   |
Dear Fred or Phil (or whomever),
I am writing another article on BTZS for a future issue of MAGNAchrom. (yes, yes I know that y'all are involved in the ViewCamera BTZS article, but this one will be significantly different and targetted to a different audience) So hopefully you won't have any issue assisting me on understanding how BTZS works "under the hood".
So here's my problem: so far I fully understand the underlying math of 99% of the core concepts behind BTZS -- after all, the concepts behind profiling your scanner/monitor/paper combination to achieve repeatable results are well-proven and essentially identical (in fact, much of the math is literally identical). There is no question in my mind that the BTZS methodology is both sound and rigourous. However, where I'm totally stumped is understanding precisely how (mathematically that is) the Effective Film Speed calculations are derived using the BTZS methodology.
Here is how the BTZS EFS calculation appears to me, a sceptical editor: the concept of the triangle appears to be a clever "hack" invented early on to enable early adopters of BTZS the ability to plot manually using nothing more than french curves and a pair of scissors. The idea of placing a triangle with a slope of 0.62 (btw, why is it the reciprocal of phi?) in the correct location allowed one to quickly obtain "good enough" estimate of the EFS without going into very complicated math. However, the underlying math is obscured by both the descriptions and the methodology. And as such, I can't help to think that perhaps there is some sort of voodoo math going on here. After all, isn't the slope of the individual curves more representative of the film speed than the curve's positiion horizontally on the X axis? And if the horizontal positioning is in fact how an accurate EFS is calculated, how come I an totally unable to replicate it using a spreadsheet approximation (unless there are some mystery constants being added to the equations)? (ironically, I have gotten very close using nothing more than the true average slopes of each curve). In other words: what is the scientific basis of the calculation and what is the underlying math? I'm from Missouri, sorry.
As part of my job as an editor, I feel it important to convey **IN LAY TERMS** how and why something works. The obscure descriptions of "placing the triangle in the family of curves" simply doesn't cut it. If I am going to be successful in my article on BTZS, then I have to eliminate such descriptions and go way beyond describing either a methodology or how a particular piece of software works. Otherwise, I have nothing more to offer than what has been written about BTZS umteen times before.
For example, consider the following representative data for a base film speed of 400 and a paper ES of 1.25 (this from one of your users of Plotter software and a big proponent of BTZS):
dev t speed G SBR
15 min 475 0.92 5.5
9 min 365 0.79 6.4
6 min 285 0.70 7.3
4 min 225 0.59 8.6
3 min 200 0.51 9.9
2 min 170 0.36 14.1
I cannot for the life of me understand how Phil derives his speed numbers (and mind you I'm pretty good with a spreadsheet). Hopefully, I'm just missing something simple. Can anyone on this great list shed some light here?
Cheers,
J Michael Sullivan
Editor/Publisher, MAGNAchrom
www.magnachrom.com
P.S. naturally, in return for any assistance I will give credit and links to www.btzs.org. Hopefully I too can help convert others to your methodology
|
|
|
|
|
|
|
