Julie,

I would avoid using "scaling" since this often means the scanner will  
be interpolating (aka making guesses) data from what it can't see on  
the original.   This would also be the case in taking lower  
resolution images and increasing their DPI after scanning.  It can be  
done, but it no longer accurately represents the original.

Regarding resolutions, this is one of those cases where it may be  
better to look at the pixels on the longest dimension, rather than  
DPI.  If your materials are all the same size you can calculate a  
target dpi for them.   Taking CDL's best practice of 4,000 pixels on  
the longest dimension would give you ~1150 dpi for 3.5 photographic  
negatives.

Some helpful tools are the Excel sheets from the Technical Advisory  
Service for Images, which let you estimate storage requirements based  
on resolution, bit depth and size.  http://tasi.ac.uk/resources/ 
toolbox.html

Setting your scanner to the "highest resolution possible" might not  
buy you anything, as film or prints have their own resolution.  At  
some point you may be capturing more information than the original  
film holds, which may not be efficient use of storage space (and lets  
not forget that good digital preservation means you have multiple  
copies to prevent a single point of failure and should be included in  
your storage estimates).

Richard Urban
rjurban at uiuc.edu

On Nov 21, 2006, at 3:32 PM, Julie Grob wrote:

> Hello,
>
> We are about to begin scanning a large group of early 20th century
> negatives. They are about 3.5" square. We will be creating master  
> TIFFs of
> course, but we would like to be able to print larger than 3.5"  
> images. Is
> it better to scale up and scan them at something like 200%, or to  
> increase
> the resolution from 600 dpi to a higher dpi?
>
> Thanks in advance,
> Julie Grob
>
>
> Julie Grob
> Digital Projects and Instruction Librarian
> Special Collections
> 114 University Libraries
> University of Houston
> Houston, TX 77204-2000
> (713) 743-9744
> jgrob at uh.edu
>
> _______________________________________________
> You are currently subscribed to mcn-l, the listserv of the Museum  
> Computer Network (http://www.mcn.edu)
>
> To post to this list, send messages to: mcn-l at mcn.edu
>
> To unsubscribe or change mcn-l delivery options visit:
> http://toronto.mediatrope.com/mailman/listinfo/mcn-l


Reply via email to