speed point, 0.1 over base plus fog or 1/3 of gamma?

[alanrockwood]

I am not sure if this post goes best in the exposure section of the darkroom section.

I understand that early-on Kodak folks considered the best choice for speed point would be where the slope of the characteristic curve is 1/3 of gamma. Later they joined the rest of the world and went with a density of 0.1 over base plus fog as choice of speed point.

I suppose for a short toe film there's probably not much difference, but for a long toe film it might make a difference. If so, and if one could do accurate measurements of the slope of the characteristic curve (admittedly a challenge), for determining one's personal film speed (EI), would 1/3 gamma be a better choice than 0.1 over base plus fog? Intuitively it seems to me that it would be a better choice.

When I mentioned a short toe film it is based on the idea that if the curve was a straight line with a perfect break to zero slope at the toe (i.e. no toe) there would be no difference between the two methods. That's why I speculate that for a short toe film the two choices are probably not much different.

Opinions?

 

[ic-racer]

0.1 over is an approximation of 1/3Gbar. Do you have a computer program that can recover 1/3Gbar. If so that is great.
Delta-X and W-speed are two more approximations that have surfaced in the literature. My spreadsheet uses W-speed, because the functions in my spreadsheet software can solve the equations.

 

[Alan9940]

I've used a densitometer for many years when testing films and I've always shot for about 0.10 over fb+f; though I'm not obsessive about it. If I'm going to err, so to speak, I'll err on the side of more exposure. For example, if I get 0.12 over fb+f I'm good with that; 0.08, for example, would not be acceptable. Whether that speed point is 1/3 gamma I have no idea, but I've always been satisfied with my exposures after testing. I've treated both long- and short-toe films this same way.

 

[Stephen Benskin]

I am not sure if this post goes best in the exposure section of the darkroom section.

I understand that early-on Kodak folks considered the best choice for speed point would be where the slope of the characteristic curve is 1/3 of gamma. Later they joined the rest of the world and went with a density of 0.1 over base plus fog as choice of speed point.

I suppose for a short toe film there's probably not much difference, but for a long toe film it might make a difference. If so, and if one could do accurate measurements of the slope of the characteristic curve (admittedly a challenge), for determining one's personal film speed (EI), would 1/3 gamma be a better choice than 0.1 over base plus fog? Intuitively it seems to me that it would be a better choice.

When I mentioned a short toe film it is based on the idea that if the curve was a straight line with a perfect break to zero slope at the toe (i.e. no toe) there would be no difference between the two methods. That's why I speculate that for a short toe film the two choices are probably not much different.

Opinions?

Don't forget the other part of the method. 0.3x of the average gradient. The purpose is to have a methodology that produces the same quality results across as many film types as possible. The ISD B&W film speed standard uses Delta-X Criterion to calculate film speed. I'll upload the equations and example later. Simply put, if Delta D is the same, then Delta X will always fall the same distance from 0.10 no matter the curve shape.

With all due respect to IC-Racer, I have to disagree concerning 0.10. It's really more of an easily found measurement point.

 

[Stephen Benskin]

Stephen, I was going to type a response to Alan but thinking about it, would you mind posting your Delta-X/hiding in plain sight paper? I always thought that was excellent for people with questions just like Alan's since it is a good primer on the evolution and relationship of:

Print judgement->fractional gradient->"simple methods" (Delta-X)->ISO criteria (0.1 density w/contrast parameters)

Michael,

I'll do it when I get home from work, but I don't think it can beat your concise description.

Stephen

 

[ic-racer]

With all due respect to IC-Racer, I have to disagree concerning 0.10. It's really more of an easily found measurement point.

Yes, that is what I was trying to convey. No computer or complex methodology needed for that, but of course fractional gradient is related to fixed gradient under 'certain parameters.'

