Archived posting to the Leica Users Group, 2001/07/31

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: RE: [Leica] Scanning
From: John_Poirier@gov.nt.ca
Date: Tue, 31 Jul 2001 09:56:27 -0600

- ----Original Message-----
From: darkroom /unix [mailto:darkroom@ix.netcom.com]
Sent: Tuesday, July 24, 2001 5:05 PM
To: leica-users /unix
Cc: darkroom /unix
Subject: RE: [Leica] Scanning




> Ok, so if as you suggest 2700 dpi is approaching the 
> do people bother to have drum scans made and why do Nikon bother
> to produce
> a 4000 dpi scanner?
>
> Steve

1) Because the premise that 2700 SPI is approaching the limit of film is
wrong.
2) Drum scanners do not have the bloom and smear problem that CCD 
scanners
have in the red channel.
3) Drum scanners have a higher dynamic range...both less noise, and an
expanded range...but the expanded range is only useful for chrome
- -------------------------------------------------------------
I think I'll throw a little curve at you folks.

Another factor in determining appropriate scanner resolution may be a 
phenomenon called "grain alienation", which in my limited understanding 
is akin to a moire pattern.  Grain alienation is an interaction between 
film grain size and pixel size, and is said to result in exaggerated 
grain under certain conditions.  An introduction to the issue is at this 
site:

http://www.photoscientia.co.uk/Grain.htm

My direct experience with the issue is very limited.  I've compared drum 
scans at about 3700 ppi with desktop scans of the same negs at 2700 ppi 
and observed differences in grain that could be a result of grain 
alienation. 

I expect to do some serious testing over the next year as I am gearing 
up to use digital imaging for permanent preservation of some archival 
images. (I've been  doing scanning for access purposes for many years.)

In terms of most real world uses, I am quite happy scanning even 
Leica-shot (to keep sort of on-topic) transparencies at 2700ppi. It is 
possible that in certain unusually sharp images there is a loss of very 
fine detail, but generally speaking it I get very pleasing results that 
can be enlarged to quite large sizes using technology such as Genuine 
Fractals. 

My experience in scanning thousands of archival images is that the 
limiting factor in image quality is almost always the original image 
rather than scanner resolution.

I make no claim to be an expert on grain alienation, so I encourage 
those who have time to follow up on this one!

Long time lurker,

John Poirier
Coordinator of Technical Services
Northwest Territories Archives
Yellowknife, NWT, Canada