Is composite color editing my print media ads in photoshop a frowned upon practice?

Question

Obviously with print media, the norm is to design the ad in vector, CYMK, in a tool such as InDesign, and export straight to PDF or something.

But I like to put a vivid "edge" in the coloring of my design. To do this, I export my design from InDesign to PNG at 700+ppi, which keeps texts and images very crisp through the rasterization process. I use PNG because it's the only exportation option that I find accurately reproduces the design's effects in a form I can use with Photoshop.

So in Photoshop, I use all of the powerful RGB color alteration tools to bring my colors and darks and lights into a more vivid and more interesting composite result. For example I might add some slight blue into the black rages for an abstracted, rich feel.

I always find that this makes my ad much nicer and more professional looking.

I convert to CYMK, export to EPS (Photoshop creates massive EPS and PDF file sizes) and use Adobe Distiller to convert the EPS to PDF. Distiller somehow keeps this 700+ppi image at the correct resolution and drops the file size down to <20MB and my ad is done!

It sounds like alot but with practice it only adds 10 minutes to my design workflow and creates a product that feels far superior to what I could do with InDesign alone.

But my co-workers hate this! They look down on the idea of rasterizing anything for print, having been taught in college "Never rasterize!!". They say I could brighten my colors in InDesign. They say no design firm would do what I do.

But I feel that many skilled designers must be doing this, judging by the advanced color effects and curves I see in magazine ads. And I know that rasterizing upward of 700ppi creates the same image crispness of a vector, just a slightly larger file size.

Maybe my co workers could somehow be right to look down on this practice. Or maybe they don't have an eye for perfection and extra effort in a design, feel my way is too complicated, without a good cause. So I wanted to ask the design experts:

Would my design practice be frowned upon by experienced print media designers / experts?

Answer 1

Don't take me wrong anithing I will write here.

One thing is extra work and one thing is usless (not usable) work.

1) The 300 ppi output for a raster image is not due people want to save pixels, it is for an intrinsec limitation on a print workflow. Yea in a high quality print you can use a 400ppi image... But 700ppi? No, not really.

People do not realize this. On a comercial magazine the resolution you are seing is in fact 150 lpi, not 300 ppi!. The 300 was only for a specific part of the process. After that you only have 150 units of information per inch.

Again... 700. (If you are using an extra resolution just as a precaution for future posters its ok)

2) you do not rasterize a text just "becouse".

Here is a macro photo I took to a negative, you can see a string of dust. All raster text gets "screened".

(enlarged image of a 150lpi negative)

This is that the text is just now a part of a big photo. If the text is big, there is not much problem, but on small texts this will blur them and can be a problem.

3) The "vivid" colors you have on a rgb image on your screen WILL be transformed into cmyk. The problem is that thoose values will NOT be as acurate as if you put them by hand on cmyk directly.

And by acurate I mean, if you want a pure cyan 100% you need to write cyan 100% on a cmyk mode.

Here is an image, one of a color wheel simulating a rgb file transformed into cmyk using adobe 1998+Swop.

_{(source: otake.com.mx)}

You can notice how a pure cyan is moved away a little bit.

The matrix used on an rgb conversion is a little complex. In this page you can see an animation of a simulated 3D RGB space vs a simulated CMYK Space.

http://www.otake.com.mx/Apuntes/RGB-CMYK/RGB-CMYK-4.htm

4) You can have issues with the color of a text.

For example a black text on a white paper needs to be printed in black, this is 0C0M0Y100K

In a RGB file the black WILL be printed as a combination of all screened inks C+M+Y+K. A real mess on small text.

5) You do not have any controll on the overprint. You also can have problems, depending on the case.

Use raster images when you need them

Yes some effects need and should be done on RGB mode alone, but that is a prior work. Probably you need to review your workflow a bit. Probably you are doing it right, but NOT becouse you think you are limited by a cmyk workflow.

Aditional notes.

and drops the file size down to <20MB and my ad is done!

File size has nothing to do with anithing. Al thoose magic passes from one aplication to another... they have no sense to me. All you are doing is avoiding making a cmyk file, (and probably using lossy compression somewhere)

A raster image with all thoose not usable resolution only increase process time on the rip. and will be downsampled at some point.

When to work in rgb all the way?

When you are making a digital print. And gess what... the resolution needed there is just 200ppi.

Epilogue

Would my design practice be frowned upon by experienced print media designers / experts?

Again the part of using a RGB file for some effects is right, the part of doing an extra resolution is righ too if you are trying to make a larger print on time.

You can do whatever workflow works for you, but do it becouse you understand it and controll it, not becouse you just "feel" its ok, and everyone else needs to follow your steps.

