Untitled Document

We print once a month on newsprint. Not the best newsprint but not the worse. The last couple of press runs have been heavy on ink. I don't know if I should try to adjust for it or not. I output to PDF...the x1 preset if memory serves. What I need to know is should I try to adjust to what they are doing or am I doing something wrong? Since I do most of the color pages I have some control...but there have been some other pictures by people who know less than I that have printed better? Did they luck into a better setting than I?

Any ideas?

Thanks,

Gene Palmiter

____________________________________________________________________________

On Sep 12, 2005, at 1:24 PM, Gene Palmiter wrote:

Newspaper printing is a train wreck, and ripe for improvement. They have all sorts of challenges such as having to print on multiple presses each with different behavior and possibly completely different mechanical make-up. It's not uncommon for presses to be modified to print four color instead of being four color presses, and depending on how the print units are organized, you can get more paper stretch from the first to second units than other presses. So to deal with this, they print yellow first which shows paper stretching the least. Hence, that press has different print sequence than the others.

Believe it or not, most newspaper presses don't run color bars. So how in the world can they have effective process control when they aren't running a color bar? And most who do run a color bar are only running a 3 color gray bar. This tells us nothing about the key channel. If it goes crazy and separations use heavy black generation, the images will get clobbered. If you use less aggressive black, then the image is more susceptible to hue shift as CMY oscillate, and also more susceptible to poor register.

All of the problems with newspaper printing have to do with process control. Some of these problems are real, and aren't solvable without upgrades to hardware, but others have to do refusal of newspapers to treat their product like the money generator it is. Newspapers can't claim to want high quality newspaper printing and not have color bars to enact effective process control measures. Some can't have a trim area for the color bar, and don't want to incorporate them into the newspaper anyway (i.e. they look bad). I say, too bad. Either trim it, print it, or don't claim quality is important.

The specifications that have to do with newspaper printing are completely absent on specifying anything other than 50% and 100%. The real problem is below 50%. Since there is no standard for what 10% or 25% tone reproduction is supposed to be, either they have to know so you can make separations accordingly, or it's a guessing game. Presumably they are using SOMETHING as a source profile for producing proofs. I'd ask for that profile. There are a bunch of newspaper profiles and they are all useless if you aren't selecting the one used for proofing because there is so much variation in newspaper printing.

Also, the media they print on has lower quality than toilette paper, even though the phrase "printing on toilette paper at 90 miles an hour" is commonplace in the newspaper industry. They get what they can get, it's cheap, it's highly recycled, and contains a varying amounts of chemicals that can resist ink. So day to day the media is different which obviates compensation through profiles. Obviously it must be done by compensation through RIP transfer curves and other best practices process control techniques.

Until your newspaper gets a clue and modernizes their processes and possibly also their equipment, keep asking for credits.

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

____________________________________________________________________________

The Newspaper specification is SNAP. It has the density and gain numbers for 25-50-75%. I have written several articles on process control at <www.newsandtech.com> search

under articles ' remaley'. High GCR is necessary for quality newspaper printing. Photoshop

has a newspaper set of values - but who uses them? Also go to <www.ifra.com> they have established

that in Germany ALL newspapers will have a gray bar made of 40-C 30-M 30-Y for process control

- prints the wisth of the sheet.

Dan Remaley PIA/GATF

____________________________________________________________________________

Lots to read there. I will get started on it. Who does the GCR? Is that built into the software somewhere or does it have to be done by someone....in our case that would be me.

____________________________________________________________________________

On Sep 12, 2005, at 6:24 PM, Henk Gianotten wrote:

Even ISO 12647-3 does not, insofar as I can ascertain, define a target highlight value, nor does it specify specific colorimetric behavior. That means it's still a barndoor that means adherence to ISO 12647-3 can still result in considerably different behavior among printers. 0% to midtone can vary among printers because only TVI, TAC, SID and maybe printing contrast are in the specification. If the value of 10% for any or all channels or overprints is specified, I'd like to know where the spec says this.

Chris Murphy

Color Remedies (TM)

www.colorremedies.com/realworldcolor

____________________________________________________________________________

--- "Gene Palmiter" wrote:

If you are converting to CMYK, then the profile or separation table is where it happens - it is a part of the conversion process into CMYK from other (single channel or three channel) contone modes to four colour process.

It used to be set by the drum scanner op in days gone by, or in a proprietary prepress system. Then along came Photoshop...

