Understanding EI???

[Stephen Benskin]

Steve

All the theory aside, what PE explains here matches very well with practical B&W photographic experience, and years of film and paper testing.

As you well know, B&W negative film, exposed at ISO speeds, is not very forgiving to underexposure, but it needs excessive overexposure before highlight compression becomes a real problem.

Yes it does and I agree and so does the theory I'm talking about because most of it is based on psychophysical testing. One thing we should point out is in all psychophysical testing such as the first excellent print tests or any type of evaluation of quality for that matter, there will be a range of opinions as to the degree of quality. I suggest we should keep that in mind when suggesting any preferred approach. As you might have noticed, I don't advocate any one approach. When I talk about the ZS testing method, I deal with the correctness of the speed values and not whether the resulting exposure works (it does BTW). For some it might be nit-picky, but to me the knowledge is freeing.

I also agree there is a great deal of overexposure latitude. Anyone who is familiar with the results of the first excellent print test can attest to that. In fact, I believe you have a graph of those results in your book on page 141. While my metering technique is rather complex and is contingent on the subject and the desired results, I too generally place values slightly higher on the curve for better shadow separation and so that in printing I have a greater range of choices on how to print them. Still, there is a one stop underexposure latitude just in case. While many discerning observers might not appreciate the diminished shadow separation, the safety factor is there.

We are in violent agreement on the desired results and application. We just have disagreements on the theory involved, and for me, theory is fun to talk about.

 

[Photo Engineer]

One, it was Stimson who wrote it. Two, I'm showing it because it represents the standard model for which film speed and light meter calibrations are based upon. I'm not advocating anything. BTW, point 4 is the 0.10 fixed density speed point. You should re-read my posts about how exposure actually falls at least 1/3 stop over speed point according to exposure theory.

I am referring to your post #70 here: (there was a url link here which no longer exists) and in particular to the graph and point (4) on it which is on the toe.

Ralph and I both commented on it in subsequent posts explaining as I have said, that to correct for this "error" one would have to add about 1/3 stop exposure to fix any shadow detail problems in an average scene.

My advice is practical, not theory, derived from literally thousands of tests. Now, overdevelopment, and subsequent higher contrast changes the entire playing field. Underdevelopment, ditto!

PE

 

[RalphLambrecht]

A big difference to me is when people think they are getting one thing when they are not. I like to understand what is really going on so I can make the decisions myself and not just follow what I've been told to do. That makes a big difference to me.

We agree on the part about the potential benefits of additional exposure. For some reason, people tend to assume that just because I am explaining how ISO and exposure is designed to work I am a stickler for the principles in practice. Just the opposite, I believe a thorough understanding of the concepts involved frees you up.

I understand how you feel.
To me, there are four levels of knowledge:

1. unconscious incompetence
(people who don't even know that they don't know)
2. conscious incompetence
(people who know that they don't fully understand)
3. unconscious competence
(people who know it works as long as they follow a given recipe)
4. conscious competence
(people who know how to make it work or fail)

I'm usually somewhere between level 2 and 3.

 

[Chuck_P]

Please, take a look at the google link to 'The Manual of Photography' in post #134. It's all in there, including an explanation the the five different methods of measuring film speed. It's a nice overview for all who want to follow or participate to this thread.

Yes, I found that article very enlightening regarding the various methods of measuring film speed. Thanks Ralph.

 

[Stephen Benskin]

I am referring to your post #70 here: (there was a url link here which no longer exists) and in particular to the graph and point (4) on it which is on the toe.

Ralph and I both commented on it in subsequent posts explaining as I have said, that to correct for this "error" one would have to add about 1/3 stop exposure to fix any shadow detail problems in an average scene.

My advice is practical, not theory, derived from literally thousands of tests. Now, overdevelopment, and subsequent higher contrast changes the entire playing field. Underdevelopment, ditto!

PE

Again, I was not advocating anything. I was presenting how film speed, exposure, and the light meter work together that was derived from literally thousands of tests. That's it. No recommendations on how to expose from me on that. Then I gave the reference for the source so people who are interested can read more about it and not just take my word for it.

Okay, I'm going to play with you a little here. When you suggest that by add 1/3 stop to fix shadow detail problems, that must be for a reason. You must have a "theory" as to why it works better. There is always theory behind any conclusion. Isn't the use of the word "error" implying some conclusion or "theory" resulting from testing a hypothesis? We shouldn't be making such a distinction between theory and practice. Surely, they work together. I don't have to tell a scientist that.

 

[Ray Rogers]

AFAIK, Kodak never sold any film with a "corrected" ISO value penciled in on the box.
To us, such film was scrap! It was not sold. PE

I will attempt something I have never done... sending an attachment along with this post.

PE,
Here are a couple of examples of how Kodak advised users of off target speeds... Usually the speeds were very close to BOX speed but the 2nd image here shows a one stop difference. (EI 20, 25 or 32 depending upon exposure time) I had remembered that the numbers were in red when they were not on target, but the examples I found doesn't really support that memory as the color varries and values are given even when they were on target.

