On Dec 15, 2005, at 11:51 AM, Godfrey DiGiorgi wrote:
... I also use 4x oversampling and dICE. The extra scanning
time is worth it.
I did many experiments with oversampling and only rarely saw much
real improvement come out of it. Only with either *extremely* thin
or overly dense/contrasty negatives do I find much use for it.
I've found good results with 4x sampling, but my scanner is reading
much darker overall due to a grain diffuser that I'm using in the
light path. Because of this I need to be very careful to minimise
noise in the shadows as some of those shadows end up as midtones in
the final image.
I experimented with 16x the other day but it's far too slow (would
probably take about 1 hour for a 35mm slide) and the resulting
difference in shadow noise was miniscule. I might play around with
lower settings next time I have the scanner running, to try and
quantify it a bit better. With the amount of scanning I'm doing now,
any improvements to my workflow can result in big time savings.
It pays to optimize exposure and processing for scanning if it's
your primary means to render negatives to print. I normally shoot
negatives for scanning to be *just* dense enough to capture the
shadow details clearly, unblocked highs, and use minimal agitation
to reduce grain growth.
I shoot colour slide film and with its limited latitude there's not
much room for optimising exposure to scan. It doesn't take much to
clip the highlights and no amount of scanner adjustment will bring
the detail back.
- Dave