As it turns out, the fixed density method has a direct relationship to the fractional gradient method when certain parameters are maintained. From a paper that should be required reading "Delta-X Criterion"

 

[Stephen Benskin]

Yes, that is what I was trying to convey. No computer or complex methodology needed for that, but of course fractional gradient is related to fixed gradient under 'certain parameters.'



View attachment 271124

That is my table. I put together. I was planning on uploading it as it can be used instead of the equation.

 

[ic-racer]

That is my table. I put together. I was planning on uploading it as it can be used instead of the equation.

Yes, of course! The whole text should be a sticky too.

 

[ic-racer]

My spreadsheet broke a little when I changed software, but in its last version, after calculating W-speed, it displayed an ASA triangle. But required user to OK the curve fit inside.
The next version was to use that table (or similar) to iterate the numbers and by the spreadsheet's trial-and-error logic operator, find the Delta-X criterion that fit the film curve, thus the ability to automatically get the Delta-X speed from any curve, the same way it obtains W-speed from any curve.

 

[Bill Burk]

A thought Stephen often mentions is that the 0.3x average gradient is the actual ASA speed reference agreeing with 0.1 speed point, when the contrast parameters are met.

What often differs in practice, people feel a need to follow that 0.1 mark when they develop more or less to get different contrasts. They think the speed must have changed.

I like to find the speed at the ASA contrast and then, if I reduce contrast to make better pictures I ignore the fact the speed point moved. (Actually I use a Delta-X gauge that confirms the 0.3x average gradient hasn’t moved).

Pushing is a special case I don’t like to mess with because I like my shadows. But sure, if you develop longer you can pretend the film is faster if you want to

 

[Stephen Benskin]

Attached is my article on Delta-X. The scientific paper is Nelson, C.N. and Simonds, J.L, Simple Methods for Approximating the Fractional Gradient Speeds of Photographic Materials, JOSA, vol. 46, n. 5, May 1956.

One of the concept I find interesting about using a fractional gradient method is that there is an inverse relationship that exists between ΔD and ΔX. With higher than normal processing, the distance between the two points is reduced and the reverse is the case when the processing is lower than normal. The results is the film speed doesn't increase or decrease to the degree it would if using fixed density method.

Notice in the table above how a ΔD of 0.80 produces a ΔX of 0.296. This is the ISO standard. When the film is processed according to the parameters of the standard, ΔD = 0.80, the fractional gradient speed point will fall 0.296 log-H to the left of the 0.10 fixed density point.

With a ΔD of 0.70 resulting from reduced process has the fractional gradient speed point at 0.346 log-H from the fixed density point effectively producing a higher film speed than what would result from the fixed density method.

With a ΔD of 1.09 coming from increased processing, the ΔX is 0.178 log-H or about 2/3 of a stop shorter than with ΔD of 0.70 and about 1/3 stop less than with ΔD = 0.80. This will keep the speed from extending processing from raising as quickly as it would with the fixed density method.

The table below is an example of how speeds compare between the two methods. Please keep in mind that the film speed equation is different when calculating from the fixed density point and the fractional gradient point. So just because the fractional gradient point is almost a stop below the fixed density point for the ISO standard, doesn't mean the film speed is one stop faster.

 

[Stephen Benskin]

This was just recently posted in another thread. The Delta-X equation and the ISO parameters.

 

[ic-racer]

Here are some results of the 'film speed shootout' comparing the speed formulas to observer's choice of best print.

0.1 was the worst. In this chart, Delta-X and W-Speed were both actually BETTER than 0.3G.

 

[alanrockwood]

Here are some results of the 'film speed shootout' comparing the speed formulas to observer's choice of best print.

0.1 was the worst. In this chart, Delta-X and W-Speed were both actually BETTER than 0.3G.

View attachment 271136
View attachment 271137

Those charts look a lot like quality control charts, depending on what the x axis signifies? What is the x axis in these plots?

 

[alanrockwood]

Here are some results of the 'film speed shootout' comparing the speed formulas to observer's choice of best print.

