"nobody nowhere" wrote:

The amount of sharpening I apply has been
monotonic decreasing over the two years I've been scanning MF. It's
simply not a panacea.

Would you please elaborate on "monotonic", what it is/does (in simple
words please).

Sorry, "monotonic decreasing" just means always getting less without ever
increasing. Just me being pretentious and too lazy to come up with a better
turn of phrase.

Some people advocate no sharpening at all, but this
seems to me impossible, one surely needs a certain amount of sharpening
to correct the softness that comes from scanning. Would you agree?

Yes. Some sharpening is required. It's just that the amount of sharpening I
use has decreased over time.

Sorry to repeat myself, but when people look at the images, they see near
645 quality, not 35mm quality, images.

When I scan my 27 years old 35mm slides with a Nikon 8000, I get
digital files which often yield acceptable A3, prints. This size would
have been out of the question "in the old days", with a traditional
enlarger/chemical system.

No! Don't say that! The digital naysayers depend desperately on the Nikon
8000 being _worse_ than wet chemical projection printing for their claims to
hold up, since the people doing 1Ds vs. 35mm comparisons (and finding the
1Ds better) are scanning their slides with 4000 dpi film scannersg.

Seriously, though, 4000 dpi scans of ISO 100 or slower slide films will make
fairly nice 12x18 prints from 35mm, as long as you keep them at arms length.
Not particularly sharp, but grain's not a major problem. But I'd be
surprised if it really were impossible to do at least nearly as well with
projection printing.

What I am trying to say, unless I am wrong,
is that the 645 image quality might be inherent in the digital process,
rather than a special quality of the Canon 1Ds and its much praised
sensor. If one accepts this, for the moment at least, the combination
film + scanner would make more sense than Canon 1Ds (for some, or most
amateurs).

I don't agree with the theory you're presenting (but the conclusion's right,
albeit for other reasons). At least for the sorts of prints I like, any
frame size/pixel count has its limitations. The question here is what
metric/theory predicts those limitations best. The "film "resolves" 100
lp/mm" theory seems disproved by the existence of a better technology that
clearly relies on a lot less than 100 lp/mmg.

So maybe true "1Ds quality" images requires Bobm's and my mythical 16MP
full frame sensor.

When this mythical thing is on the shelves, wake me up! but not before!
:-).

Yup. That's my line. But I won't be sleeping. Until then it's 645 and a big
tripod for me.

David J. Littleboy
Tokyo, Japan