Some time ago, one of the Epson tech support
people told me that the “internal” resolution of the current printers
(2200, 4000, 7600, 9600) was 360 ppi. In other words, if it received an
image with something other than this resolution, it would interpolate it to
360, then go about it’s process of creating variable-sized dots, etc.
I have also heard from people who have the
time to do such tests that they can see differences in the output up to 360
ppi. But higher resolution files produce no discernable difference in
output. This would seem to be real-world evidence of what the tech
support guy said.
Does anyone have info which contradicts
this?
AND, if you’re using a RIP instead
of the Epson driver, does that potentially change the “ideal” PPI?
(where “ideal” = “produces the best quality prints possible”)
Michael
From: colorsync-users-bounces+news=email@hidden
[mailto:colorsync-users-bounces+news=email@hidden] On Behalf Of email@hidden
Sent: Monday, August 15, 2005 7:41
AM
To: email@hidden;
email@hidden
Subject: Re: Image PPI and Epson
Printing
In a message dated 8/15/05 9:14:09 AM, email@hidden writes:
An Epson Pro dealer
told me the other day that the resolution of an
image will change the color integrity of the image when it is printed
using an Epson printer. They said that printing an image at 300PPI will
provide a color shift because the image must be converted to a
resolution that the printer can use like 288. The dealer said all
images that are to be printed with an Epson should be at a PPI
divisible into 1440/2880. If the image is at a different PPI (like
300PPI) there will be color shifts because of the conversion to a PPI
that fits into the Epson printing resolutions. Is this true?
The complex algorithms used in inkjet printing are not as simple as that, and
can't be defined by magic resolution numbers. However, the color constancy at
different resolutions can be tested by printing different resolution versions
of the same color target. Certainly Draft mode does not offer the same color as
higher resolution settings, but with a given printer resolution, images with
varying image resolutions in the same color target offer results with minimal
variation in the resulting L*a*b* values of the printed patches... about as
close as different prints and measurements of same resolution targets. Please
note that this only tests the flat color segment of the algorithms (the only
area we have tools to easily test), not what it produces in high contrast of
other variable areas.
So with all the other color issues you could be worrying about, I wouldn't give
this one much consideration.
C. David Tobie
Product Technology Manager
ColorVision, Inc.
email@hidden
www.colorvision.com