RE: [Digital BW] Print Quality -- MF film (was From A Nikon D1)
2002-05-30 by Paul Roark
Austin, >...waiting for 16M pixels ... Concur -- and monochrome at that. >... IMO, if you want to improve your photography (depending on what >your style is that is), moving to a decent MF can make a >HUGE difference. Also, a big concur. In fact, with digital I'm finding the extend to which I can manipulate the image without obvious artifacts is making my MF equipment marginal -- the film grain seems to be the limiting factor. > I pretty much end up using 35mm for snapshots ... Another concur. In fact, I've moved to the MF rangefinders -- Fuji 645 Zi & Bronica RF 645 -- and find them to be good compromises that can also do snapshots. For family affairs/travel, the Fuji Zi with T400CN film in it is an amazingly flexible, easy way to go. But, to get to my question -- I suspect that the old "rules of thumb" that I/we used in the darkroom with respect to B&W film may no longer apply to film that is shot for scanning. With modern scanners that are designed -- perhaps primarily -- to be able to capture the wide density range of slide film, I would think a low contrast B&W negative might be wasting the scanner's capabilities. So, what I'm experimenting with is shooting Tech Pan and developing it at a higher contrast than I would if I were going to print the negatives. Since grain is my limiting factor with TMax 100, TP helps solve that problem nicely. I don't really have objective tests that tell me whether I'm closer to finding the scanner's (in my case the Nikon 8000) optimum range, but the bottom line is that the scanned images are clearly better. It could, of course, be mostly the film that is resulting in the better images. However, with traditional enlarging, TMax 100 developed to maximize sharpness (not minimize grain) won this contest and was my standard for years. My current formula for TP is to shoot at about 40 ISO and develop in Xtol 1:3 (all distilled water for this developer). At 75 degrees f., 18 minutes with agitation at 1 minute intervals seems to do the trick. TP in Xtol is much less prone to uneven development, but I still am very careful with agitation. I use the Kodak vertical shake method (I use 20 shakes of the small metal tank, as fast and hard as I can), but I alternate that method with a normal inversion agitation. I use a "plain sky" (ground glass over the lens) shot to test the agitation methods. The results are a characteristic curve with at least 12 stops of useable image range (all that I've tested). The slope at the mid-gray (per the meter reading) is 1.25. The mid-range (mid-gray plus and minus 3 stops for a total of 6 stops) has a contrast index (average slope) of about 0.88. At close to 9 stops over the meter-read gray point the slope is still close to 30%. This point is about 7.5 stops above the film base + fog point. (I read the densities with a meter, so they are in stops, not log density values.) The Nikon 8000 is still able to separate the values at this density and with the 30% slope -- but barely. I can't see that noise is much of a factor anywhere. However, I haven't had enough experience with real world images using this combination to know for sure. I have developed the TP to higher contrast, but the image quality seemed to deteriorate. The resolution seemed to decrease and grain increase. Again, this may have been mostly the film, not the scanner, characteristics that I was seeing. I'd be curious if there are any generalities about how to squeeze the best performance out of these CCD scanners, or whether they are too different to generalize. In most of the technical systems I'm familiar with the end points are usually marginal, but using too little of the mid-section range and having to increase contrast that much more is also probably bad. So, is there a general "sweet spot" compromise with these scanners that you've been able to ascertain? What B&W film/contrast have you found optimizes the quality of your scans? Paul http://www.PaulRoark.com