Tag Archives: EPS

The Evil EPS

5 Feb

Now why do we call the EPS file format, more formally referred to as encapsulated postscript, evil? Because, with EPS, we are never quite sure what we are going to get. The format has less than consistent support for transparency advanced clipping masks and spot color. In addition, certain export settings resemble to the old style “flattened” version of digital art, such as the PDFX1/a.

Let’s take a look at how wacky this can get by revisiting our sample art from a previous posting. Notice that the same spot color is used on several different objects. Also, take note of the white portion of the BPI logo; it employs a feature called a drop shadow which overlays the custom Reflex Blue spot color.

Now, let’s make an EPS from Adobe Illustrator with the export compatibility set for CS3. First, let’s take a look at our initial result (Banner_Art_eps).


Notice that the layer structure is intact. It is unchanged from the Illustrator source file. Now, let’s examine the art in outline view (see Banner_Art_eps_Outlines).

Banner Art_Outlines

Notice the complete absence of slices. As you can see in the Banner_Art_eps_LiveScripts screenshot below, the Illustrator scripts for fills and drop shadows are still active. A number of newer RIPs (Raster Image Processors) can read these appearance scripts and correctly interpret them.


So where’s the problem? Let’s toss a bomb into the mix. Let’s place the same art in InDesign and export to EPS and use the highest quality settings available (see InDesign_Export_eps screenshot).


Let’s take a look at the results in outline view (see InDesign_eps_Outlines).


Instead of a digital file made up of vector objects and live scripts, we now have a mixed bag of vector and raster objects. We also have a pretty obvious slicing error. But wait, it gets worse! Take a look at the color values in the logo (see Color_Error1 screenshot below).


That’s not the only potential color problem. Take a look at the cmyk values of the image slice in the logo (see Color_Error2 at left). That particular slice butts up to the vector object with the Color_Error_2original spot color value (see Color_Error3). Now things are getting really complicated. In the large format world, a RIPColor_Error_3 calculates color values and it usually calculates a spot color, vector object in LAB. However, a raster image, like the one in Color_Error3, will be calculated in cmyk. Print this file and you will have two very different shades of blue piled next to each other. Furthermore, print this file and you better be ready to do this job over.

Frankly, the future of the EPS file format is solidly in the past. EPS is rapidly becoming an outdated file format which is being replaced by PDF just like PostScript itself is also being phased out and replaced by PDF. Check out what Dov Isaacs from Adobe said in a discussion on a Print Planet forum about the future of PostScript:

“ …Adobe will continue to support EPS as a legacy graphics format for import of non-color managed, opaque graphical data into Adobe applications (such as InDesign and Illustrator). Although we certainly do not recommend that new graphical content be stored in EPS format (except to satisfy the need to import data into page layout programs that aren’t quite PDF-centric), our user base should feel comfortable that there is no need to worry about a need to convert their very sizable libraries of EPS-based graphic assets.”

When a principal scientist with Adobe employs the phrase “legacy graphics format,” it’s time to move on.

Flattening and transparency are pretty confusing subjects. If you’re still curious, or just plain lost in the woods over flattening and transparency, check out the online tutorial below for more information.


How to Have a Bulletproof Workflow

21 Feb

So how do you manage your projects so you get predictable results?  Answer: manage the two different forms of digital assets (raster and vector) so that the color range of your print file fits within the range of your output device.  This is sometimes called a CMYK neutral workflow.

Let’s talk a little bit about print files.  The basic types of print file are the JPEG, TIFF, EPS (encapsulated postscript) and the PDF aka…Acrobat file.  JPEG’s and TIFF’s are strictly for raster/image/photo file types.  EPS and PDF/Acrobat files can contain raster and vector art.  You may have heard of postscript (PS) files, but it’s used less and less these last few years. Postscript has essentially been replaced by encapsulated postscript (EPS).  For the sake of a more compact post, let’s concentrate on the EPS and PDF file types.

There are several structural differences between the two file types.  But, the main difference is ICC profile support. The EPS file does not support the inclusion of ICC profile data; Acrobat does.  And this is why so many printers and large format service providers use EPS as their preferred print file.  What your output provider wants to avoid is a conversion phenomenon akin to double-dipping.  Most RIPs look for profile data.  When they find it, they will (unless managed by the RIP operator) apply another instance of image compression to the print file to fit the color range of the specified ICC.  In other words, your compressed photo gets re-compressed.

ICC’s are not really a perfect model for vector art.  Think back to the diagram, InkjetVsSWOP.png in a previous post. If you strictly interpret the 3D lab image for offset printing, you would conclude that a perfect green (100% cyan and 100% yellow) can’t be achieved on an offset press.  We’ve all read enough magazines to know you can get bright yellows, reds, greens and rich colors from offset printing.  So, what’s going on here?