Answer 2

That would be interesting to see one of your design example and one of your co-worker design to see what you mean.

RGB, CMYK

There's no issue in working in RGB and then converting to CMYK if you know what you're doing. Anyway some filters are not even available in CMYK mode.

What counts is the final result; 1) does it look awesome and 2) Is it well printed or could it have been better?

I do the same thing as you to get maximum impact and that luminosity. It brings a dimension to the images, it's not "flat." And because I don't use fancy effects that (I think) should not be done in vector, my stuff prints well everywhere. I still use Distiller too.

In fact, there's a lot of situation when it's better to use composite effects rasterized instead of the vector option; for example some gradients will have that bending effect amplified in vector while you can make them smoother with Photoshop and some noise & blur. And if you do ads for online use, it looks ugly to have those white lines caused by transparency in the PDF because it was done in InDesign or Illustrator.

For the RGB converted to CMYK sometimes it's easier this way. And then you simply boost up your colors AGAIN in CMYK. It's easy and I guess that's what you do. As long you end up with CMYK, it's alright.

Regarding the rasterizing part:

Yes, your images will ultimately get converted back to a lower ppi at the print house. But if you're using a higher ppi it doesn't hurt, in fact you might be able to reuse these designs easily and save time if you need to work on a bigger size layout. It's smart.

You need to be careful with the text and only use these things on your picture montages when you can. But let's make things clear; we're talking about body text, subtitles and the like. Also, if you need to do a map for the subway system for example, yes you should prioritize clarity and vectors above "shinnies." Always put the priority first on the main goal of the ad or layout.

When you create a poster or a flyer, you can freely use the main title and the text boxes with an effect and have fun with your design the way you do already, but make sure you rasterize WHEN IT'S NECESSARY. If you can do something in vector, for example a text title with no effect and has a simple color fill, do it in vector.

One way I like to do this, for example, is to create the montages that need to be composite in Photoshop but if I have text I typed that can be easily converted to vector, I do it at the last step of my design; I'll keep the shadows or highlight effects in Photoshop, remove the text layer, and I'll add that text part in vector on top of my composite montage in InDesign or QuarkXpress. If I can't replicate that effect in vector then I don't and I leave it in Photoshop.

Some effects are meant to be done in Photoshop. For example, a designer can swear vectors are the best and decide to do a golden letter effect in vector... but compared to one done in Photoshop with all the shine added to it, that vector will look fake and plastic! So in that case you use Photoshop because it gives the best final result. Another example: They'll use clipping paths while in fact a channel mask would be nicer and smoother. At this point, it's laziness that can be blamed.

Others

Some designers think using vector will make them better but when you look at their layouts, they're boring as... well they're boring. But yes, their boring design is sharp. They might not attract attention but, hey it's sharp...! My philosophy at this level is: Microsoft Word too can output sharp designs, but it doesn't always add value to the eyes of the target market to who the ad is destined to be sent to.

If you use the maximum of all the available features and the max quality you can get for print, go for it. Some designers forget that SOME clients don't care about a big title being in vector or not, they don't even know how to appreciate good printing but they ALL know how to appreciate what they feel is awesome. So that's the challenge basically, using the max of everything. It's your job to make sure you don't only prioritize style only but clarity too.

If you were hired by one of the most prestigious magazine to do their layout, they will expect you use the maximum of rasterized images and vectors combined when it's possible; but they will find it idiotic to limit yourself in a certain kind of design because "it was always done in vector." The response will be something like "I don't care, it look like... well it looks boring. Do something with more 'pizzazz'!"

So make sure you use vectors or even line-art bitmap @1200ppi+ colorized when you can. For the rest, keep doing what you're doing. Some designers are simply annoyed seeing others using different techniques because it makes them feel insecure that you get better results with your own techniques. But people who dare to do things differently and get good results are the ones who are called innovative. The others are just followers or replicators.

Please do not use small text or shapes that could easily be done in vector in Photoshop. If you do that then yes, it's wrong. Find a way to make that text look good without rasterizing it. Or put the emphasis on something else.

Can't you show an example anyway of how you split your vectors and your rasterized elements? It's still possible your co-workers are right too. You can't rasterize everything, that's for sure.

---

Some info on ppi and lpi

And extra trick if you like to do small layouts in Photoshop or import your text easily to add it back on top of your composite images in InDesign: How to keep the text in vector in Photoshop without rasterizing it or flattening the layers when exporting to PDF?

Sometimes, rasterizing is necessary and why it works better with Distiller

Careful with rich black when converting back to CMYK and this with more details

Optimizing PDF even more (You can still output a .ps and use Distiller, and then use this in Acrobat!)