Is there a trick to pasting a pixel-perfect vector image from Illustrator into Photoshop as a smart object, and not introduce a bunch of anti-aliasing that was not in the original file? My images are perfectly aligned with the pixel grid in Illustrator, and the pixel preview shows them correctly.
This page is helpful for dealing with vector shape layers, but the techniques he talks about don't work for smart objects. Strangely, he says he doesn't have a problem with smart objects, but I do.
Here is an example of what I'm talking about: a simple 13x13 pixel image in Illustrator turns to crap in Photoshop when pasted as a smart object (I know an image as simple as this could be pasted as a shape layer, but the real images I'm working with are much more complex, and cannot be pasted as simple shape layers).
I'm using Photoshop CS4.
Rendering a vector shape -- will generally get antialiasing.
Photoshop is rasterizing what Illustrator sent, but you can't guarantee that it will rasterize exactly the same because of many factors.
The antialiasing is there because it is supposed to be there -- so it is correct.
CS6 has the option to snap the outer bounds of smart objects and vector shapes to the pixel grid -- but the inner edges can still be antialiased. And that is still correct behavior.
> Photoshop is rasterizing what Illustrator sent, but you can't guarantee that it will rasterize exactly the same because of many factors.
So Illustrator can display the pixel preview perfectly well without any such anti-aliasing problems, but that same pixel arrangement can't be translated to Photoshop intact. Not good enough. It's not like Adobe makes both products or anything.
You are taking vector information from a native vector application (Illustrator) to a native raster application (Photoshop). Even when you bring it in as a Smart Object, it still is being rendered into pixels in Photoshop. This is the same type of thing as translating from one language to another, there isn't always an exactly perfect equivolent between one and the other. Try placing your Illustrator object into a larger Photoshop document, less antialiasing will have to happen and give you better results.
KrisHunt wrote:
…Illustrator can display the pixel preview perfectly well without any such anti-aliasing problems, but that same pixel arrangement can't be translated to Photoshop intact. Not good enough. It's not like Adobe makes both products or anything.
Inconsistency between or among applications in the artificial "suites" should come as no surprise.
The "suite" concept is a fabrication of Adobe marketing and bean-counting types. The engineering teams are totally independent of each other, they are not only in different buildings but in different cities and states of the American Union, even in different countries.
The fact that they have little if any communication among them is highlighted by requests occasionally made in these forums by top Adobe engineers to let the other teams know when there are problems in one application that impact our workflow in another one.
The fact that they have little if any communication among them is highlighted by requests occasionally made in these forums by top Adobe engineers to let the other teams know when there are problems in one application that impact our workflow in another one.
I would be interested to learn what’s a good way to let the Illustrator team know something; the members seem to avoid participating in these Fora, so a Bug Report seem the best approach – it’s just that with my »favorite« Illustrator-bug having been maintained for six versions by now I wonder how much good that is.
(And I am aware that this is not quite fair to the engineers … but the experience is a bit disheartening.)
While I've seen many times that user misunderstandings are at the root of what are initially considered bugs, when I see enough experts commenting here that something's not right, well, I start to wonder whether something may not be right.
Could how to use these tools be better documented so that when someone tries to do what the OP above wants to do, he can actually accomplish his goal? Are you implying the answer has already been given (e.g., enable ''snap to pixel grid'' in Ps CS6) but KrisHunt is simply ignoring it?
-Noel
>> Could how to use these tools be better documented so that when someone tries to do what the OP above wants to do, he can actually accomplish his goal?
I'm not sure. He seems to be expecting behavior that should not be expected. So I don't know what he thinks is happening, or what part he misunderstands. I can document how math and applications work, but not all the ways people incorrectly assume that they work.
The snap vector to pixel grid option will handle the outline of the shape - but the inside may be antialiased depending on the position and size of the shape plus the definition of the original vector shape. Most of the time, vectors are not going to align to the pixel grid, but we can force the outside edges to align. Inside edges generally can't be snapped to pixel boundaries without distorting the shape badly (in the cases where it's possible at all).
Hi Kris - you're not alone. I noticed this on Day 1 of CS6, as I have about 300 icons that I had designed at 512x512 (smart object brought in from Illustrator) that work pixel-perfect in 64x64 when shrunk to exactly 12.5% of their original size. Meaning, all of the hard edges used to stay perfect in previous versions.
But now, in CS6, I'm terrified whenever I have to edit or resave any of these icons, as NONE of them look exactly the same as before. Specifically, hard edges now bleed the color to/from adjacent pixels despite all coordinates being mathematically perfect - like there's an error in the way anti-aliasing is conducted on smart objects. Here's the interesting part: I can open the original PSD, and if I export as-is, it's fine. But if I ever touch the Transform of that smart object, even if I don't change the values, the new anti-aliasing routine takes effect.
This is not a difficult thing to reproduce:
1) Start a new document in Illustrator. Canvas Size 512x512px.
2) Draw a black square with coordinates: X: -64 Y: -64 W: 640px H: 640px
3) Draw a white square with coordinates: X: 0 Y: 0 W: 512px H: 512px
4) Use pathfinder to punch the white square out of the black square
5) Copy the object into Photoshop as a smart object, and look what it does to the inner edge of the left and top sides.
Now, for a control, go back to your Illustrator file, undo the Punch so that you still have both squares. Now copy both squares together and paste into Photoshop as a smart object. No bleeding.
So, I think I can flat out say "something's wrong".
North America
Europe, Middle East and Africa
Asia Pacific