Some testing results - unexpected outcomes.

[BarryS]

The image order has got to be reversed and your scanning technique is messing up the comparison. Essentially, you're adjusting each scan to each image and that doesn't seem like the comparison you intended. Can you scan all the images at the same scanner settings? You need a densitometer more than a scanner. Looking at the negs--the windows should go from a medium or medium-light density to fairly dark as you increase the development time.

[Ken Lee]

The photos are no longer visible. Did you take them down ?

I can't see any of them, but if they are 4x5 you can scan 4 of them at a time, so they will be scanned at the same setting.

It's best to test under ordinary circumstances where reciprocity is not an issue. Also under balanced lighting (not sunrise or sunset) where some light meters can be fooled by issues of color balance.



Here's a test I did recently with TMY and Diafine. (Diafine is very "compensating" and even under noon day sun the high values are guaranteed to be printable, so these negatives may look a bit soft. Here we are more interested in how faithfully the low values are rendered.)

Exposed under noon sunshine, there are objects whose tonality is easy to recognize and compare after the shoot: some black cloth, white plastic, metal, gray plastic, etc. The white plastic object should not be pure white, because there is a reflection of the Sun on it too. That reflection can be pure white of course.

There is a white card in each photo, which shows the ISO at which the exposure was made. In the "correct" negative, even the white card has a bit of texture.

Another helpful suggestion is to consider that a "standard" gray card of 18% should be placed on Zone 5.5, not Zone 5. Kodak themselves recommends this in so many words. If we consider that the average scene is 7 zones, then Zones 8,7,6 and 5 correspond to 100%, 50%, 25% and 12.5%. Note that Zone 5 is 12.5%, not 18%.

The photos were all made at the same shutter speed - a typical speed like 1/60 - changing only the f/stop. No view camera adjustments were made. The subject was far enough away to eliminate the need for bellows compensation.

For the record, going back to 1970 every test I have ever made has directed me to shoot film at 1/2 the manufacturer's recommended ISO.



Here's a test of TMY and Divided Pyrocat HDC, made on a gray day. Note the natural appearance of the low and middle values: the black cloth, the blue painter's tape and the gray plastic spray bottle. That bottle is an ideal testing prop, as good or better than a face. It's already gray so there's no guesswork involved. The same is true of the "Sharpie" marking pen.

[Ken Lee]

the middle grey was set to 1.0 for each scan...as in ensuring my histogram wasn't clipping and the middle tone was set to 1.

The middle setting is for Gamma, the contrast curve. When it's 1.0, no curve is applied, and the result is "linear". I put that in quotes because with this kind of scanner and software we have only partial control: much is out of our hands.

The Epson scanner has a dynamic range that is equivalent to printing on Grade 1 paper. Therefore we can expect that the histogram of a "normal" negative will not occupy the entire range of the scanner. There should be some wiggle-room at both extremes.

[welly]

Oops. Deleted the files accidentally and now I can't edit my original post. I'll put those images here:

3'30"


6'30"


9'30"


12'30"


I'm beginning to suspect this is more about my (lack of) scanning ability than anything. As you guys have suggested "they're the wrong way round!" but I'm looking at the negatives at the moment and the 3'30" negative is by far the "lightest" negative, as I would expect. Why that would scan in as it has with the 12'30" negative being so dark is what has puzzled me for some time. Unfortunately I don't have a light box to show but the negatives look as I'd expect them (as far as density goes). I think I'm going to need to spend a bit more time with my scanner.

[Ken Lee]

I'm beginning to suspect this is more about my (lack of) scanning ability than anything. As you guys have suggested "they're the wrong way round!" but I'm looking at the negatives at the moment and the 3'30" negative is by far the "lightest" negative, as I would expect. Why that would scan in as it has with the 12'30" negative being so dark is what has puzzled me for some time. Unfortunately I don't have a light box to show but the negatives look as I'd expect them (as far as density goes). I think I'm going to need to spend a bit more time with my scanner.

It's best to scan them all at the same time if possible. Then you are comparing apples to apples, as they say. Each relative to the other.

It's hard to know which image is the literal representation of the scene because the lighting is non-uniform and non-trivial. It's hard to know which of the many shadowy areas were placed on Zone III. It might be easiest to do another test under more routine conditions, and make notes of the actual values as they were metered and where they were placed.

Placing a human face in the scene will also make it easy to determine which version feels the most like light.