Re: determining optimum dot gain
Re: determining optimum dot gain
- Subject: Re: determining optimum dot gain
- From: Roberto Michelena <email@hidden>
- Date: Wed, 24 Nov 2004 08:07:46 -0500
Mark,
There are four steps to this task:
1) linearize your plate
2) print and measure a strip to determine the press "natural" (measured) dot
gain
3) check this "natural" (measured) dot gain curve to see if its shape and
value falls within specs (or common experience) for your press, inks and
media. Because if not, before making an adjustment curve you should try to
correct the mechanical or material defects that cause a wild behaviour.
4) select a "target" curve you want to match. I would say ISO 12647 curves
(adjusted for negative plates if you're in the US), but you can also choose
GraCol or SWOP; I believe however that these standards don't provide
complete curves just selected dot gain values.
5) build a "compensation" curve to match your "natural" one to the selected
standard. Depending on your software (if your QuickMaster is driven by a
Harlequin rip, for example) you can just enter the "natural" curve and the
"target" curve in two parts of the Rip interfase, and it will calculate the
compensation automatically; furthermore, the Harlequin rip comes with
various target curves preloaded. In other Rip systems, you have to calculate
the curve yourself and enter it.
To calculate the compensation curve, substract the "natural" curve values
from the "target" curve ones, and then add this result to a linear curve.
For example: (just five values to illustrate)
natural curve: 0=0 , 25=37, 50=74, 75=88, 100=100
target curve: 0=0, 25=34, 50=70, 75=85, 100=100
linear curve: 0=0, 25=25, 50=50, 75=75, 100=100
compensation for 25= 34-37 + 25 = 22
compensation for 50= 70-74 + 50 = 46
compensation for 75= 85-88 + 75 = 72
so "compensation" curve is 0,22,46,72,100
again, it's important to know where in your rip to enter curves properly;
some rips have place for only one curve, some for two (plate linearization
and dot gain compensation), some for three (plate linearization, measured
[natural] press curve, target press curve).
when you say optimum... nowadays with electronic retouching there's no
"optimum" per se; the "optimum" curve is the one that matches the way your
images were prepared. So if in Photoshop you've got your preferences set to
"swop" and your images are separated and viewed as such, the best for you
will be to have "swop" as your target curve.
All standards curves represent printing conditions with good contrast and
detail. Anything else lies in image preparation.
But of course by matching a curve different to the way the images were
prepared, you get some effect you might want. Some printers for example,
when using CTP, prefer to run a linear plate instead of matching analog
plates. This results in a lower dot gain, more similar to European
standards. And this lightens up the images, and sometimes they prefer that
"sharp" look even if it's not accurate. After all, it's a matter of taste!
best regards,
-- Roberto Michelena
EOS S.A.
Lima, Peru
On 11/23/04 8:48 PM, "Mark" <email@hidden> wrote:
> I've got a Quickmaster DI press at work and I'm want to adjust the
> dot gain and I need some help.
>
> I can basically set any dot gain I want by first linearizing the
> press (plates are imaged directly on press) and then selecting a dot
> gain curve in the RIP. There were some curves supplied with the press
> but I wasn't too impressed with them. I've got a curve that's working
> ok at the moment but it was just put together by looking at the
> little sample graph and adjusting values to get the shape of the
> curve we wanted.
>
> Can anyone provide a more scientific way (without making my head
> explode from too much color theory) for setting up an optimum dot
> gain curve?
>
> Thanks,
> Mark
> _______________________________________________
> 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
>
_______________________________________________
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