Digital BW, The Print

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Loring Palleske 
<lorpal@m...> wrote:
> NO. Dithering is not done on a pixel by pixel basis.
> 
> On Sunday, July 27, 2003, at 02:33  PM, Peter Nelson wrote:
> 
> > Think of it this way - at maximum printer resolution it can only
> > represent 46 distinct tonal values in the surface area of the 
paper
> > where the pixel might have any value from 0-255.
> Regards,

Actually, we don't know WHAT the dithering algorithm they use is.  
BUT there's a simple test you can do, which I've done on all three 
of my inkjet printers (HP 970CSE, Epson 870, and Epson 2200):  
Create a test pattern of individual pixels and one-pixel-wide-lines 
separated by one-pixel white space.   Create the image and print it 
without doing any rescaling, so the pixels and 1-pixel wide lines 
are sent to the printer AS 1 pixel.   Just for fun, do this at 
different resolutions - I've tried it at 240, 300, 360, 400, and 600 
PPI.

You will find that they are individually rendered and distinct under 
magnification at resolutions up to 360 or so DPI, depending on the 
printer.

In reality we have some good evidence that the 2200 resamples before 
dithering.  But if they did a stochastic dither over an area greater 
than 2 pixels-width than adjaecent pixels would blur together.