Velvia 50 and 100 Reciprocity Correction

Hello All,

This may seem retarded and all, but I'd love to know the answer in a plain way. For example, with Ilford Delta 100 and PanF the way to find out your adjusted shutter speed for measured times greater than 1/2 second is to do this algorithm:

Y=1.465^X+1 (Measured Time + 1 to the power of 1.465 = Adjusted time - where 120 seconds adjusted is 18.76 minutes)

Does anyone know this way of figuring out the adjusted time for Velvia 50 and/or 100? The only thing I can get off of the Fuji website is to use a magenta filter and to add 1/3 of a stop to 2/3 of a stop. But that should be just to offset the filter it recommends.

Thank you!

-Brian

The corrections listed on the Fuji site are to both correct for reciprocity failure and for the color shift that occurs with long exposures.

For Velvia 50 you need to increase the metered exposure by an extra 1/3 stop at 4 seconds (or the aperture); at 8 sec. 1/2 stop; 16 sec. 2/3 stop; 32 sec. 1 stop:

http://www.fujifilm.co.uk/profession...e_velvia50.pdf

I already understand that way of doing it as I noted in the last part of the first post. Now, if I'm not using the filters (As I intend to fix the color shift in Photoshop) do I need to bother using the extra stops as the stops are there, in my opinion, to offset the filter density? Or should I do it anyway - even if I'm not using the filters?

Thanks!

-Brian

Quote:

---

Originally Posted by Brian_A

I already understand that way of doing it as I noted in the last part of the first post. Now, if I'm not using the filters (As I intend to fix the color shift in Photoshop) do I need to bother using the extra stops as the stops are there, in my opinion, to offset the filter density? Or should I do it anyway - even if I'm not using the filters?

Thanks!

-Brian

---

Yes, you need the extra exposure to account for the reciprocity failure, with or without the color correction.

Quote:

---

Originally Posted by Ron Marshall

Yes, you need the extra exposure to account for the reciprocity failure, with or without the color correction.

---

The Fuji recommendation does include the adjustment for the filter. However, if you are excluding the filter at image capture then you simply need to strip out the filter adjustment factor. I measured this at about 1/3rd of a stop for a 10M filter. When NOT using a filter I have used the following RF adjustments:

4s - no adjustment, 8s add 1/3rd stop, 16s add 1/2stop, 32s add 2/3rd stop.

Obviously I've rounded the adjustment to the nearest 1/3rd of a stop. Incidentally, instead of using the M series of filters you could try either a Lee 85c, or the Tiffen 812.

Cheers

Paul

Brian, based on experimental evidence acquired by another photographer my father plotted and derived the following formula.
Corrected = 0.90842 * Indicated^1.24
Applies only once the indicated exposure hits about 2s and agrees pretty closely with the information posted by Ron and Paul.
Should note the resulting value assumes a colour correction filter is in place that will have a filter factor of 1.3 to 1.5... I may see if we can work out a formula for unfiltered.

I should add that is for Velvia 50.

I always wondered why reciprocity failure wasn't plotted to an equation just like you posted. Thanks, I will definitely be using this!! Do you have any other equations for other films?

Brian...My personal guidelines for Velvia 50...I rate the film at asa 50...I do not use color correction filters...4 sec-5, 8 sec-11, 16 sec-24, 30 sec-54, 60 sec-115, 120 sec-260...The longer exposures require a bit of intuition as I find the light values often change during exposure...This has worked for me for years...As to the 100 speed films I have no experience with them...Hope this works for you Doug

An almost perfect fit for the reciprocity failure of Ilford black & white films (they provide the same graph for every film) is the equation:
y = 1.953x + 0.1064x^2
Where x is the metered exposure and y is the corrected exposure.
I have a table of corrected values for Provia 100F I will plot and derive an equation, if anyone has any other verified film reciprocity data for example Kodak Ektar, Kodak E100G, Kodak Portra 400 or any other film for that matter, please share :)

Provia 100F is a very stable film, so testing is a little tricky. From all the measured data I've been able to find, the best fit for it's reciprocity characteristics is as follows:
y = 0.9727x + 7.629E-7x^3
The E means it is *10^-7 which makes it very small... 0.0000007629 to be precise. Sorry bout that! It's what happens when a film has very very little reciprocity failure.

Here is an approximation for unfiltered non-colour-corrected velvia.

y = -0.9718 + 1.088x^1.1
Sure makes Velvia 50's reciprocity failure not seem as severe as before!

It's not as neat because filter factors are hard to *extract* from exposure information. Obviously as with all of this stuff it is all somewhat fuzzy but should be 'good enough'.

Kodak Ektar 100 fits a very nice curve, from the only available test for reciprocity I was able to find. This is unfiltered of course.

y = 0.759x + 0.555x^1.3

Most information seems to specify that Kodak E100G doesn't suffer noticeable reciprocity failure even at 10 minutes so I won't bother plotting that one.
Portra 400 has scarce information but being a 400 speed film I probably won't do any long exposures with it.

That's all the films in my fridge atm, so unless someone else wants a particular film plotted, I've done all I need to :)

Thanks, great info! Do you know if the old Provia (RDP II) has the same or similar reciprocity characteristics as the newer? I have 3 old boxes I'd like to experiment with doing long exposures.

Old Provia RDP-II has worse reciprocity characteristics and colour shift than Provia 100F. Fuji recommend 1/2 stop at 32s and a full stop at 2min with RDPII... Whereas 100F has essentially no reciprocity failure until around 6 minutes and even then it's only about 1/10 stop.

With only the information of no correction up to 16s, 1/2 stop at 32s and 1 stop at 120s I can derive an equation of
y = 0.6046x ^1.25
For RDP-II. Good luck! :)