<< I have read some postings regarding the feasibility of printing enlarged
negatives on transparency material using the Epson Stylus 800. I am using
Photoshop 4.0 as an image editing program, and am curious as to what output
resolutions and printer resolutions people are using to get the highest
quality enlarged negative results with this printer. >>
Oh, this is a difficult question to answer. It depends on what process you are
using. Your question is analogous to "I want to output an image to be used in
printing my book. What resolution is best?" It depends on what printing method
you are going to use: xerox? quick print? offset? gravure?
But since output from a printer is so easily generated, I won't get too
technical into calculation and stuffs like that because you can easily test it
out. Just generate a scale (digital step tablet). An easy way to do that is to
marquee an area, use gradient fill to fill it from black to white, and then do
"image, adjust, posterize," select 21 steps (you can use any amount, but 21
steps give you 5% each step). Note that the tones of this 21 steps would *not*
appear equally spaced (like a Stouffer step tablet, e.g.) because % coverage
is not the same as density. Percent coverage is linear whereas density is
logarithmic.
Now you can print this "step tablet" out with different resolutions. Then go
through your printing process and print all these tablets. You might be
surprised (you shouldn't) to find that your printing process might not resolve
all the tones with higher resolution. What happened is if your don't expose
enough, the tiny highlight dots get washed out during printing. If you expose
it longer, the shadows gets merged into a blob.
In order to fully use the fine resolution, your printing must be able to hold
even a single dot. That is, the lightest area should be a one-dot tone, and
the darkest area should be all black but one-dot clear (I am talking about the
"negative" now). If you find that your process cannot resolve that but have to
print, say, from rgb=25 to rgb=230, you are relying on the overlapping of
dots. You might think that you can easily use Photoshop to change the output
level to go from 25 to 230, but what you are doing here is trading off
printability with bit depth because now you don't have 256 levels but 230 - 25
= 205 levels, thus you are not gaining anything compared to if you are
printing at lower resolution. Mathematically the two are the same although
because of dithering algorithm you might get a slightly different look.
That is why I said in my earlier post that higher resolution might not give
better result. You will reach a limit. Of course you will get better result if
the output of the printer is the *final* result that you want (otherwise why
would they even try to increase the resolution); but if the printer output is
used as a negative for your alt. process, your resolution will be limited by
the process. Frankly, I don't think a hand-coated material can resolve a very
high resolution. This is not looking down on anybody's ability to coat, but it
is just because we are not machines.
But as one respondent said that he printed with not just black but all 4
colors (or something like that) and then use the result for enlarged negative,
and then he said he wasn't sure whether this would give better result, but it
pleased him. I thought that was cute. It is so true that in arts sometimes you
don't have to sit down with your calculator and do all the calculation before
you make arts. And I can agree with him in some sense. But for those into
technical details, I hope this helps a little.
And sorry to dump this to you with your simple question. I didn't mean to
address this just to you, but I thought the general information might be
useful for others as well.
For those of you who are relatively new in digital imaging but are seriously
thinking about how it works, you can check out my temporary illustration at
http://members.aol.com/fotodave/dignegs.html
If you truly understand the diagram on the left (the "digital" negatives), you
will know (find) that it can actually answer *every* question you might have
with digital negatives. It is simple and nothing new, and many of you already
know it, but every time you have any question about digital imaging, just
think about this "digital" dots type of thing and I am sure you can answer the
question yourself.
Dave
PS: <sigh> I am typing this online again without much organization. Maybe
oneday I should organize this and make it a more organzied article and put it
on my Web site, but for now, just bear with me. :)