0.1 was the worst. In this chart, Delta-X and W-Speed were both actually BETTER than 0.3G.

View attachment 271136
View attachment 271137

It looks like there is a high correlation between the charts in b, c, and d, but less so in a. Also, the 2x sigma of b, c, and d are quite close (perhaps the same within statistical error(?)), and a lot less than a.

 

[alanrockwood]

Here are some results of the 'film speed shootout' comparing the speed formulas to observer's choice of best print.

0.1 was the worst. In this chart, Delta-X and W-Speed were both actually BETTER than 0.3G.

View attachment 271136
View attachment 271137

Also, it looks like there is an offset (bias) in a, but not in b, c, or d. It looks like b, c, and d are probably just about equally good.

 

[Stephen Benskin]

Notice how each is a "vs Print Judgement Speed"? Print Judgement Speeds are the results of the psychophysical testing from the First Excellent Print test. This was an incredibly laborious effort so a sensitometric method needed to be determined which would most closely approximate the results from the psychophysical test. Different methods to determine film speed were applied to each of the negatives. The graphs represent how far each came from matching perfectly to the print judgement test, point 0. The tighter the spread, the more accurate the method over the various film types and curves.

 

[alanrockwood]

Notice how each is a "vs Print Judgement Speed"? Print Judgement Speeds are the results of the psychophysical testing from the First Excellent Print test. This was an incredibly laborious effort so a sensitometric method needed to be determined which would most closely approximate the results from the psychophysical test. Different methods to determine film speed were applied to each of the negatives. The graphs represent how far each came from matching perfectly to the print judgement test, point 0. The tighter the spread, the more accurate the method over the various film types and curves.

Does each point on the horizontal scale correspond to the evaluation of an individual print?

 

[Stephen Benskin]

This also comes from the paper Simple Methods for Approximating the Fractional Gradient Speeds of Photographic Materials. These examples shoe the inability of the 0.10 fixed speed criteria to predict the effective speed of film with different curve shapes. The difference between the 0.10 fixed speed point, S', and the print judgement speed, S, ( the fractional gradient speed point is indicated by the small arrows) varies depending on the curve shape. In these cases, toe characteristics, left 3 examples, as well as average gradient, right 3 examples.

 

[Stephen Benskin]

Does each point on the horizontal scale correspond to the evaluation of an individual print?

The individual negatives that produced the prints that were judged, yes.

The abstract and first two paragraphs from the introduction.

 

[alanrockwood]

Question: Is the delta x supposed to estimate the G_naught/3 exposure point relative to the 0.1 over base plus fog point?

(I think this post and your last two posts "crossed in the mail".)

 

[Stephen Benskin]

Question: Is the delta x supposed to estimate the G_naught/3 exposure point relative to the 0.1 over base plus fog point?

(I think this post and your last two posts "crossed in the mail".)

Yes. The Fractional Gradient method is the most accurate method, but obtaining it was difficult and error prone. Europe didn't want it as an international standard, so Kodak figured out how to obtain the same results using an easier technique.

In effect, most people have been misinterpreting the fix density part of the ISO standard.

 

[Stephen Benskin]

Does each point on the horizontal scale correspond to the evaluation of an individual print?

The vertical axis is log-E or today log-H which I have to believe is the difference in where the film speed fell from the aim of 0. 0.1 would be 1/3 stop. The plus and minus would be considered faster or slower speeds than the judged speed.

 

[alanrockwood]

Stephen, I found the paper you referenced (Simple Methods for Approximating the Fractional Gradient Speeds of Photographic Materials) and downloaded from my university library. It should make for interesting reading.

 

[Stephen Benskin]

Stephen, I found the paper you referenced (Simple Methods for Approximating the Fractional Gradient Speeds of Photographic Materials) and downloaded from my university library. It should make for interesting reading.

Enjoy.

The internet these days. You know how long it took me to track down all these papers? Searching through the stacks at USC and UCLA, and all those hours standing over a hot copier machine.