Also, the "Effective Speed" is given as an ISO value and not as an "Exposure Index".

Ian,
All the T-max data I looked at (ca 1986) for various products used EI.

 

[Chuck_P]

Well, this thread has certainly been a whirlwind of thought, very interesting. IMO, theory, at some point, gives way to practical application of it, but this does not dismiss the importance of understanding the theory itself.

Chuck

 

[2F/2F]

So, what is EI anyhow? Is that that number that I set my light meter at that affects the recommended exposure? It is? Cool. Thanks.

 

[Ray Rogers]

The use of T-grain emulsions simply highlighted a potential problem with the standards that required an updating.

Specifically, what was that potential problem?
Can you summarize it succinctly?

Thanks,

Ray

 

[Photo Engineer]

Okay, I'm going to play with you a little here. When you suggest that by add 1/3 stop to fix shadow detail problems, that must be for a reason. You must have a "theory" as to why it works better. There is always theory behind any conclusion. Isn't the use of the word "error" implying some conclusion or "theory" resulting from testing a hypothesis? We shouldn't be making such a distinction between theory and practice. Surely, they work together. I don't have to tell a scientist that.

I've already answered that question before. Remember that the gamma of any point on a print = gamma film x negative paper gamma. This is taken point by point and the slope here is the first derivative of course. So enters calculus.

In this case, the film slope is less than the 0.63 optimum that was designed into it, and it is being printed on the shoulder of the paper which has lower gamma. If you move the exposure up by 1/3 stop, the slope or delta density / log E (dD/dLogE) becomes greater by the difference in the slope of the toe vs the slope in the mid scale. Therefore the perceived dD/dLogE that we see in shadows is greater and there can be more of them thus giving higher detail in the shadows.

PE

 

[Photo Engineer]

I will attempt something I have never done... sending an attachment along with this post.

PE,
Here are a couple of examples of how Kodak advised users of off target speeds... Usually the speeds were very close to BOX speed but the 2nd image here shows a one stop difference. (EI 20, 25 or 32 depending upon exposure time) I had remembered that the numbers were in red when they were not on target, but the examples I found doesn't really support that memory as the color varries and values are given even when they were on target.

Also, the "Effective Speed" is given as an ISO value and not as an "Exposure Index".

Ian,
All the T-max data I looked at (ca 1986) for various products used EI.

Ray;

Thanks. I had never seen nor heard of that before. Of course, I never worked on reversal film, so I guess there were some problems.

OTOH, the one on the right seems to indicate that this "correction" is for reciprocity failure as it gives the data and says it applies to 3" exposures. (or longer?)

In any event, the one on the right is common if it refers to reciprocity corrections.

Thanks.

PE

 

[Stephen Benskin]

I've already answered that question before. Remember that the gamma of any point on a print = gamma film x negative paper gamma. This is taken point by point and the slope here is the first derivative of course. So enters calculus.

PE

Sorry, I was joking with you about a concept. It won't happen again.

 

[Stephen Benskin]

Specifically, what was that potential problem?
Can you summarize it succinctly?
Ray

I had heard that there was talk before the t-grain incident questioning the relevance of the standard's developer because it wasn't one used in the real world and may not reflect real working conditions for certain films. The T-Max films emphasized this concern.

What year are the boxes from?

Steve

 

[Ray Rogers]

What year are the boxes from?
Steve

T-Max?

The oldest I saw looked at were from 10/86.

 

[Ray Rogers]

Ray;

Thanks. I had never seen nor heard of that before. Of course, I never worked on reversal film, so I guess there were some problems.

OTOH, the one on the right seems to indicate that this "correction" is for reciprocity failure as it gives the data and says it applies to 3" exposures. (or longer?)

In any event, the one on the right is common if it refers to reciprocity corrections.

Thanks.

PE

Ron,
You're welcome!

The one on the right was, I think, designed for exposure times of 1/10 to 100 seconds;
other examples I have for the same film do show much less LIRF. (but beyond reciprocity proplems, this film was out of spec as no exposure times were near box speed.)

Actually, the reporting of such supplemental information was a sign of Kodak's high standards and professional integrity.

Ray

 

[Photo Engineer]

Ray;

At that time, in the mid 80s, we formed the Committee for Emulsion Lab Standardizations, (CELS) due to problems encountered in replicating a product all over the world. AFAIK, this may have been an example of a problem making E6 films in Europe for example vs the US. IDK, I do know that there were some problems, but not if this is related to it. It may have just related to the raw emulsions which could have been tweaked into final compliance with an aim.

In any event, regarding t-grains, I know of no instance in our development projects where t-grains performed differently in "standard" developers. I do know that the products had to have tailored t-grains to meet specs for speed, grain and sharpness. This is always the case.

Steve, if you would be more specific, I would love to play your "game". But from re-reading the post, it really doesn't sound much like fun. Sorry. I worked with several people who had input to the ANSI standards and at least one member of the Committee. I have read their internal reports and designed products to meet their criteria. So, I guess I did at least answer you in a practical fashion.

