edge to edge sharpness from flatbed

My tests are showing me that film flatness seems to be more critical than I'd

previously thought. I'm wanting to compare my findings with others' to see if

I'm on the right track. I scanned the same 120 tranny in my Canon 9950F, a

friend's Epson V700, and a Nikonscan 8000 (with the regular non-glass film

holder). Keeping all things somewhat equal, the Nikon was very sharp in the

center of the scan but soft on the edges, The V700 was almost as sharp in the

middle and equally sharp at the edge, and the Canon was not quite as sharp in

any location, but fairly even across. Applying stages of USM to each image

brought each of them to a point where an 11x14 image looked almost equally

sharp, but at 16x20 the Nikon started to edge ahead, but ONLY in the center

sharp region. The V700 was still noticably sharper at the edge.

My goal in this particular project is sharp 16x20 prints.

So here's my resulting hypothesis, which I'd like evaluated:

1) I'm not going to make it with this Canon scanner.

2) The Nikon, as everyone says, really does need the glass carrier to create a

16x20 print. And they would be great.

3) An Epson V700 with Doug's film holder and the ANR glass insert would be

better than the Nikon scanner with its regular holder.

4) The V700 MIGHT be as good as the Nikon at 16x20 if you used the Doug/ANR

holder AND were working from a negative that you could get the right dmax out

of.

Your thoughts?

4) A Nikon LS-8000 is sharper with a glass carrier than with the standard carrier

5) Depending on the quality of the source negative, a Nikon LS-8000 is nearly twice as sharp as a V750 at it's best. Not all negatives are equal (including lenses and technique).

6) A 16x20 enlargement is not big enough to make a critical comparison, being only an 8x enlargement (comparable to 35mm prints at 8x10 inches). The corollary is that for 16x20 inch prints, a V750 might be adequate.

My experience is with a HP 4890 flatbed (have no others for comparison).

On another thread, thanks to advice from others, I discovered that the clarity of my 35mm colour negative scans was critically linked to how flat the negative was held. [i had originally thought that having the emulsion side up was important, I have since found that it had more to do with the curve of the (relatively flat) negative and how its orientation altered its distance from the optimal focal plane.]

The only larger negatives that I am scanning are all old (40+years) 616 and 620 b/w negatives. For whatever reason these seem quite insensitive to how flat they are held, and even to their distance from optimal focus. I am not sure if this insensitivity is related to the b/w or the age of the negatives (or both).

Whatever... given my experience with the 35mm negatives I intend to create custom mounts that hold the negatives as flat as I can and at a distance that seems optimal for 35mm (in my case closer to the glass than the HP provided holders).

Doug's anti-newton carrier and his adjustable-focus carrier are nearly equal, but the glass does increase dust...I use the adjustable, non-glass version, which does offer slightly better sharpness. I have both types, as well as the most basic non-adjustable, non-glass version.

All of Doug's carriers hold 120 flatter than would a conventional enlarger film carrier (such as Omega or Beseler), far flatter than Epson standard carriers.

I rented a Nikon 8000 for a few hours a couple months ago just to see how much I'm missing out in only using my Epson 4990 (which I'm VERY pleased with). Bottom line is unless you're printing consistently over 16x16 or so (I shoot mostly square MF lately) the extra thousands of the Nikon's are simply not worth it. With proper equipment, attention to important things like flatness and good scanning workflow and proper post-scanning technique in Photoshop (standard levels, curves, use of unsharp mask pre-printing, etc.) one can get incredible prints from a scan off an Epson flatbeds.

Do the Nikon's get a bit more sharpness, dmax, and perhaps tonality? In theory, yes. In practical terms of being able to tell one from the other on a print, framed on a wall at typical gallery viewing distance? Nope.

That said if money was no issue or someone offered me a Nikon 9000 with glass ANR holders for cheap I'd grab it. But for the $400 for my 4990 vs. $2500 for the Nikon with the holders, no thank you, not needed.

When I need to print larger I'll get a drum scan.

Total film flatness is a given when fluid scanning, so are adjustable

film planes to match the scanner's.

Julio

www.scanscience.com