Digital BW, The Print

I'm certainly no expert on the finer scientific points of
interpolation. In fact it was all against my religion until this past
year when I started getting more and more more digital camera files
that people wanted fairly large output from. I lived by Frasiers
statement that the one thing to know about interpolation is - don't do
it. My feeling about it now is that I might as well get used to it
because its going to be with us for awhile until these camera chips
get better. Sure people have been doing this 10% increment thing in ps
for years, and many still do with with good results (better than GF by
the way for all the hype that surrounded it). I have also heard that
the CS BC Smoother function is also quite good and I don't know what
the math is behind it. 

It certainly would be nice to take the same file and res it up
different ways to do a comparison test. I'll be glad to do the Miranda
test of it and or PSCS2. You know all this is to a degree image
dependent also. With the digital camera files and any ccd capture for
that matter noise is always a nagging problem in the extremes, whether
it is upsampled or not. I assume that is why Miranda has added this
new smart blur filter between each step of the action. I have to read
more about it to discuss it intelligently but it does work.

You guys have made me want to look into all this furthur. I gues with
all these Cannon cameras out there I'll have no choice. Now ... back
to drum scanning, my real optical files...

John





--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Paul D. DeRocco"
<pderocco@i...> wrote:
> > From: Jeff Medkeff
> >
> > Stair interpolation refers to iteratively computing a spline over and
> > over again, usually in fixed proportions, until you get the output
size
> > you want. For example, if you used bicubic interpolation and increased
> > the file size by 10% each time, until you got where you wanted to be,
> > that would be stair interpolation. If you used S spline the same way,
> > that would also be stair interpolation.
> >
> > Several web sites report that FM SI Pro is bicubic stair interpolation
> > at 12% per iteration, and several note that the results with "bicubic
> > smoother" in CS2 is just as good. This sort of exchange is typical:
> 
> It seems to me that any linear filter applied multiple times is just
another
> different linear filter, and can be precomputed as such in advance, and
> applied in the same amount of time it takes to do any other single
filter.
> Doing it iteratively seems like the dumb way to do it (if you're
writing the
> software, that is), and might also build up noise due to accumulated
> roundoff errors, especially if computation is done in 8-bit mode.
> 
> --
> 
> Ciao,               Paul D. DeRocco
> Paul                mailto:pderocco@i...