PE

 

[Ian Grant]

Thanks Ray, I was sure the early Tmax only had an EI speed not an ISO, and like you I remember using Ektachrome with the corrections marled as ISO, the first time I saw EI used on a film bow was the when Tmax was released in the UK.

In the UK there was some surprise at the time that Tmax had no ISO speed but then John Sexton's tests & articles and other independent writers had already indicated that for most uses it was better to use a much lower EI, and in fact even Kodak recommended 50 EI in teir datasheet for fuller tonality. Because of disquiet Kodak then hastily began marketing Tmax developer which allowed better tonality at box speed.

Many didn't and still don't like Tmax films, I tried TMX & TMY and have been using them ever since, for personal and commercial work although alongside initially APX100 &25, and more recently Ilford Delta 100 & 400 because of poor local availability.

The issue of the changing ISO standard methods just seems to cloud the issue but it highlights the necessity to test films for yourself and fine tune your personal EI's for a particular emulsion.

Ian

 

[Ian Grant]

Ray;

At that time, in the mid 80s, we formed the Committee for Emulsion Lab Standardizations, (CELS) due to problems encountered in replicating a product all over the world. AFAIK, this may have been an example of a problem making E6 films in Europe for example vs the US. IDK, I do know that there were some problems, but not if this is related to it. It may have just related to the raw emulsions which could have been tweaked into final compliance with an aim.

PE

Ron, I guess this relates to cleaning up previous film anomolies.

When I first used Tri-x and a very rare ocassion when I bought a Kodak developer the Datasheet had recommended Film speeds/Development times for Tri-x from 3 different locations (& presumably coating lines).

Kodak USA
Kodak, Canada
Kodak, UK

It's quite probably I still have a datasheet as I have 2 packets of Microdol and 1 equally vintage D76

Ian

 

[Chuck_P]

The issue of the changing ISO standard methods just seems to cloud the issue but it highlights the necessity to test films for yourself and fine tune your personal EI's for a particular emulsion.

Nicely said.

 

[Ray Rogers]

In any event, regarding t-grains, I know of no instance in our development projects where t-grains performed differently in "standard" developers. I do know that the products had to have tailored t-grains to meet specs for speed, grain and sharpness. This is always the case.
PE

Yes, thats the price of admission, so to speak.

But I think you said it best earlier when you wrote:
"As for the developer, T-max performs "better" in a developer tailored especially for T-grain films. IDK what that means, as IDK the exact formula for that developer."

Exactly what this "better" means is what I would like to figure out.

Ian has commented on this while I was composing this post, (I peaked!) and what he wrote, combined with what I extracted from a glance at one of the two data sheet links you posted, along with your own statements about T-max performing "better" in a developer tailored especially for T-grain films, seems to be in harmony in pointing to a slower ISO speed for T-max films in D-76 than in T-max developer. (I still have not checked what formula was the used as the ASA or ISO standard film developer formulas but I suspect it was something close to D-76) Now, Was that speed the same as Tri-X in D-76? Perhaps, I don't know. PE seems to be saying it was. If true, what does all this indicate?

Ray

 

[Ray Rogers]

Thanks Ray, I was sure the early Tmax only had an EI speed not an ISO...

The issue of the changing ISO standard methods just seems to cloud the issue....
Ian

You're welcome, Ian!

Clouds, yes - although I do think I can see two sides of the coin; on one side, if you force all emulsions to be compared in a standard developer, the user knows how they compare in the test developer, which may not be optimal.

OTOH, by allowing another, more optimal developer to be used, the manufacturer can publish a more attractive (higher) ISO rating for the film.

Essentially, the standard seems to have been changed from one that compared emulsions to one that compares processes.

Ray

 

[Ray Rogers]

The issue of the changing ISO standard methods just seems to cloud the issue but it highlights the necessity to test films for yourself and fine tune your personal EI's for a particular emulsion. Ian

Nicely said.

Yes!

But let's not forget that that is exactly what Kodak has been telling us publication after publication after publication, and is printed on every Kodak data sheet I can remember reading!

 

[Chuck_P]

Yes!

But let's not forget that that is exactly what Kodak has been telling us publication after publication after publication, and is printed on every Kodak data sheet I can remember reading!

Yes, let's not----, back at ya

 

[Ian Grant]

Ray, of course with only one US manufacturer of B&W films by 1993 it's easier for EK to get the ASA part of the ISO standard changed to help their own goals, that list Stephen gave has no rival manufacturers at all..

The ASA standard test developer was close to Agfa Ansco 17 [Agfa (Germany) 44] than D76 and almost identical to Adox Borax MQ, and both these developers give approx a 1/3 of a stop more film speed than D76 with most films.

Ian

 

[RalphLambrecht]

...The issue of the changing ISO standard methods just seems to cloud the issue but it highlights the necessity to test films for yourself and fine tune your personal EI's for a particular emulsion...

Sums it up nicely.