CPorter -
Right, 0.1 log E units above fb+f is the "practical" determination of film speed. Why, because, IMO, it is within the response of the paper's emulsion to be able to produce a tone just perceptibly lighter than Dmax------what we call Zone I, there is no shadow detail in a Zone I exposure, only enough useful density on the negative for the paper to print a Zone I print value. Negative density outside the response capability of the paper's emulsion seems useless to me.
What matters is what can be printed with todays papers. Who cares about any measurable density below fb+f density. What value is it to anybody with a film camera in their hands?
If by "useful" you mean that a -0.29 log E unit can be printed on paper, well I guess I would have to see it to believe it. I can't get any thing on paper below fb+f, paper that I buy anyway.
First of all, density is not the determining factor for determining print quality. It is contrast. That is why there was the fractional gradient method of determining film speed and why there is the Delta-X Criterion method of determining film speed which is incorporated into the current ISO standard. The fractional gradient method had the speed point where the curve’s gradient was 0.3x the average gradient of the curve. Loyd Jones found that contrast determined quality in his historic First Excellent Print Test back in the late thirties. As the quote I posted earlier states, film speed is not a fundamental concept. It is based on the ability to produce perceived high quality results in the print. That means there is a psychological element to it. Up until Jones, no one took that into consideration. According to W. F. Berg in is book
Exposure Theory and Practice,
“with this criterion, (Jones) completed the circle which started from a purely academic linking of the speed with an almost entirely arbitrary property of the characteristic curve, then led us to the purely practical concept of speed chose criterion based.”
The fraction gradient point was chosen because it was the base point of acceptable exposure, but it was never intended to be the point where any exposure was to fall. That would leave no room for underexposure error. It was used as a speed point, or the point where the speed of the film is determined. Take for instance that for normal development, the fractional gradient speed point falls around a stop lower than the 0.10 fixed density point which would indicate the film speeds should be one stop faster than if you used the 0.10 fixed density point; but in actuality, they were one stop slower. How? Once the point was found, it was plugged into an equation. The amount of exposure necessary (in meter candle seconds) is divided into a constant. If the constant is 1, then the speed will reflect that exact position, if not, it will shift the placement. Transparency film speed is determined on the curves mid-point which is around 10x higher than the amount of exposure need to obtain the 0.10 fixed density point for a given speed yet transparencies don’t have speeds 10 times higher than b&w films. They use a different constant in the equation. A given speed point for a type of film is based on the particular characteristics of the material and the desired results, then the speed equation makes it possible to determine their exposures with a single type of light meter instead of having special light meters for each type of film (color neg, transparency, litho, aero, etc).
This is no different with the 0.10 fixed density method today. 0.10 isn’t necessarily the aim point of density for the shadow exposure. According to the standard model, the statistically average scene is 7 1/3 stops. The highlights fall around 0.93 log units above the (~3 ¼ stops) above the mean log luminance (meter exposure) and 1.27 log units below (~4 1/3 stops). Well, as the 0.10 fixed density speed point falls only 1.0 log units below the meter reading wouldn’t that mean the shadows will fall about one stop lower? Yes they should, but average flare from the average scene is 1 to 1 1/3 stops. The reason for the range is because the standard model uses one stop while real world average for 35mm users is around 1 1/3. One stop flare brings the exposure up just above the 0.10 speed point and the real world flare takes it up 1/3 stop higher. Add to that the constant in the b&w speed equation is 0.8 and not 1, that adds 1/3 more stop to the exposure (1/.0064 = 156 vs 0.8/.0064 = 125 a 1/3 stop difference). In fact, the standard model pretty much agrees with Ralph's placement of the shadow placement at a higher density, but for different reasons. Not only is the target for shadow exposure not at a density of 0.10 but that there is an underexposure safety factor of at least 1 1/3 stops.
