You said in your message...
>I always imagined that if you have 300 lines you also have 300 dots / inch in
>that direction.
>
>Could you help me to understand it better?
Oh, boy! There is an appendix chapter in my book called "Pixels, Samples,
Dots, Spots, Lines and Specks Explained." It's there because of good
questions like yours. It's a confusing issue with all these dpi, ppi,
lpi, spi terms falling upon us!
Briefly (very) the "lines," as in lines per inch (lpi), are the tiny ink
"spots" you see in an image printed by a printing press. Yes, the term
"lines" IS misleading (actually comes from the way they used to break up
a continuous tone image with a glass plate that was etched with so many
"lines" per inch), but basically it refers to how many ink spots there
are in a linear inch. Newspapers use about 80 or so lpi, a fine-art book
might go to 300 or even higher. And most magazines are in the 135-175 lpi
range.
Now each of these ink spots is formed by the imagesetter's laser. But
it's not on a 1:1 relationship. For most printing press jobs, the shades
of gray are made by varying the SIZE of the ink spots on the paper. These
spots are arranged in rows like cars in a parking lot. The imagesetter
uses its much-higher resolution (going above 10,000 dpi in rare cases) to
make different size spots (remember, the "lines" are really those spots
of ink). The higher the imagesetter's resolution, the more different size
spots it can create and the more different shades of gray are produced.
Though I won't go into the math right now (gosh, it's early and I haven't
even finished ONE cup of coffee yet), most of the time you strive for an
imagesetter resolution that is 16 times your desired lpi output. So if
you are printing a magazine at 150 lpi, you'd want an imagesetter that
has a resolution of 2400 dpi (16 X 150
= 2400). For the kind of negatives we make for contact printing on
alternate or silver processes, lpi of 300 or higher are desirable (to
make the spots invisible to the unaided eye) so your imagesetter
resolution should be much higher, in the range of 4800 dpi for best
results. (There are all kinds of exceptions and stuff that I won't
attempt to cover in this response.)
It gets better. The way some of us make imagesetter negatives (myself
included), instead of creating different SIZE spots, we vary the DOT
POPULATION instead. These "specks" are much smaller than the normal
"spots" used in most commercial offset printing. This is the heart of
Stochastic printing: the specks are arranged RANDOMLY and are much
smaller. For stochastic printing, the ratio of imagesetter "dots" to
negative "specks" is much lower, often only 1:2. I routinely make 1200
ppi negatives for which the imagesetter runs at 2400 dpi.
There is much discussion (arguing) over which method of making
imagesetter negatives is better, line screen or stochastic. I've seen
wonderful prints from both so it's really more like choosing a religion
than planning a rocket launch.
I'm heading back to the Mr. Coffee with fears that I've perhaps confused
you more. Sorry if that's the case!
Dan
This archive was generated by hypermail 2b29 : Tue Jan 11 2000 - 12:10:49