Utter reciprocity confusion

N

nephilim

Established
Local time
5:15 AM
Joined
Feb 8, 2010
Messages
59
Going to take some long exposures in a few weeks but I've never done this with film before. So I went to the Kodak website and had a look at the data sheets for 400TX, TMX, TMY and the Portras. The first confusing moment was when I digest the TriX sheet which says

indicated 1s -> add 1 stop -> 2s (fine)
indicated 10s -> add 2 stops -> 50s (? isn't it 10s *2 *2 = 40s?)
indicated 100s -> add 3 stops -> 1200s (??? 100s * 2 * 2 * 2 = 800s)

On the other hand, the TMAX 100 & 400 sheet are comprehensible.

Next I found Gainer's method (http://www.apug.org/forums/forum37/11566-reciprocity-misbehavior.html) which adds to the confusion.

According to Kodak:
400TX @ 10s -> 50s
TMX @ 10s -> 15s
TMY @ 10s -> 15s

and

400TX @ 100s -> 1200s
TMX @ 100s -> 200s
TMY @ 100s -> 300s

According to Gainer, the corrected exposure is a function of the coefficient tc,1 only. These are

400TX: 0.169
TMX: 0.069
TMY: 0.061

Using this formula (Tcorr = Tmeasured + tc,1*(Tmeasured^1.62)), 400TX @ 100s should be 400s, which seems to be very low. Moreover, TMX has a greater tc,1 than TMY and therefore a shorter corrected exposure. This completely contradicts Kodak's sheets, both at 10s and 100s.

Please tell me, I'm stupid and made a terrible mistake.

Cheers,
Torsten
 
Torsten, since you have a few weeks before you take the actual pictures, I'd suggest that you do some testing, and determine the actual exposure by experiment rather than relying on conflicting theoretical numbers.
 
Never heard of the "gainer method" before - but I am somewhat sceptical regarding an attempt to reduce such a complex multi-factorial phenomenon as real world film reciprocity (there must be dozens of substances with each their own reciprocity factor in a modern film) to one single factor. The tables published by Kodak et.al., which usually are measurement results, are far more complex than that.

When stating actual times, the makers will probably use the data from their measurements rather than coarse rule-of-thumb multiplication factors, so that approximate factors and sample times will differ. In any case, be aware that the data sheet only provides starting values for your own trials.
 
Some explanation: The Gainer method is a fit of data gathered from a bunch of experiments with different films by Howard Bond back in the first half of 2000. Both data and fit were published in Photo Techniques magazine. Apparently, all b&w films behave in a similar way and allow for this seemingly simple fitting formula. A guy at apug.org did a comparison with another film database for astrophotography purpose and Gainer's formula again agreed very well. Hence, I hoped I wouldn't need to do the experiments but rely on the results of others :)

However, one thing that came to my mind that might explain one of the discrepancies above is, that the emulsion nowadays is TMX-2. I hope I can find the sheet of the old TMX somewhere... And honestly, the difference in seconds might sound large but in terms of stops it's not that large.

UPDATE - Ok, found the old TMX with identical exposures as the new TMX-2. Interestingly, PlusX has identical numbers and plots as TriX. I'm wondering if Kodak updates these info for each film.
 
Last edited:
Kodak and presumably other manufacturers wanted to be as accurate as possible to keep their customers happy. You should be able to use their data as a starting point.

But as Chris101 said, you need to experiment for yourself. Your personal preferences as well as the accuracy of your photo gear will need to be accounted for, and only experimenting can do that for you.
 
nephilim said:
Going to take some long exposures in a few weeks but I've never done this with film before. So I went to the Kodak website and had a look at the data sheets for 400TX, TMX, TMY and the Portras. The first confusing moment was when I digest the TriX sheet which says

indicated 1s -> add 1 stop -> 2s (fine)
indicated 10s -> add 2 stops -> 50s (? isn't it 10s *2 *2 = 40s?)
indicated 100s -> add 3 stops -> 1200s (??? 100s * 2 * 2 * 2 = 800s)
Click to expand...

Hmmm... I thought they usually mean (and sometimes it is written on the datasheets too) that you should open up the lens given amount of stops for the given amount of time. This is because if you make the exposure longer the reciprocity failure will get worse even if you make it longer because of the reciprocity failure. By keeping the time constant, it is a simpler equation.
 
The Kodak material says to either adjust by n stops OR use the increased exposure time.

I tend to use the formula as starting point, bracket in the field (including the recommendations from the Kodak sheets) and make notes.
 
Elektrojänis said:
Hmmm... I thought they usually mean (and sometimes it is written on the datasheets too) that you should open up the lens given amount of stops for the given amount of time. This is because if you make the exposure longer the reciprocity failure will get worse even if you make it longer because of the reciprocity failure. By keeping the time constant, it is a simpler equation.
Click to expand...

Making the exposure longer does not compound the reciprocity failure (I believe).

Ronnie
 
Thanks for the link. That's exactly the data, Gainer used for the formula posted above. And it also sorts out the discrepancy with TriX.
 
nephilim said:
Thanks for the link. That's exactly the data, Gainer used for the formula posted above. And it also sorts out the discrepancy with TriX.
Click to expand...

I heard an interview with a Kodak engineer discussing TMY-2. He stated that the only thing to change from the old TMY was a perfection/reduction in grain size. He didn't say that reciprocity would be different but I 'feel' it is the same. Gainer's formula was just too much for me so I copied Bond's chart and take it with me.
 
I got a free calculator app for my android phone and saved the formula for each film I use. I hope my long exposure sessions won't last longer than the phone's battery ;-)
 
Peter^ said:
can never find the dials on those damn things
Click to expand...

Do they make one (?), that's what I want; to replace this, my old phone; which I gave to my daughter (I shouldn't have: a couple of years ago when I updated to a Princess):

2653602827_a9e311b326.jpg
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom