Are you reading the first or second version?
Concerning exposure and film speed theory, can you outline some of the issues? Do you have trouble with how he frames the various historical methods and their underlying principles/assumptions? Fractional gradient? Or is it his method for an in-camera test for "effective EI"? Etc.?
Michael,
I have the second edition. So far, Ive only been able to review some of Chapter 6: Film and the Making of a Negative, and scan through the rest. Havent read this book in more years than Id like to admit. Still hate the formatting with the dense type, but now Im even more impressed with the sheer volume of testing and detail.
In order to help keep this from getting too long, Ill touch on a few points I found interesting. The section on page 137 Geometric Variation in Illumination of the Film in the Camera nicely covers an important topic rarely touched on. He notes that the average light loss at an optical angle of 15 degrees off axis averages around 20% (12 degrees is the value the exposure meter standard utilizes for calculating K and q). Depending on the focal length and f/stop, the fall off can be easily ½ stop with over one stop not uncommon.
While Henry doesnt touch on it, the first thing that came to my mind was how this information would effect in camera testing with a contacted step tablet. The portion of the step tablet / film where the degree of accuracy is most critical falls towards the outer edge of the image circle where the greatest proportion of light loss occurs.
His summary of the history of film speed determination was good up to the fractional gradient part and most importantly the relationship between the fractional gradient and the current method. I believe he need to flesh out more about the concept of film speed and the reason why the fractional gradient was chosen. Even though Henry references the Nelson and Simmons paper that introduced the Delta-X Criterion, which the current ISO speed method is based, he gives the impression that using a fixed density of 0.10 alone has a valid correlation to the fractional gradient method. I dont know if he just muddled this part or meant to imply the fractional gradient method and its principles were abandoned for the fixed density method, but thats what Im left with.
One page 142, Henry wonders why in the 1979 speed standard there was a statement in the forward, not considered the official part of the standard, that read, For any specific developer, the sensitometric speed rating method of this standard includes a 1/3 camera stop underexposure safety factor. He was rather puzzled at this and stated that 1/3 stop overexposure was built in, and why wasnt it in the standard?
I think hes making more of this than is necessary. Between the fractional gradient standard and the 1960 ASA standard, the safety factor was reduce from around 2.35 to 1.2, as explained in Nelsons
Safety Factors in Camera Exposure, which is one of the two papers where the 1960 ASA speed standard was derived. Henry does talk about the 1.2 (1/3 stop) safety factor on the previous page. I believe the reason it was listed in the forward is because the safety factor is based on the statistically average scene luminance range which isnt a constant so the point where the shadow falls cannot be consistent and therefore neither can the safety factor, so such a statement couldnt appear in the body of the standard.
These might seem like minor examples (and they are), but they help evaluate his knowledge of theory (and mine). I havent gone through the entire book yet, but there are a number of things I really dont agree with and yes one of them is his preference for in camera testing, incorrect concept of metered exposure point, using the fixed density method to determine film speeds outside the ISO parameters, and more. Which we can go into in more detail later.