Comments on: Sandwiching Text or Logos Inside an Image

By: Niall Funge

Niall Funge — Sat, 23 Jun 2007 15:31:33 +0000

On the output provider side of things for images like this I have noticed problems when supplying layered pdfs to digital output bureaus - shadows turning to solid black etc. The quickest and safest workaround is to save as PDF-X/1A files. This is a cmyk flattened pdf file which almost any output device should be able to handle. This comes from experience of getting files out on anything from Postscript Level 1 devices onwards.

By: eugenetyson

eugenetyson — Thu, 21 Jun 2007 23:15:21 +0000

I would love to know what the transparency flattner is all about!

On the other note, I’m not the best at wording my things, as you can see. I can pixel edit anything and photoshop, illustrate, design anything, ask me to explain it and I can’t get the words from my brain to the page. I don’t know why. I know how to do it, just not to explain it. So forgive me if my comments aren’t completely accurate.

By: David Blatner

David Blatner — Thu, 21 Jun 2007 20:08:55 +0000

Dave, there is no doubt that ALL of the transparency effects (in this or any other program) requires some work on the part of the output provider to ensure high quality. But there is equally no doubt that you can get excellent quality, without the glitches you refer to. One of these days we’ll need to do a post or podcast on the Secrets of the Transparency Flattener!

By: Dave Courtemanche

Dave Courtemanche — Thu, 21 Jun 2007 20:04:57 +0000

One problem with this technique, done without a clipping path, is that some imagesetters will have problems with the file. I used this technique over an Illustrator logo with a gradient in it. I hadn’t drawn a clipping path and when the file was ripped, the gradient rendered differently under the transparency compared to the logo outside the transparency area. Just my 2¢.

By: David Blatner

David Blatner — Thu, 21 Jun 2007 15:31:55 +0000

No problem, Eugene. The more ideas the better. However, just so I’m clear: the point of my article is that clipping paths are probably not the best way to create this effect. Using Photoshop’s native masking is much better because you get a soft antialiased edge between the pixels and the vector logo/type.

By: Gerald Singelmann

Gerald Singelmann — Thu, 21 Jun 2007 15:30:47 +0000

>Of course, clipping paths is the correct way to perform this.

Not with furry images, it isn’t. There David’s procedure to use two PSD layers is far better.

By: Eugene Tyson

Eugene Tyson — Thu, 21 Jun 2007 15:11:51 +0000

Guys, I wasn’t trying to suggest my way was better , just a different approach for people that can’t pixel edit.

Of course, clipping paths is the correct way to perform this.

By: Gerald Singelmann

Gerald Singelmann — Thu, 21 Jun 2007 10:13:39 +0000

>So why did I bother?
Because some images allow path edges and in those cases I prefer to have the image placed only once.

I find this technique especially valid when the image already has a (clipping) path saved in Photoshop you can use in connection with the Pathfinder to build the mask for the text / logo.

By: David Blatner

David Blatner — Thu, 21 Jun 2007 03:06:16 +0000

This idea of using the pencil tool got me thinking: You could use Paste Into to nest the text (or other vector logo) inside another frame. Then select the Pencil tool, click down on part of the outer frame, then drag along the shape you want to mask. Let go of the mouse button when the cursor is back on top of the frame again.

This works because the Pencil tool replaces portions of a frame when you start drawing over it.

But of course, it’s still a hard edge! So why did I bother? Because it’s fun to try.

By: David Blatner

David Blatner — Thu, 21 Jun 2007 02:51:26 +0000

Eugene, that’s good thinking “outside the box,” but this has several problems. The main issue is that it’s a hard-edged clipping path. You could fake anti-aliasing with feathering, but it’s rarely good for anything other than a comp. If you didn’t own Photoshop, I suppose it might do in a pinch. But ultimately, it comes down to this: As soon as you want to do stuff at the pixel level, you want an image editor.