Velvia 100F: Reciprocity Failure

I have been debating how to interprete my results of longer exposures up
to 64 seconds under daylight-balanced fluorescent light(D50). My early
interpretations and images are online at:

http://www.ephotozine.com/equipment/...fm?test_id=314

I have subsequently created a new comparison using the same data as
shown below:

http://www.tomoko-yamamoto.com/test/Wilten.html

The background color is close to cyan rather blue.

My question is how one can identify multiple color shifts if any exists.
I first interpreted my results with a wrong idea that a color shift in
reciprocity failure of Velvia 100F was green. According to the Fuji
literature, it is supposed to be yellow.

Can my results be interpreted with a green cast plus a small but yet
progressive yellow shift? Can the green cast worsen upon longer
exposures? Does the yellow shift only manifest itself after 1 minutes
or does it start before 1 minute?

The results I am presenting are from within the same roll.

Since this is an online review, the editor at ephotozine.com will allow
me to update my review.

Tomoko Yamamoto
http://www.tomoko-yamamoto.com/

Tomoko Yamamoto wrote:
My early interpretations and images, and later data, are online at:
http://www.ephotozine.com/equipment/...fm?test_id=314
http://www.tomoko-yamamoto.com/test/Wilten.html

Thanks, interesting review.

My question is how one can identify multiple color shifts if any exists.
I first interpreted my results with a wrong idea that a color shift in
reciprocity failure of Velvia 100F was green. According to the Fuji
literature, it is supposed to be yellow.

Can my results be interpreted with a green cast plus a small but yet
progressive yellow shift? Can the green cast worsen upon longer
exposures? Does the yellow shift only manifest itself after 1 minutes
or does it start before 1 minute?

I think Fuji is wrong and you are right: this looks like a green shift,
mostly in the highlights. I thought Velvia 100F did not have as much
resistance to reciprocity failure as Provia 100F, but the 64 second
exposure looks relatively good to me.

The way I got colors in the the left-hand 16-second image to match those
in the right-hand 64-second image was to decrease brightness by 24,
increase contrast by 12, then increase Green (highlights only) by 28,
using GIMP. The hat-top white feather is still too bright and must be
selected and have green saturation reduced.

It might be a good idea to do this test with a Macbeth chart in various
types of lighting (full sunlight, open shade, electronic flash).

Thanks for your feedback.

I have tried to emulate the 64-second version by modifying the 16-second
version using Photoshop.

I tried your formula using Photoshop. I think it comes close except for
the hat. I could also get a close match using a combination of green
and yellow, but the hat was a problem.

I have just bought my first film scanner from B&H and in the next few
days I hope to be able to scan a few more frames from my test including
32-second and 8-second exposures. With more data, I might be able to
figure out whether there is any yellow shift as well.

Bill Tuthill schrieb:
Tomoko Yamamoto wrote:

My early interpretations and images, and later data, are online at:
http://www.ephotozine.com/equipment/...fm?test_id=314
http://www.tomoko-yamamoto.com/test/Wilten.html


Thanks, interesting review.


My question is how one can identify multiple color shifts if any exists.
I first interpreted my results with a wrong idea that a color shift in
reciprocity failure of Velvia 100F was green. According to the Fuji
literature, it is supposed to be yellow.

Can my results be interpreted with a green cast plus a small but yet
progressive yellow shift? Can the green cast worsen upon longer
exposures? Does the yellow shift only manifest itself after 1 minutes
or does it start before 1 minute?


I think Fuji is wrong and you are right: this looks like a green shift,
mostly in the highlights. I thought Velvia 100F did not have as much
resistance to reciprocity failure as Provia 100F, but the 64 second
exposure looks relatively good to me.

The way I got colors in the the left-hand 16-second image to match those
in the right-hand 64-second image was to decrease brightness by 24,
increase contrast by 12, then increase Green (highlights only) by 28,
using GIMP. The hat-top white feather is still too bright and must be
selected and have green saturation reduced.

It might be a good idea to do this test with a Macbeth chart in various
types of lighting (full sunlight, open shade, electronic flash).

Where can one get a Macbeth chart? By the way I have once shot the
background only inside and in sunlight. In sunlight the background
looks less green on Velvia 100F.

Tomoko Yamamoto
http://www.tomoko-yamamoto.com/

Tomoko Yamamoto wrote

[SNIPPED GOOK DRIVEL]

This is an American forum. No Gooks or Chinks allowed.

Tomoko Yamamoto wrote:

I tried your formula using Photoshop. I think it comes close except for
the hat. I could also get a close match using a combination of green
and yellow, but the hat was a problem.

You are right, the hat fabric it too green under my formula, and the
yellow flower is oversaturated. This is why reciprocity is called
"failure" -- because it messes up color balance.

Where can one get a Macbeth chart? By the way I have once shot the
background only inside and in sunlight. In sunlight the background
looks less green on Velvia 100F.

You're saying that Velvia 100F emphasizes greens more indoors than
in sunlight? Maybe it was the fluorescent lighting, which is usually
greenish. You can buy a Macbeth color checker for under $70 at any of
the NYC retailers. The B&H website has a picture of one.

Bill Tuthill wrote:
Tomoko Yamamoto wrote:

I tried your formula using Photoshop. I think it comes close except for
the hat. I could also get a close match using a combination of green
and yellow, but the hat was a problem.


You are right, the hat fabric it too green under my formula, and the
yellow flower is oversaturated. This is why reciprocity is called
"failure" -- because it messes up color balance.


Where can one get a Macbeth chart? By the way I have once shot the
background only inside and in sunlight. In sunlight the background
looks less green on Velvia 100F.


You're saying that Velvia 100F emphasizes greens more indoors than
in sunlight? Maybe it was the fluorescent lighting, which is usually
greenish. You can buy a Macbeth color checker for under $70 at any of
the NYC retailers. The B&H website has a picture of one.

The difference is very small, but if I look at the transparencies under
a loupe, there is a subtle difference. The indoor shots show more cyan
than the outdoor shots.

By the way I realized that I had seen Macbeth chart tests on six films
(three new Fuji films including Velvia 100 and three Kodak films) in a
Japanese photo magazine. Compared to the picture on the B&H website,
the magazine article shows mostly overexposed charts. I don't know
whether this was due to printing or what, but the black on the chart is
not coming out black on the Fuji films, but is more black on the Kodak
films. Yet the picture captions say that the black hair of a woman
model, who is holding the Macbeth chart, is not good on the Kodak films
compared to the Fuji films.

Tomoko Yamamoto
http://www.tomoko-yamamoto.com/