Re: dpi and line screen -- Conventional Wisdom vs Reality
Re: dpi and line screen -- Conventional Wisdom vs Reality
- Subject: Re: dpi and line screen -- Conventional Wisdom vs Reality
- From: "Richard Frederickson [Contr]" <email@hidden>
- Date: Fri, 30 Dec 2005 11:10:29 -0500
Title: Re: dpi and line screen -- Conventional Wisdom vs
Reality
Thanks for the feedback, Jim.
We use an Imacon 949 scanner to directly
capture images from histologic specimen slides, and it does a
wonderful job. The math just works out nicely that at our maximum
capture resolution, we get 40 times life size (a nice scientific
sounding round number)--only its 200 dpi at final size (about 1.5
times for a 133 line screen or 1.3 times for a 150 line screen).
200 reproduces nicely to our photo-quality printers, but
everyone has been indoctrinated into the 300 dpi mantra for offset
print.
In an article by Richard Anderson
(http://www.peterkrogh.com/Pages/digital/buyingdigital.html), he makes
an argument that digital capture contains more detail than scanning
film (since you're not dealing with film grain) and, therefore, the
rule of thumb no longer applies.
The article seemed more opinion than
fact-based on hard evidence, so I was hoping for something a bit more
definitive on the subject--I will follow-up with your
leads.
The follow-on question is how
AM/FM/hybrid screen modulation affects the equation?
Again, my thanks for your assistance.
Richard
There have
been a few articles on this and all done agree.
Here is an
almost exact quote that is taken from the original Photoshop in Black
and White Book that was published in 1992.
"The rule of thumb for input resolution is based on the NyQuis
Therom. That is, when you are going to print, use twice as many input
pixels (2:1) as the halftone screen ruling. While this 2:1 rule of
thumb is a good starting point to determine image resolution, some
imaging professionals have experimented with reducing input resolution
and have found that it is possible to achieve good results with as low
as 1.5:1 input image resolution for halftone printing. To determine
input resolution, the best method is to run several tests at different
resolutions and choose the best results. "
Also in 1992 we ran some tests for halftone printing using ratios from
1:1 to 2:1. We found that in some cases when using less than 1.5 :1
that the images were pretty safe. However, we also determined
that if you use a lower resolution image (sch as with the 1.5;1 ratio)
you can compromise the image quality if any of those lower resolution
images are scaled up say in Quark. And if that happen you then
might have to rescan the original at a higher resolution. If that
happens every now an then its not a problem, but if it happens say to
hundreds of images your hosed.
As for images below the ratio of 1.4:1, we found that you can
get away with the lower resolutions but it became image dependent.
That was then.
Today, for offset printing my sense is that those guidelines still
apply. But for other processes like inkjet printing you can get away
with less image resolution. You just have to run a few tests to
determine the optimum input resolution.
When it comes to resolution, mileage will vary.
Jim Rich
On 12/29/05 3:21 PM, "Richard Frederickson [Contr]"
<email@hidden> wrote:
Please pardon my
being a bit off topic with this post, but I thought with all the
experts out there who monitor this forum, I'd get the most reliable
response to the following:
Conventional wisdom says to provide twice the resolution of your line
screen for offset printing. I seem to remember reading an article
debunking that myth and promoting a resolution equal to line screen
scenario--but I can't put my hands on it.
Can anyone point me to an authoritative source I can quote?
Many thanks!
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list
(email@hidden)
Help/Unsubscribe/Update your Subscription:
.com
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden