Michael, I think if you do a test where half the frame has a black card and half has a white card you'll find that the density of the black card will fall off slightly as you move away from the white card.
Better yet, substitute the sky for the white card.
Greater luminance range so the effect would be more pronounced.
Originally Posted by Michael R 1974
An interesting point about the speed point is that it's not necessarily where you want a specific exposure to fall - the shadows for B&W negative material and the metered exposure point for color reversal film for example. The purpose is to have a point that has meaningful information about the material. One of the functions of in the determination of the speed point is that it defines the limits of the material being tested. When the B&W ISO parameters are met, the minimum useful exposure that will produce a high quality print falls approximately one stop to the left of the speed point density of 0.10. If the processing is different, greater or less, then the relationship of the minimum useful exposure point and the 0.10 density point changes. In contrast conditions outside the ISO parameters, a different method should be applied such as the Delta-X Criterion.
Originally Posted by Rafal Lukawiecki
Film speed is then determined using the speed point, but as I said, exposure doesn't necessarily fall at that point. The ISO film speed is just a different approach to determining the fractional gradient speed point. It is basically the Delta-X Criterion. As we know, the fractional gradient speed point falls around a stop lower than the 0.10 density point, yet when this was the standard, film speeds were one stop slower than film speeds produced using the current standard. This sounds contradictory to our understanding on how speed works. Shouldn't film speeds increase the further left the speed point falls? One of the points missed in almost all general purpose photography books is that the speed point is only part of the determination of film speed. There is also the speed constant.
Film speed is part of the exposure equation used in exposure meters. This is why the ratio between the speed point and metered exposure point is so important. For black and white the speed the constant is 0.80 and for color reversal it's 10. This means that there is 1/3 stop difference between the color reversal speed point and the metered exposure point for a given film speed. With black and white film it's a 3 1/3 stop difference. The speed constant for color reversal films used to be 8. Without changing the speed point, the change in the speed constant caused a 1/3 stop increase in the film speed. Changing the constant instead of the speed point preserves all the important information about the material that the speed point is associate with.
This can be done with the black and white speed constant too. WBM uses a density of 0.17 as the speed point. With normal development, this produces approximately a 2/3 of a stop decrease in film speed compared to the 0.10 speed point, depending on the shape of the film curve. There's nothing wrong with wanting a little extra exposure as a safety factor or perhaps to achieve better results with your personal metering techniques. The problem with the 0.17 speed point is that it isn't in good agreement with the limits of the material. If you want the extra film speed, it's better to simply use the 0.10 with a speed constant of 0.50 instead of 0.80, or for those using relative film speed, adjust the EI by 1/2, 2/3, or whatever amount works best for you.
Last edited by Stephen Benskin; 09-01-2012 at 11:42 AM. Click to view previous post history.
Sponsored Ad. (Subscribers to APUG have the option to remove this ad.)
WBM has the best set of gradient values for diffusion printing that I've seen. I probably think this because they closely agree with mine. The difference is in how they were determined. Mine incorporates flare into the equation, WBM doesn't.
Originally Posted by Rafal Lukawiecki
The equation to determine the CI for any luminance range is:
CI = NDR / (log subject luminance range - flare)
The luminance range of a scene is reduced at the film plane by flare. The range you are developing for is the one that actually strikes the film. With a one stop flare factor, a scene will a 7 stop luminance range effectively becomes a 6 stop range at the film plane. If you develop for a 7 stop range, you would be under developing. One of the ways the various methods that don't factor in flare use to still produce good CI values or accurate processing is to use higher negative density range values. The average log exposure range (which equates to the aim negative density range) for a grade two paper printed with a diffusion enlarger is 1.05. WBM uses a 1.20 and the Zone System uses a 1.25 range. This effectively compensates for the lack of a flare adjustment to the luminance range. And while this approach will produce a good negative, it doesn't correctly reflect the process.
One consideration to maintain when factoring in flare into the equation is that the degree of flare isn't a constant. Flare varies with different luminance ranges. It's virtually impossible to determine the exact flare value for any given scene but flare tends to change by 1/3 stop per stop change in the luminance range. If the value of flare is kept constant, the CI values for shorter luminances will be too high producing contrasty negative and with longer luminance ranges, a fixed flare value will produce CI values that are too low. The developmental model that best reflects the shooting conditions would use a variable flare model.
Another factor that needs to be considered is that there isn't a perfect relationship between the negative density range and the paper LER. Loyd Jones found that “for the soft papers, the density scales of the negative (NDR) should in most cases exceed the sensitometric exposure scale of the paper (LER), whereas, for the hard papers, the density scales of the negatives should in most cases be less than the sensitometric exposure scale of the paper (LER).” In other words, it’s not only acceptable, but preferable to have a slightly contrastier negative for contrasty scenes and a slightly flatter negative for flatter scenes.
This is why my developmental model averages the fixed flare method and the variable flare method. In the following graph, all the models are based on a 1.05 NDR. An increase in the aim NDR with the no flare curve will shift it upward to be even with the others. The similar shape of the no flare and practical flare model curves is why the WBM with its higher NDR and no flare approach and my practical flare model agree so closely.
In practice with the use of variable contrast papers, personal taste, the uncertainty of the flare, etc, how critical are these distinctions? In most situations, not very. There's always been a large range of acceptability in photography. Afterall, it's an art form that uses science. I just think it's important to understand how it really works.
Last edited by Stephen Benskin; 09-01-2012 at 01:04 PM. Click to view previous post history.
Michael, no need to apologise, your questions about flare testing conditions are very relevant to my question.
Stephen, thank you for our extensive explanations and the time you put into them. I feel like I would like to back up my understanding of practical sensitometry with some additonal studying. You mentioned threads on which you explained aspects of it, would you be so kind to recommend also any other relevant reading?
I am looking forward to retesting when I am back in Ireland from my current trip to Wyoming. My overall goal of being able to find a working set of development times and EI, based on a set of sensibly chosen CIs, still remains. I like the completeness of the WBM approach, but I would also like to hear your suggestions as to the alternatives. I suppose I also ought to do a paper response test to validate my chosen CIs at some stage, but I am not sure how much of a personal variable those would happen to be.
Hope you are having fun shooting Wyoming!
I am working on a revised graph of your data with a compressed LogE scale in the toe to counteract the effect of flare.
I think it's the necessary approach to graph a test that was shot with flare.
One interesting thing my new graph will do. I will end it abruptly at the speed point for 320. That's because effectively flare guaranteed there isn't any step wedge sample that received "less" light than that.
Thanks, Bill. I am amazed by the amount of care, insight, and sheer time you, and others, have been putting into helping me understand, and control, film testing. I have never realised how complex the subject was, until I have tried the procedure. The funny thing is, that with the material being so forgiving, I realise the exactitude of such testing is probably not necessary. I just happen to be obsessive enough to want to know the answers. I am very grateful to all of you for your assistance. Thank you!
PS. Wyoming is amazing. Leaving the high, people-free, mountain wilderness areas tomorrow, for a week in the Tetons and Yellowstone.
Last edited by Rafal Lukawiecki; 09-02-2012 at 11:10 AM. Click to view previous post history.
Reason: Added PS.
I keep wanting to post a comparative flare/low flare test for illustrative purposes (my personal testing is low flare which matches the conditions of most of my photography) but interestingly I'm having quite a hard time coming up with a suitable test with flare, one that is otherwise identical to or at least similar to my low flare tests. Intuitively this seemed simple to construct, but I'm still stumped. Essentially I need to photograph some type of calibrated grey scale against a white background but this is not as easy as it sounds. Just some thoughts on test conditions.