Re: Scan Backs
Re: Scan Backs
- Subject: Re: Scan Backs
- From: Dick Busher <email@hidden>
- Date: Fri, 19 Jul 2002 15:26:58 -0700
- Organization: Cosgrove Editions
>
on 7/18/02 6:29pm Terry Wyse wrote
>
>
I find this topic fascinating and of great interest. A couple of
>
points/questions:
>
>
* Isn't it true that transparency films only have a real dynamic range of
>
about 5-7 stops compared to high-end digicams with 10+ stops? The scan of
>
film might look great compared to the "original" but that film has already
>
compressed the original scene by several stops, no?
7+1/3 stops if you do not count black as the starting point. However that should not be a
limitation. In the studio environment we control the lighting, and we should be able to
light the subjects so that all of the detail we are interested in capturing falls within
the dynamic range of the film.
>
* What's the practical "resolution" of transparency film, say ISO 100-200? I
>
thought I heard many years ago that any scanning resolution greater than
>
about 4,000dpi was beyond the resolution of the grain itself. True or are
>
toady's films much better than that? So a file size of about 1-1.5GB would
>
appear to be the practical limit of capturing real detail.
The film I use has an ISO of 100. We scan often scan 35mm transparencies for others with
ISO 50. I have heard others say the same thing you have said above. However, a colleague
scans 35mm chromes with resolutions as high as 10,000 dpi for the Ultrastable printing
process. Does such a scan actually contain more information? I'm not sure (another
experiment to add to my schedule). However his output device does need the higher
resolution. He claims that a 4000 dpi scan does not result in a print as good as one made
from an 8000 dpi scan. I can only take him at his word.
>
> In the final analysis the proof is in the pudding.
>
>
Good point. Back in my RIP training days, I used to do tests for folks and
>
show them the relative relavence (now there's a combo!) of text resolution.
>
We used to output film/proofs and sometimes press sheets where we'd RIP the
>
job from, say, 3600dpi down to about 900 dpi. Put a 12x lupe on the film and
>
above about 1800dpi was almost indistinguishable. Put it on press and even
>
900dpi looked as good as the higher rez type. My point is, I wonder how much
>
of that extra detail and resolution with images really makes it to the press
>
sheet or even to the proof. Could I see the diff between a high rez film
>
scan vs. a good digital capture? Maybe, maybe not.
The higher image setter resolution may not be as important with vector graphics/type.
However, in my experience it makes a hugh difference with images, especially with images
containing fine details. I started printing with 175 lpi conventional screens. I made a
progression to 200>225>240>20 micron stochastic>10 micron. With each upgrade I could see
differences. With each upgrade I had to increase file sizes accordingly in order to take
advantage of the higher screening frequency. With 10 microns I need to have a 600 dpi
file at the output size. (The 2:1 rule for conventional screens does not seem to apply to
stochastic.) That translates to 5,000 dpi scans from 35mm originals for 8.5 x 11 bleed
images.
Dick Busher
Cosgrove Editions
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.