What most people don’t know is that the fractional gradient method wasn’t really eliminated in favor of the fixed density method. There is an equation built into the current ISO standard that effectively makes the fixed density method equate with the fractional gradient method, but only when certain contrast parameters are met. The reason why the ISO standard requires a very specific set of contrast conditions (Δ log-H of 1.30, and Δ log density of 0.80, the +- 0.05 has been eliminated) is because under those conditions, there is a correlation between the fixed density method (which is easier to determine) and the fractional gradient method (which is more accurate). Only under those conditions does this happen. In T.H James’ and George C. Higgins book
Fundamentals of Photographic Theory they write,
“This procedure (Delta-X Criterion) is to be used as the new American standard method of determining speed (1960 standard), and since it incorporates the fixed-density feature of the DIN method or speed measurement there is a possibility that it will be adopted internationally.” Film developed outside the contrast parameters shouldn't use the fixed density method for accurate speed determination but should use the Delta-X equation. The reason for this is explained in the next paragraph. The Delta-X Criterion is perhaps the biggest secret in photography today and not understanding it leads to misinterpreting film speed which is almost universal.
CPorter - Why, because, IMO, it is within the response of the paper's emulsion to be able to produce a tone just perceptibly lighter than Dmax.
The paper black method of printing isn’t really a valid way of determining proper film speed when you take into consideration that density isn’t deciding factor. One of the reasons why the fixed density method wasn’t accepted as the preferred method to determine film speed was because, as C.N. Nelson writes in his paper
Safety Factors in Camera Exposures,
“The fixed density criterion tends to underrate films that are developed to a lower average gradient and overrate films that are developed to a higher average gradient.” The fractional gradient method has the speed point move up and down in relation to the 0.10 fixed density method depending on the over all contrast. As it is the shadow contrast that determines quality, that means the overall density of the film will increase and decrease depending on the development. This means there can’t be a specific print exposure to determine “correct” shadow except normal development. This renders that concept of paper black moot as an argument for proper film exposure. However, the method is a good learning tool and can be helpful for normally processed films, but it isn’t a rule or good theory. It’s just conventional wisdom.
The problem with using the Zone System is that it isn’t real sensitometry nor is it part of exposure or film speed theory. It is a short hand for exposure. Adams had nothing to do with the establishment of film speed. He wrote no scientific papers and is not considered an authority. In fact, he acknowledges Mees (read Jones) for their assistance (which he misinterpreted) in helping develop the Zone System. So, when you talk ZS, you are only talking about exposure and not film speed.
This is why I think it’s important not to confuse film speed with EI or exposure. Ralph is not talking film speed when he says to rate the film 2/3 stop higher. He is talking about exposure. I agree with him that additional exposure can’t hurt and in many cases helps, but too many people get confused between a good rule of thumb exposure suggestion, like Ralph’s, and the actual speed of the film. It’s important to know the difference so that the decisions you make have a strong foundation. I know that depending on the scene’s flare, with 80% coming from the subject, there is a potential for the shadows to fall within a two stop range of the speed point. This is film speed. From this, I don’t attempt or expect to nail a specific point of density with my exposure because I know it’s virtually impossible. That is exposure.
CPorter - What matters is what can be printed with todays papers. Who cares about any measurable density below fb+f density. What value is it to anybody with a film camera in their hands?
If by "useful" you mean that a -0.29 log E unit can be printed on paper, well I guess I would have to see it to believe it. I can't get any thing on paper below fb+f, paper that I buy anyway.
I'm sorry, but you’ve totally misinterpreted what I was saying. I never said anything about densities below fb+f. The 0.29 log units was log-H units (not density units) below the exposure required to make 0.10 density above film base plus fog.
P.E. - In other words, all zones must be on a straight line portion of the H&D curve that does not include bows upwards or downwards, nor should it contain any portion of the toe or shoulder. This is the way to measure a negative film and how to get the best photos.
This kind of sounds a lot like the concept behind the inertia speed method which was unanimously rejected by the Ninth International Congress of Photography in 1935 because the inertia speeds varied as much as 500 percent with increasing gamma. It also didn’t work because like Berg said, it was
“a purely academic linking of the speed with an almost entirely arbitrary property of the characteristic curve.” It also goes against findings with psychophysics and perception. I've probably just misread your post.
Ralph - below Zone I or 0.1, negative density is too low to print, and consequently, cannot be called 'useful' in my opinion.
The thing is, it is just an opinion. It is not film speed theory. I think it's important for people to understand the difference between when someone is giving an informed opinion and when it is established fact. Jones found that “first excellent print” could be made from a negative where the shadow density fell about one stop lower than 0.10 density point. The "useful" came from Jones who conducted those tests and from which all modern film speed, exposure, and tone reproduction theory is derived. As far as I know, no one has ever said that 0.10 is the minimum useful density in any scientific paper. It's only in popular texts.