You’ve probably guessed it.  At some point, your output provider is discarding the ICC profile information.  And it sounds like heresy after all you’ve heard about the necessity of working with agreed upon profiles for color management.  But don’t let your head explode just yet.  Let’s get back to image management.

Image management is simply making sure all your elements (text and photos) conform to the range of your final output device.  Pick the appropriate workflow and work with the end product in mind.

Here are a few tips for you budding graphic designers:

Consider color conversion a one-way street for your vector art.  Your classic light blue (100% cyan) will appear similar when you switch your file to RGB. But, if you were to switch your file back to CMYK the values are transformed to roughly 70% cyan and 14% magenta.  We always make my vector art in CMYK. That way, if we have to make an RGB graphic from it we don’t have to worry about missing a corrupted value for print.  Also, our web graphics will more closely resemble my print files.

Do NOT convert vector art to an ICC profile. Occasionally, you’ll see this option when “Placing” (linking to an external file) or pasting other art into your current working document.  The color values will shift in an unpredictable manner, just think back to what happens to 100% cyan when you change it to RGB and then change it back to CMYK.

Make sure all your scans, photos and Photoshop art share the same profile and color space before assembling your art.  We know most applications have a “Convert to Profile” export option.  This is not a reliable way to work.  When you employ blind processing, you’re asking for an unpleasant surprise. You’re much better off doing the image work yourself.

Use TIFF’s instead of JPEG’s (and ZIP compression when exporting to PDF). Once upon a time, JPEG compression was the recommended way to save images for print due to limits on memory and processing power. One way JPEG compression controls file size is by consolidating color together within a certain range of values.  You may have heard of JPEG’s as a lossy (or lousy, if you’re as picky as we are about image quality) form of compression.  If you’ve ever wondered where those boxy looking squiggles or blocky halos around text came from in digital art, now you know.  In addition, the default compression for saving as a PDF is JPEG.  When you pile JPEG compression on top of JPEG compression you’re just asking for an unattractive result.  We now have more memory, much more powerful CPU’s, graphics cards and RIP software.

If you want more information on JPEG compression, check out the Wikipedia page: http://en.wikipedia.org/wiki/JPEG

And finally, CYA!  Conserve Your Assets!  Always keep customer supplied originals, unaltered.  The client may ask for them at some point.  You may need to compare appearances or justify edits and output with your client standing over your shoulder. CONSERVE YOUR ASSETS; it bears repeating. 🙂

PS: You may have noticed the inclusion of the Light Jet and Durst Lambda under Large Format RGB. Conventional large format inkjet imaging has progressed to the point of photo quality output which can (IMHO) outperform the Durst and Light Jet machines.  Also, new archival, acid-free papers and fade-resistant inks are eliminating the advantages of large-format photo output for lifespan.

Flawless Printing

17 Jan

Like every other skilled field, large format printing is rife with hidden issues just waiting to inflict damage to your health, sanity and pocket book.  If you can avoid the lions, tigers and bears in the Forbidden Forest you’re half way to the Emerald City.  So here are my top four print production pitfalls.

1: Stop trying to save money with inexpensive software applications.  Often those applications, which shall remain nameless, are designed by companies who refuse to purchase a full postscript license.  Just like LAB values drive all color values for digital printers; postscript is the language which drives all digital printing.  They endeavor to find their own unique solutions to digital imaging outside of the accepted standard.  This may work fine when you print from within the application to your desktop printer; but will often fail to produce a quality image when you attempt to save as a PDF.  More than occasionally, those PDFs will fail in the real world of large format and offset printing.

2: Don’t try to do everything with Photoshop or one of its competitors.  You need to be able to deal with both raster and vector art for quality design.  In addition, unless you are extremely knowledgeable concerning design for print, you can really paint yourself into a corner with your digital art.  There are numerous technical challenges just waiting to swoop down like a gang of flying monkeys on the unwary innocent when it comes to subjects like color conversion, image manipulation, text management and PDF creation.  We always recommend that design personnel be equipped with Photoshop and Illustrator at a bare minimum.

3: Work with people that understand print production.  They may not be the least expensive hires in the world, but they will save you and your firm time and money.  Why?  You will avoid the most expensive business expense in existence – doing something twice (or three times).  Your print order will be completed on-time and on-budget.

4: Watch out for phony vector art.  Just because a file is saved as an eps (encapsulated postscript), it does not mean it is vector art. Over the years we have received several bogus logos from large corporations.  Someone would place a low-resolution jpeg (usually snagged from a company website) inside an Illustrator file and save it as an eps.  Then, submit that art for a large banner project.  Sorry folks, there is no magic eps dust that converts tiny web graphics into 4 foot by 8 foot banner ready art.

We can hear what you’re thinking now. You’re thinking, “Okay, smart guy; tell me how things should be set up and the explanation better make sense”. And, that’s what I’m going to do.

Next time.

%d bloggers like this: