Stephen
I cannot entirely agree. I agree, the ISO speed is not arbitrary, it is not purposely overrated by manufacturers and it is not a guesstimate, but it has to be personalized, because the ISO speed is determined by methods and standards, which may differ from what the photographer wants to achieve.
This is the reason to conduct personal film testing and the need for EI vs ISO. And, since these personal test consistently return a speed that is lower than ISO, the ISO standard seems to differ from what photographers want to achieve, and therefore, it differs with customer 'reality', which is an unavoidable fact. How can ISO know what developer I'm using?
This is not a criticism towards the ISO speed, method or the people who put it together. The ISO speed is a valuable comparative measure, but it is not the all-encompassing answer of the actual film speed. For that, the photographer needs to do his or her own testing.
Ralph,
This might just be about semantics, but I think it's an important distinction in order to really grasp certain concepts. There is film speed and there is personal taste in exposure or personal experience with exposure (EI), but it shouldn't be considered film speed.
Again, I've stated why the personal testing isn't so much about gaining a personal EI that is really different from the ISO. If you really evaluate the structure behind the "personal film test", it really doesn't reflect anything. It introduces multiple experimental errors and there are many factors no body considers. It also isn't an apples to apples comparison with the ISO test, so it can't comment on anyway with the the ISO speed, which for me has to be considered the actual film speed. Anything and everything else is choice. I think it's important to have a believable standard. The ISO speed may not work for a particular individual, but that doesn't invalidate the method.
The almost universally obtained lower results from the personal testing is really the result of using two different methods that don't relate. It's really like one person using the standard prior to 1960 and another using the post 1960 standard. The results will have a constant 1 stop difference in speed between the two.
People who use the ZS testing method will obtain results that are consistent with the type of results obtain prior to 1960. Remember the ZS testing procedures haven't changed since the forties while the ISO speed standard has. Zone System speeds agreed more readily with the speeds obtained from the pre 1960 film speed standard. The results obtained with film shot prior to 1960 were excellent as are the results for the people who obtain similar EIs doing ZS testing today.
But, let's remember the reasoning for the speed adjustment in the 1960 standard. Many people were already automatically rating their films a stop faster. There was a two stop safety factor which was no longer necessary because of the improvement in exposure meters and lenses (lens coating included). Smaller cameras meant the additional exposure will have a noticeable affect on sharpness and grain because of the higher degree of enlargement as well as longer printing times. Now, because of the discrepancy between the two testing systems (because the ZS testing was never updated, and because of many other problems with it) people are obtaining EI speed values 1/2 to 1 stop slower than the ISO speeds. So, how does the concept of the pre and post 1960 speed example compare with the ISO and personal ZS speeds? It is the same concept. There isn't any difference. Speeds were once a stop slower and people complained. Now, they are a stop faster and people want to change back. Or do they? I don't think people really are aware this is the difference. Most incorrectly think they are obtaining more accurate speeds for their personal working conditions than the ISO speeds. Don't agree with my premise? Break down the Zone System testing method and compare it to the ISO method (you will also have to overlay the standard exposure model onto the ISO method). I think this could be educational for a lot of people. Personally, I do sensitometric testing with a calibrated sensitometer and obtain speeds relative to today's ISO speeds.
I have to disagree with the idea that "the ISO standard seems to differ from what photographers want to achieve, and therefore, it differs with customer 'reality', which is an unavoidable fact." It is way too speculative, vague, and subjective to be supportive. Jones found that any additional exposure above the first excellent print didn't show any appreciable increase in perceived quality. And the ISO standard is already one stop over the first excellent print. Even if you want to justify the use because of increased shadow contrast (which is valid), some could argue that the perceived increase in "quality" in a such a small area of the tonal range doesn't justify the loss of sharpness, increase in grain, or increase in printing times. Also, many factors can negate the "advantage" such as a higher flare factor or various paper curves. Saying that the customer wants a higher safety factor is something that can be supported.
I also think people are thinking they are getting the "real" speed of their film using the ZS testing method so therefore they think it is superior, but in actuality, they are just unaware of anything else. I mean, how diligent are most people with their testing? For example, I know most don't take into consideration latent image keeping. It's also almost impossible to do accurate real world confirmations on the results. One would have to be able to measure flare for that. What about those who test using a meter with a high red sensitivity under open shade (blue light) using a lens without 1/3 stop increments, obtains their personal EI then goes out and shoots a scene using an orange filter with a lens they didn't do the test with? How many do that unaware of all the influences to the exposure results (notice not film speed) yet are confident in their EI over the ISO because they determined it themselves so that it reflects their personal working methods? So how good are personal speeds when the assumed testing method could be over a stop off and/or imposes so many variables that it probably doesn't reflect actual shooting situations?
Basically, since you really don't see an appreciable increase in quality with the increase in exposure, the photographer not knowing this assumes the additional exposure is better because it reflects his approach and because some popular book told him so. They are happy with the results they get. They have less loss because of accidental underexposure, and they never do a head to head analysis because, let's face it, most of us aren't scientists and just want to get to the business of shooting. Thus, they conclude their system is superior to the ISO standard - a subjective conclusion.
What about different developers? The current ISO standard doesn't use a standard developer anymore. The T-Max films kind of initiated that change. The reason why they only had EIs when they first came out was because they did poorly in the ISO developer. Kodak ran their tests according to the ISO standard except for the developer used but since you can use the ISO prefix only when the standard has been explicitly followed, Kodak had to use EI. In order to reflect a better real world usage, manufacturers can use any developer they wish in their ISO testing but they must note it. Chances are Ilford films are tested in ID-11 and Kodak films are tested in D-76, T-Max, and/or Xtol. Anyway, most general purpose developers will produce speeds around the ISO speed. Specialty developers should be tested for, but again, how accurate can you guarantee the testing to be?
Now, take into consideration how film speeds don't really change much with the Delta-X Criterion / fractional gradient method and the need for testing for your personal EI really comes into question. This might sound cryptic to some, but it really is an important consideration.