Image sizes and resolutions
Image sizes and resolutions
- Subject: Image sizes and resolutions
- From: Gary Smith <email@hidden>
- Date: Mon, 22 Apr 2002 22:18:00 -0600
Hello all. I've been dragged into a bit of a mess here.
I am a former owner of a service bureau. I have owned 2-color
presses, color laser copiers, large format ink jets, desktop ink
jets, film recorders, scanners and various RIPs. I am very versed
with scanning and preparing art for print. I successfully use color
management and can't imagine life without it. I now have a Linocolor
Opal Ultra and a Nikon Supercoolscan 8000 ED.
My 15+ years of experience with these various scanners and output
devices has led me to understandings about image size and resolution:
1. For example, One (8" x 10") photo scanned 3 times in greyscale.
Once at 100% (8" x 10") at 100 ppi, again at 50% (4" x 5") at 200
ppi, and again at 25% (2" x 2.25") at 400 ppi. All three are 782 K
and 800 x 100 pixels. All three will yield identical results when
output at the same size, i.e. all three at (8" x 10") or all three at
(4" x 5") or all three at (2" x 2.25"). It is my understanding that
the purpose of discarding excess resolution is to prevent slow RIP
times, make smaller files etc., not to prevent "plugging up" or other
problems with image quality. Too much total ink and dot gain is the
cause of that. The RIPs discard the unnecessary pixel data.
The local newspaper has the usual guidelines for file prep. 85 lpi
screen, images should be between 1.5 x and 2.5 x for resolution at
print size.
If I were to, in a page layout application place either the (2" x
2.25") 400 ppi image or the (8" x 10") image and transform either to
(4" x 5"), the pre-press department at the newspaper claims the small
image at 400 ppi will cause "plugging up" while the large 100 ppi
image has insufficient image data. Note I am talking greyscale here.
They find the (4" x 5") image acceptable, yet all three have
identical pixel data.
Ultimately, I like prepare PDF files with Distiller Job Options set
to downsample to 220 ppi so any excess resolution it gets discarded
at that point.
This lack of understanding at this pre-press department of the
fundamentals of scan resolution, scaling and choice of resolution at
scan time, and the effects of size transformations is a huge red flag
to me. I believe them to not have a clue.
I am currently preparing art for ads on behalf of a client destined
for this newspaper and encountering problems with color images being
too dark. I use Illustrator 9/10 and Photoshop 6/7. Getting proper
Photoshop CMYK separation setting from these guys has been futile. I
have tried using US Web Uncoated and a couple of Newspaper profiles
that came with Linocolor. I finally got some custom Photoshop CMYK
setting info from one of the guys that has no clue about image size.
It is said to me it did not work. I have not seen the results as the
images were modified at the newspaper.
I suspect they discard embedded profiles when encountered.
And this is the problem: My very computer illiterate client is
becoming frustrated and much of it is directed at me. He is turning
to the pre-press guys for answers and buying their line of bull
(whatever it is). I insist that their advice so far has been akin to
a mechanic suggesting the use of oil for a two stroke engine in a
car. BAD NEWS!
I have been following this forum (quite passively) for a few years
now and respect the knowledge of it's frequent visitors. I search the
archives for solutions very frequently. I would appreciate any help
anyone has to offer regarding this situation or any support of my
position that yes, there is the unfortunate yet common lack of
understanding of basic pre-press principles by some pre-press
personnel.
Finally, where can I get a good newspaper ICC profile or Photoshop
CMYK separation settings?
Thank you,
Gary Smitht
--
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.