Pushing and Pulling Transparency Film

[marcookie]

I made a video on the subject, comparing N-1, N, and N+1 development.

I will also write down my findings and summarize them in the post below.


https://youtu.be/yyGpNA0YRv0

If interested, please see this other post on the possibility of recovery highlight details by under-developing Velvia 50. (https://www.largeformatphotography.info/forum/showthread.php?151855-Pulling-Slide-Film-to-Increase-Its-Dynamic-Range)

[marcookie]

Hello,
This is a written synopsis of the video above.

Following up from my previous post on pulling Velvia 50 to recover some information from the highlights, I conducted some more “accurate” tests on the effect of development on transparency film, in this case on Fujifilm Provia 100F.

Fuji states in the “manual” of this film that it can be pulled up to ½ (half) stop, and pushed up to 2 stops.

This is how I conducted my test: I exposed three sheets of 4x5 Provia 100F for the same amount of time, and then asked the lab (PrausProductions/4photolab) to expose them at N-1, N, and N+1 development.

I gave the same exposure to three sheets of 4x5 film because I wanted to see the effect of changing only one variable: development time (I am actually not sure if the lab achieves a different development strength by changing the time or the temperature of development).

The scene was low in contrast, and it was exposed from a reading of Zone IV (darkest leaves in the top part of the images below the grass) to a reading of Zone VI (lightest yellow leaves in the center and in bottom left corner).

I first scanned the three transparencies by using the same level adjustments in the scanner, so that to have an “absolute” comparison.

https://i.ibb.co/mt61K5r/absolute.jpg

It is apparent that the development has a very large effect. The N-1 is dark and shows color shift—not surprising because Fuji does not recommend going below N-½ development.

The N looks quite normally exposed as it should (I did not mess this up too much!).

The N+1 looks super bright and contrasty—the highlights have been pushed with development to Zone VII or more, very close to the limit of the film dynamic range, and therefore they show barely any texture. In fact, the histogram shows that there is loss of detail in the highlights. The contrast is also off the chart—it seems that the image was captured in full sun, while the conditions were very dull. This is not bad, and it appears that pushing slide film allows for unprecedented levels of contrast that I have never seen in a one-stop push in negative film. Just be aware that what meters in Zone VI may already risk going too bright in the final transparency. The color of the N+1 looks quite true to how I remember the scene, just very bright and therefore apparently lower in saturation.

I then played with the levels to get the images as close as possible in overall brightness, so that to better compare the colors.
https://i.ibb.co/47bsHx1/relative.jpg
The color shift of the N-1 is clearer here. The greens are less “green” in the N-1: both red and blue hues seep into the grass color (I also checked with the 11x11 average eyedropper tool). The warm tones look more orangey to me.

https://i.ibb.co/kSYMtS7/d-n-1010-0-33x.jpg

The N+1 looks brighter, but overall with a similar tonality compared to the N.

I also checked film resolution with a good 10x loupe but found no apparent difference in the three transparencies.
I hope you find this interesting and/or helpful.

Please also consider watching the video posted above, where a go a little more in-depth showing the images and discussing the interpretation based on the zone system (well, my qualitative interpretation at least—I don’t do densitometry but I find it useful to have some relative/personal references using the zone system).

See this link (https://www.largeformatphotography.info/forum/showthread.php?151855-Pulling-Slide-Film-to-Increase-Its-Dynamic-Range) if you want more (qualitative) information on pulling Velvia 50.

Any comment or suggestion is appreciated.

[Tin Can]

Excellent, thank you!

[Kiwi7475]

Good work, thank you for sharing this!

[Corran]

Good work. Previously I had thought you were developing yourself. It would be interesting to know what is changing and how much for these results. Tonight I dev'd some Velvia 100 and my temps got a little hot, resulting in similar results to your N+1. I will remember to keep highlights at Z5 for pushing.

