Re: CoolPix 950 question [off topic]

["Keith Davison" <keith.davison@xxxxxxxxx>]

> >A generally accepted resolution for "photo quality" (i.e. as good as
> >an original high quality photo) is 300 dpi. The 950 is 1200x1800
> >resolution, so it would yeild a 4"x6 print at 300 dpi. At 150 dpi,
> >yeilding a lower-quality image with some visible digital effects
> 
> Here's my take on image resolution for printing...
> 
> The 300 dpi rule is from traditional halftone printing (press printing and older desktop computer printers). Here the rule of thumb is image resolution (dpi) should be twice the line frequency (lpi) of the halftone. The lpi is device dependant and related to the dpi of the printer. Any more 
resolution than this is wasted data (and space from big images).
> 
> Most color inkjet printers today use stochastic screening or similar technology that needs much less image resolution to make good looking images. Some even interpolate image resolution on the fly if needed.
> 
> Anyone have a rule of thumb for needed image resolution for the new generation of (affordable) inkjets? Particularly the HP 970C I've got my eye on.
> 
> ~Todd
> 

I don't have a rule of thumb, but I have done extensive testing with an 
Epson Photo Stylus 700 and EX and images from scanned negatives. 
For these printers the cut off point for colour prints is 340ppi, there is no 
improvement in printed quality beyond this figure. Below 340ppi quality 
dergrades, gradually exhibiting the usual pixelation effects.
Be careful when comparing pixel counts from consumer digital cameras 
to those from other sources such as scanned film. Digital cameras such 
as the 950 use a single CCD with a tricolour mosaic filter in front of it, 
they interpolate colour information from several neighbouring pixels to 
generate the value for each individual pixel. This filter will still be present 
in monochrome mode and will have to be compensated for in some way, 
I imagine this compensation will also involve interpolation.  So although 
you have 1200x1800 images, they will probably have had some degree 
of interpolation applied even in monochrome mode, and less image 
detail than the pixel resolution would suggest.

Keith Davison

Neurobiology Dept.
Medical School
University of Newcastle upon Tyne
England
*
****
*******
******************************************************
*  To remove yourself from this list, send:          *
*         UNSUBSCRIBE INFRARED                       *
*       to                                           *
*         MAJORDOMO@xxxxx                            *
*----------------------------------------------------*
*   For the IR-FAQ, IR-Gallery and heaps of links:   *
*  http://www.a1.nl/phomepag/markerink/mainpage.htm  *
******************************************************