Re: [Gimp-print-devel] 1440x2880 mode?

[Ralph C. <ra...@in...>]

Hi,

> I'm experimenting with a 1440x2880 mode (that's correct -- not
> 2880x1440).  The reasoning behind this is that dither artifacts in
> 6-color printers often look like "holes" behind dark dots, and that
> oversampling in the Y direction covers them up.  It does seem to look
> very smooth, perhaps smoother than 1440x1440, although currently I
> don't have the matrix working correctly with it.
> 
> Anyone have any idea what to name it, if I do decide to put it in?

I've been confused by the naming so far so here's my 2p's worth.

Distinguish between `physical' and `pseudo' resolutions.  Am I correct
in thinking that most exactly match a printer resolution and it is used
for all calculations?  No super-sampling?  I'd call that a physical
one.  Ones that are over-sampled before being converted back to a
physical one could be `pseudo' resolutions.  The printer isn't really
printing at 1440x2880.  Each pseudo resolution should have an
accompanying physical one.

Hmmm.  I guess I'm really saying there are two resolutions involved all
the time, logical (better than pseudo) and physical.  Sometimes they're
the same, 1440x720-1440x720, and sometimes not, 1440x2880-???x???.

Add onto that a method of error diffusion/dithering naming and have we
a canonical format where all parameters are given?

    logXxlogY-phyXxphyY-diffused

Sorry if this is wide of the mark because I don't understand the whole
process.


Ralph.