If you are in CMYK, then it would be a human doing this with, selective colour, hue/sat, channel mixer and other tools, not fun but I am sure that there are old time separators who can do this blindfolded (although Macromedia Xres used to have a UCR/GCR operation/filter which was applied to a CMYK file, which I never used on a live job and I never had the time to figure out what it was doing).

More on newsprint in these archives:

http: //www.ledet.com/margulis/ACT_postings/SeparationIssues/ACT-Newsprint-Setting.htm

http: //www.ledet.com/margulis/ACT_postings/SeparationIssues/ACT-Preparing4_localnewspa.htm

Regards,

Stephen Marsh.

____________________________________________________________________________

GCR is applied in Photoshop (color settings)under 'Light, Medium, Heavy, Maximum'.

Also, images that are already cmyk can be converted to Lab then converted back to cmyk with GCR applied.

(Quick Tip) - Ever had a problem with Powerpoint files that have solid y-m-c under type or black only panels? Simply convert with max. GCR in Photoshop and all that color under type goes away!

Dan Remaley

____________________________________________________________________________

Found it with the help of this PDF...for the CGR settings one must first choose to make a custom CMYK. So now my CMYK reads SWAP (Newsprint) 30% GCR Heavy....any reason to believe this is a mistake for the next press run? Given that the last 2 press runs tended to run dark with heavy ink coverage?

palmiter_gene@verizon.net

http: //www.photoshopforphotographers.com/pscs2/download/PSCS2_colmanage.pdf

____________________________________________________________________________

OK...I am going into Edit/Color Settings in PS-CS2. I see no mention of GCR. Could it have been moved somewhere else? If I leave the photos in RGB for conversion by InDesign then is any GCR getting done on them when they finally do get converted? I will have to look through InDesign to see if GCR is mentioned.

____________________________________________________________________________

gene wrote:

I understood everthing was running too dark... only reason to bet the farm on HEAVY GCR would likely be a bunch of neutral images.. maybe stainless steel cookware ad... I would suggest medium GCR and 33%(min) and would not hesitate go to 36% depending on the image.. regards, jim bean

____________________________________________________________________________

OK...I was taking a stab at it based on the curve it showed...I must have been misreading it. I know that I will have to, given the lack of science in our current workflow, make small changes and get it zeroed in over time. Hey! I am the only one who sees a problem with it now...the others just don't know any better.

To clear up my mistake...I was thinking from the curve I saw that HEAVY meant a heavy application of GCR but you say it means a heavy application of black ink.

____________________________________________________________________________

Gene Palmiter writes:

Gene, did you see my message? It should have all been there in the links too.

You have to look under the "Custom CMYK" option in colour settings or convert to profile. Then you can select the amount of GCR, dot gain, ink type, stock etc.

You sound confused.

As I wrote previously, GCR/UCR is a part of the conversion process, it is part of the CMYK profile, colour table or custom CMYK etc. Adobe do not say what GCR their SWOP Coated TR001 profile uses, but it appears to be something similar to medium in the old custom CMYK speak (but not the same).

Chromix used to offer TR001, but they did offer UCR and various GCR options and ink limits named in their profile.

These are just examples, I know that this is newsprint.

So, either find a 'proper newsprint profile' that does what you need (yeah right) or use Custom CMYK in Photoshop to make your own newsprint profile.

As for your other question above...

If you are using RGB in InDesign and you output the images as CMYK, then the GCR is a part of the conversion to CMYK.

I personally do not like or use workflows where one works in RGB in illustration and page layout for illus and layout colour builds, letting a profile convert your contone images is one thing, but letting a profile handle all your solids and tints is not for me or the type of work I do.

There are many posts here about custom CMYK, and Dan goes into it in Professional Photoshop if you have it.

On newsprint, I would be doing something similar to this in custom CMYK for a ballpark conversion, not knowing anything about the issues you face (apart from dark prints):

SWOP Newsprint, Dot gain CMY 30 K 35 (or perhaps 35/40), GCR, Medium, K limit 85%, Total ink limit 220-240%. The dot gain can be further adjusted, that is only a simple suggestion only concerned with factoring in more K gain.

You will live and die by the black plate, so factor in more gain. It will often depend on the tearsheet that you view on how good the rego and colour is...compare three to six or more if you can, it is usually all over the place.

