significance. 

 

[Robert Hart]

Slightly offtopic, the proposed new ORWO colour stock could be interesting. It is apparently based on Agfa Gevaert XT320 formulation but seems to have a different look. It's uniqueness may appeal to creative folk. 
 

 

[Tyler Purcell]

35 minutes ago, Robert Hart said:

Slightly offtopic, the proposed new ORWO colour stock could be interesting. It is apparently based on Agfa Gevaert XT320 formulation but seems to have a different look. It's uniqueness may appeal to creative folk. 

I have a feeling with the turmoil in Europe right now, we won't see a finished product for a while. All of the test film was processed and scanned by them. I think they made a single sheet just to get feedback. It does look good, so at least that's good. 

[Mark Dunn]

1 hour ago, Tyler Purcell said:

the turmoil in Europe

Eh? There's a war in Ukraine and an energy crisis, but apart from increasing its costs I don't see why that should adversely affect business in Germany.  It's 1500km from Wolfen to Kyiv.

[Tyler Purcell]

8 hours ago, Mark Dunn said:

Eh? There's a war in Ukraine and an energy crisis, but apart from increasing its costs I don't see why that should adversely affect business in Germany.  It's 1500km from Wolfen to Kyiv.

Where do you think the low cost raw materials are coming from? Not EU! 

[Robert Houllahan]

On 10/13/2022 at 12:57 AM, Sam David Zhang said:

However, I wanted to reach out to the community and ask any cinematographer who has worked a lot with film – or is currently working with film on a show – how reading the spec sheet has informed your decision on a film stock. What did you do with the information? How has it led to your testing phase? Is it even important in the end versus real life testing?

I find that most DP's will look briefly at the Kodak spec sheet but then do allot of tests with the stock and camera lens they want to work with. Also if they plan to do any push or pull processing etc. or other look creation work they test test test.

[Daniel D. Teoli Jr.]

On 10/14/2022 at 3:26 PM, Robert Houllahan said:

I find that most DP's will look briefly at the Kodak spec sheet but then do allot of tests with the stock and camera lens they want to work with. Also if they plan to do any push or pull processing etc. or other look creation work they test test test.

I'm glad someone said it...TEST!

I was going to answer this thread telling them the same thing when I first read it...but I figured they didn't want to hear it. 

Some people are crazy for this stuff...they are usually the anal people. (As in analytics.)

 

 

As I've told you all before, I like seeing the practical application and not charts and graphs so much.

Here, look at this for an example of practical application...

 

Left: Printed with Epson R2000 on matte paper with matte black ink.

Middle: Printed with Epson R2000 on semi-gloss paper with gloss black ink and gloss optimizer.

Right: Printed with Epson 3880 with the same semi-gloss paper as the middle sample using gloss black ink. 

Photograph above: The Lost Princess 2014 (Candid) by D.D.Teoli Jr.

Being this is a visual medium; we need to see the end product and not a bunch of graphs and charts. You can have the prettiest chart in the world and the end result still may look like shit! That's the bottom line.

Note: If you have trouble seeing a big difference in each print, view them on a color balanced monitor.

 

Edited October 21, 2022 by Daniel D. Teoli Jr.