Shape layers will print as vector. Rastered layers will print as…well… raster.
If you have a sufficiently high resolution output device, shape layers will deliver superior print from lower res files. There’s a balance there that depends on the resolution of your image and your final output – low res files and high res output devices will show the grestest difference between the shape layer and it’s surrounding pixels, high res files and low res output (60 LPI laser screen, for instance) the difference will be difficult for anyone to spot.
Converting to a Shape makes vector paths and means the type is infinitely scalable. A Postscript printer will print such type smoothly no matter what the size. But it won’t really be type anymore.
Rasterizing turns the type into pixels. It’s just an image then.If you scale it you will see the difference. Of course if the printer is non-Postscript it won’t make any difference.
Hi Mike,
These normally print between 100-200 ppi. Oh ah – I did not realize that converting to shape was the same as vector. Now I clearly understand the difference. Another type question along this same line is should I change the type anti-aliasing to none before I convert to shape ? I have printed several examples of the different anti-aliasing styles and really see no difference.
-James
By the way, the best thing to do is just leave it as type layers if possible.
I notice very little difference personally between the antialiasing methods (maybe they’re more worthwhile than I know) but I just did a little test. Shape layers antialias in only one way, and in my quickie 30 second on screen test, their antialias method seems closest to ‘Crisp’ text antialiasing.
So to answer your question it won’t matter what you set your antialiasing method to – shape layers *will* antialias in their own way – at least on screen. What happens inside the printer is a whole different process – vector information is processed separately and (I think) essentially disregards the antialiasing setting from photoshop. Then again, I could be wrong on that – I’m not as much a wizard on the internal machinations of postscript RIPs as I’d like to be.
I hope I helped 🙂
Edit: oh, as Ed mentioned, leaving it as type has the added advantage of being more editable, and may also have another added bonus, depending on the RIP that it’s sent to. Some RIPs handle type independently of other vector data, sometimes applying automatic overprint, sharpening, or the like. So… the variables increase!