With newsprint, finding the 'correct' conversion settings for the image type is only the first part of the battle. One often has to push contrast and clean up primary colours to print more pure than the default conversion.

Stephen Marsh.

____________________________________________________________________________

--- "Gene Palmiter" writes

Guess what Gene, they are one and the same!

The grey component of an image is formed from CMY, replace a heavier amount of black and proportionally remove the right amount of CMY from the grey component and you have the same LAB value via a different CMYK mix. Heavy GCR pretty much means a heavy black plate.

Stephen Marsh.

____________________________________________________________________________

While on the subject of newsprint, I thought I would offer this tip.

I would appreciate hearing your thoughts on this issue.

I have found that when having to separate out of gamut or pure RGB colours, such as solids in logos or in computer generated gradients, going from RGB direct to Newsprint CMYK leads to solid colours being way off lighter in tone and gradients also having similar issues with hue and brightness. It would appear that even smaller gamut's like ColorMatch RGB are obviously too different to newsprint CMYK to do a good direct baseline conversion. The newsprint profile could be Custom CMYK or one generated by a third party profiling package. Results are similar in this respect (tonal transitions are often better in the proper ICC profile though, over custom CMYK).

So rather than going direct from RGB > Newsprint CMYK, try an intermediate step.

RGB > Euroscale Coated v2 Perceptual > Newsprint CMYK.

Then compare the colour values and the look. I have not tried other intermediate CMYK spaces or edits (K channel, selective memory colours etc).

I find this interesting. Results are much closer to the original (say the RGB file has a solid logo in 'impossible' pure RGB cyan 0r255gb or a pure bright red).

With the Custom CMYK there is no perceptual render intent, but even when using a proper newsprint profile using perceptual, solids are still poor (gradations can be better though). However, the Euroscale Coated profile before the newsprint profile helps in solid colour builds.

Yes, there are other alternatives to the issue of solids, simply using the RGB channels direct in the CMYK file, or channel mixing. These do not account for gradations of colour and tone, where as the intermediate CMYK profile step does.

Regards,

Stephen Marsh.

____________________________________________________________________________

on 9/13/05 4:12 PM, Gene Palmiter at palmiter_gene@verizon.net wrote:

Gene,

Yes, but I can't say why.

I read the recent posts regarding GCR and there's no problem there (I'd also suggest medium in your case). I simply don't think that its going to be a fix. My instincts tell me something else is going on. So for what it's worth I'd say you could find more variable ink coverage conditions, especially with newsprint, as you continue.

Lee

____________________________________________________________________________

One of the few advantages of the media consolidation that's been going on is that it becomes more important for a company that prints several newspapers to have them printing under similar conditions. And this means it can be easier for us to predict what those conditions will be.

For instance about a year ago I had to prep some movie ads that were to be run in papers across the country. Since I'm in LA I called the Times and got in touch with their color guy, sure enough he said the parent company had gone to great expense to get them all running close to the same way. Even better, they had created a profile that was created from several samples that had been averaged together.

I used this profile for the final image, and later compared the actual printed piece to the file and was very happy to see I had a pretty good match.

In this case there was an icc profile that filled the bill. Perhaps Gene could be just as lucky as I was. Never hurts to ask...

Dennis Dunbar

____________________________________________________________________________

Stephen writes,

Correct. If there is detail in an area of bright colors in any RGB, there will be major problems bringing into a newsprint setting. The problem exists no matter what kind of CMYK you're going to, but if it's decent quality printing it can probably be lived with. Not newspapers, though--just too weird.

The only realistic choices are to assume that correction in CMYK will always be necessary, or to correct prior to the conversion to bring everything close to the newsprint gamut.

That is one of several ways of doing it. There are other shortcuts that may work better on certain images.

Basically I think the problems fall into two broad areas:

1) To emphasize the brilliance of the out-of-gamut color, we want to subdue all similar colors in the image.

2) We just want to retain the detail in the OOG color but we would like other colors to be unaffected.

If it's #1, a sensible way to approach it is, (if the file already is not in Adobe RGB) Convert to Profile>Adobe RGB followed by Assign Profile>sRGB.

If it's #2 the best results by a long shot seem to come from working in LAB and weakening the OOG color by moving it closer to 0a0b. Easy to do this with the AB curves because those OOG colors will be in an area of the curve well away from the less saturated ones.

Dan Margulis