[Drew Wiley]

A correct pull/pull experiment would take into account an alteration in exposure to the effect that the offset in development would lead to equivalent midtone values. Tweaking brightness afterwards is just a disguise. It doesn't tell you, in this low-contrast case, what would really happen to shadows and highlights at the ends of the scale. Using Zone System terminology further confuses the subject. Therefore I don't really see any "pull" effect in the "N-1" example, but just a case of underexposure. But it's hard to make critical evaluations about web images. You should start with a standardized target with a full scale of neutral values (a gray scale) plus a representative selection of colors, like a standardized chrome of a MacBeath Color Checker Chart. Then adjust your exposures or effective ASA's to obtain matching middle-gray values in the respective push or pull examples. That way any drift or crossover at the extremes will be readily apparent. You choice of simply bright green and saturated red (plus some orange leaves) makes any objective evaluation quite difficult. Neutrals will tell you a lot more. (Not intended as a criticism per se, but a hint how to proceed with your experiments for more objective results).

[marcookie]

Good work. Previously I had thought you were developing yourself. It would be interesting to know what is changing and how much for these results. Tonight I dev'd some Velvia 100 and my temps got a little hot, resulting in similar results to your N+1. I will remember to keep highlights at Z5 for pushing.

I emailed the lab and asked them how they achieve push/pull. They have been helpful in the past maybe they will share something.

Anyway I would assume that time and temperature would have a similar effect (if appropriately controlled), as it is for b/w negative film.

[marcookie]

A correct pull/pull experiment would take into account an alteration in exposure to the effect that the offset in development would lead to equivalent midtone values. Tweaking brightness afterwards is just a disguise. It doesn't tell you, in this low-contrast case, what would really happen to shadows and highlights at the ends of the scale. Using Zone System terminology further confuses the subject. Therefore I don't really see any "pull" effect in the "N-1" example, but just a case of underexposure. But it's hard to make critical evaluations about web images. You should start with a standardized target with a full scale of neutral values (a gray scale) plus a representative selection of colors, like a standardized chrome of a MacBeath Color Checker Chart. Then adjust your exposures or effective ASA's to obtain matching middle-gray values in the respective push or pull examples. That way any drift or crossover at the extremes will be readily apparent. You choice of simply bright green and saturated red (plus some orange leaves) makes any objective evaluation quite difficult. Neutrals will tell you a lot more. (Not intended as a criticism per se, but a hint how to proceed with your experiments for more objective results).

I am planning a proper underexposure/overdevelopment and overexposure/underdevelopment test, but I am now mostly shooting black and white negative. Stuff goes in waves so I will probably get back to it. Thank you for the suggestions.

[LabRat]

I used to pull chrome film for commercial architectural shooting (when the lighting contrast was high), and yes it flattened it at the cost of lower saturation and weaker blacks, but it works... But will you like it, and it has a different look from your (N) exposures...

I'd only use it for worst case scenes where I needed an image for a client, but would wait for contrast to even out due to time of day etc and tried to shoot for (N)...

I didn't shoot higher EI for LF, but would live at the lab waiting for the first of 4 sheets exposed alike, to see if a shot was a little dark, that might need a little push when slightly dark, but film was about $80 a box of 50 sheets of 4X5 chromes, and less than a few bucks a sheet for E6, so cheap insurance on a shoot...

I still sometimes shoot 4X5 chromes of art for artists, and have to get everything right with accurate color, and no casts... But rarely shoot LF color for personal use now, but will sometimes shoot color neg when needed...

Looked at some old good chromes of mine recently, and noticed that the film's tend to have a skinny top, but big bottom compared to color neg and B/W I have shot a lot of since...

Steve K

[Drew Wiley]

Change of routine always involves a lot of work. I shot mostly large format chromes for a long time, and color negs only for limited paid projects. I was a good enough Ciba printer to do most of my portrait work in that particular medium, which fetched higher rates, and even printed architectural shots on Ciba. But when the handwriting was on the wall about the demise of Ciba, I got a running start into a permanent switch. And now I'm finally getting to the point at which I feel my RA4 printing going forward will not only equal my past Ciba work, but even excel it. Today I printed a particular Ektar shot on both Fuji CAII (glossy RC) and Fujiflex Supergloss (20-mil polyester). Both are lovely, but the Fujiflex version has all the sheer richness and depth that Ciba offered, and is considerably less idiosyncratic. And fortunately, Kodak color neg films are better than ever.