Tale of Two Films

• Films: Print film: New Reala (ISO 100). Slide film: Fuji Velvia (ISO 50).

• Shooting: Both types of films are shot with the same lens, with the same exposure (one stop increase in shutter speed for Velvia). Print films are shot with Nikon N6006; slides with Nikon FM. Metering done by N6006.

• Processing: Prints were developed and printed by Qualex, a Kodak subsidiary, and the quality of this particular batch was better than average. Slides were developed by Fuji TruColor Lab in Phoenix, Arizona.

• Scanning: Prints are scanned from 4"x6" prints using a flatbed scanner. Slides are scanned by Polaroid SprintScan 35. In both processes, I used the scanners' default (automatic) setting, and the scanned images are about 1000x1500.

• Image Manipulation: I tried my best to keep it to a minimal level. Both scanned images are "sharpened" by "xv" program. Besides sharpening and size adjustment, almost no image manipulations were made.

The left column are from prints, and the right column are from slides. Use your own judgment to draw your conclusion(s).

• Although it is true that a print film can record wider a tonal range than a slide, when using print films, one needs to take into account the tonal range of print papers, which is the actual limiting factor, unless the negatives are processed electronically. I suspect that a print paper has a smaller tonal range than a slide. Look at the left edge of the MIT dome in the first set of pictures. In the print, the dome blends with sky so there is no distinguishable border. In the slide, the boundary is still quite clear.

• Unless custom lab is used, one has no much control over print quality. The last two examples are the typical "washed-out" color prints. We have seen far too often pictures like these. It is disappointing and frustrating. Advise given by those "experienced" is often "you had an overexposure". Actually, no, the exposures on the film are perfect, as shown by the slides. The lab messed up.

• From my experience, when a picture is dominantly dark, the result of a machine print is usually terrible: the machine tends to over expose so the picture is dominantly grainy gray. When you got such a bad machine print, you would not be confident that a properly exposed print could be any good, and thus hesitate to shell out more money to have a custom print made.

• Slides are good for mastering the exposure. From negatives, due to its color cast, it is not easy to judge by naked eyes whether a negative is properly exposed, unless the exposure is way off the scale. The wider tonal range adds another level of difficulty since, from the negative, you can still see most of the details. Slides are totally different. Any subtle difference is readily observable.

• Now I can talk about colors! When I bought the Nikkor 80-200mm/f2.8, I didn't realize that the lens uses a 81A warming filter as the protector, until I saw the slides. From slides, now I see the subtle color rendition of different lenses. From prints, all I can say is "oh, this batch has a purple tint", or, "this batch is too yellowish". I never got prints from the same negative by difference orders that look even similar in terms of colors. In my opinion, if this type of prints is all that you've looked at, you are not qualified to talk about most of lens optics.

• Slides have different visual effects as compared with prints. For a slide, lights are filtered through it as if they were emitting from the scene. A bright spot in a slide really glows, and colors are purer and more saturated. For a print, lights are reflected from the paper (or plastic?) surface, and often, the colors appear muddy. Compare the colors of the sky and the water in the first two sets of pictures!

• I found that scanning from color prints is not such a good idea. Quality lost occurs in almost every stage: lost of tonal range in printing, lost of details during printing (due to inaccurate focus), lost of color accuracy due to machine correction, lost of tonal range due to incorrect printing exposure, and further lost of tonal range in scanning since scanning is performed by reflecting light off the print. None of these problems occur in scanning of slides.

Final Remark:

Several people pointed out to me that it is not fair to compare the scanned images from prints and from slides, and what I should compare is the scanned images from negatives and slides.

My personal feeling is that the fairness depends on the ultimate form of photographs. Traditionally, prints or slides are the only forms, and negatives is an intermediate form. If the goal is to generate electronic images, the above argument is quite true. I might revisit this topic later. But, even in that scenario, slides still definitely have numerous advantages.

I would agree that it would make much more sense to compare the prints made from negatives and from slides. On that regard, I have conducted two series of comparisons. The goal is to make high-quality prints from slides economically. Check out my following reports:

• Slide Film vs. Print Film: Enlargements in which I compare various ways of making enlargements from slides, and compared with the enlargements made from negative film.
• Home Internegging in which I explore a simple way of duplicating slides into negatives using generic print films.