Topic: How to calculate film speed from the characteristic curve

[Francois]

This is a question we all get at one point or another: how to calculate a film's speed when the only information available is its characteristic curve?
Well, this had me dig through some old books for the answer. It came from Hollis N. Todd and Richard D Zakia's classic volume: Photographic Sensitometry - The study of tone reproduction. A book that is great for both the amount of information and its capacity to generate splitting headaches!
While the information in it is quite something, I can't vouch for its clarity so I decided to take the better part of the day to try and clarify things up and do a step by step description of the process used.

Now, over the years, the method used for calculating sensitivity has changed considerably. The original ASA rating method began being used in 1943 and was subsequently changed in 1960 in order to better match the European methods of measurement. Even though we now use an ISO rating, the measuring method is quite similar to that of the modern ASA rating.

Here are some slides describing the process. The curve used is the one from Kodak Vision Color Intermediate film (5242, 7242, 2242, 3242). For the calculation you will need a full page enlargement of the film's characteristic curve, colored pens and a ruler. You can also use a drawing program like I did. It will simplify the addition of fairly precise scale marks.

Now in the technique presented, it is asked that we calculate the density range, something that once the steps completed I discovered is not really necessary. The speed is set for proper exposure of the shadows. This method is only viable when it comes to pictorial applications as the requirements change depending on what you plan to do with the images.

Edit:
I've now produced a simplified technique that is based on Kodak documentation and appears to be much more simple, precise and useful.
I've now eliminated the part where the density variation is calculated to verify is the development is adequate since we can assume that the companies producing such data take every precaution in making sure that their documentation is correct.

Thank you.

[mcduff]

Thanks a lot Francois! I really hope I did not keep you up too late thinking about this and give you too many headaches!!

So I have a couple of questions:

The first:  I am not sure why you add the "0.2" in image six (the "A+ 0.2" part). In the awesome PDF that PhotoUtopia pointed out -- http://motion.kodak.com/motion/uploadedFiles/US_plugins_acrobat_en_motion_education_sensitometry_workbook.pdf -- there is no reference to adding this. Any comments?

The second: Are you saying that the whole issue of having a denstity difference of 0.80 (±0.05) is just more of a goal than a requirement, and if it does not happen to just not worry about it? In your example the film was higher contrast than that giving a delta of 1.18 and in my example it was less than desired of .65 or something.

[Francois]

I think you need to add 0.2 to make sure you're off the toe and able to record some detail in the shadow area. But the book didn't explain didley on that.
I too wonder about that... but it seems it was part of the norm.

As for the 0.80 density difference, it's more of a goal than anything else. It happens that 0.80 is the density difference required for enlarging with condenser enlargers. For diffusion type, it's 1.05 (from memory).

Reading the material, I also noticed that the goals for color negatives are slightly different than for B&W.

I'll have to read the Kodak publication to compare. Maybe they're better writers...

[Terry]

Francois, thanks for doing this and making the "walk-through" of the process.  I feel somewhat enlightened, but had to go to the Freestyle site and order a box of 5x7 Arista to make the headache go away.

[Bryan]

Francois, thanks for doing this and making the "walk-through" of the process.  I feel somewhat enlightened, but had to go to the Freestyle site and order a box of 5x7 Arista to make the headache go away.

Take two 5x7's and call Dr. Francois in the morning.

[Photo_Utopia]

Hi Francois
Thanks for taking the time to do this. One thing is confusing (to me at least) you seem to be averaging out the emergence point (Jones point) and dividing that by 10 to get the ASA (ISO) which I'm not sure is right.
There was a method common before 1963 where the exposure Jones point was divided by a constant to give a speed, I'm not sure they still use this method.
I think from the diagrams you've posted you might have an iSO of about 6 ballpark just by eye.

To verify your method try to find the curve for a known ASA film like Portra 400 and use your method to arrive at that speed-I'm not sure it will work.
I'm away from all my literature at the moment so can't find the modern formula but from memory it was the density at point (P) divided by the exposure to give 0.2 above fog.

In B&W if a film can't get 0.8D at 1.3 above base fog then the film needs more development to obtain that speed or more exposure both of which will change the pictorial representation of tone.

Which brings us to the question why bother? What does this mean to the average Joe/Jane?
Well 1.5 D log E on a curve is the equivalent to a brightness range in the negative of 32:1 (1.3 units is 20:1) so if it meets that standard a user can expect to capture a scene with that brightness range with that film.
This explains why if we expose for a wide range scene we need to lower the ISO (on the meter) and give less development, if we have a lower SBR than 20:1 (low contrast) we need to rise the ISO and give more development.
You're changing the contrast of the film, or expanding or contacting the range of tones it can record.

Lastly I noticed a comment (can't remember who) that the Red channel on a colour curve is much lower (separated from)  the green and blue. This is because of the colour masking (orange) in negative films, slide films RGB plots join up where negative ones don't but should remain parallel.

[Francois]

The division by 10 is actually 10^(average log exposure)
For me, going through this trouble is only good for those films with a mystery sensitivity like cine copy films.

Like I said, I'll have to do a few comparisons to make sure I understood the less than clear text...
But it should be OK... I hope since I don't want my head to hurt like it did yesterday!

[Francois]

OK, I did a verification against some Kodak Portra 400 and the method I described is flawed!
Well.... I'm sorry for all those who tried to understand it. My comparison tells me that Portra 400 is really a 176 ISO sensitivity. Definitely wrong.

Back to the drawing board...

I just checked the Kodak info book and it seems it works better. It tells me Portra 400 is actually a 338 ISO film.

I think I know what went wrong... the book I relied on was written in 1969. According to Wikipedia, the measurement methods were revised in 1979 and again in 1986...

I'm starting to get a feeling of why most films were of such low sensitivity back in the old days!

[Francois]

OK, now the document is corrected and should more adequately reflect reality.

[mcduff]

A quick response from me is that this seems to be the way I have been reading how it is done. I am saying this fully acknowledging that I am just learning this from the stuff PhotoUtopia and Francois have been teaching me! But I think we are getting there. The turbo-nerd part of me is very appreciative of this, BTW.

[Francois]

I wonder if the document that accompanies the workbook is available somewhere?

[mcduff]

I wonder if the document that accompanies the workbook is available somewhere?

Sorry, Francois, which document? Do you mean that pdf you posted in the other thread? http://motion.kodak.com/motion/uploadedFiles/US_plugins_acrobat_en_motion_newsletters_filmEss_06_Characteristics_of_Film.pdf - it is a good one too! I hope it will not be contrary to the kodak workbook, I think I am finally starting to figure stuff out from it, haha. I have printed off that workbook and am doing it as we speak! I know I should go out and take pics, I will do that later (its cold here!)

[Francois]

If it's cold at your place, it means it's on its way here...
It's dull and gray... I can't even remember what sun is like...

I checked and the workbook is all there is and contains all the information needed in it...

But the characteristics of film document does explain a few things like the arbitrary 0.1 over base+fog...