Join Tim Grey for an in-depth discussion in this video Basic output workflow, part of Photoshop CS6: Optimal Output.
Whenever I'm preparing an image for Output. I always go back to my master image. And that's the image saved at it's original resolution. In the Photoshop PSD or TIFF file format. That master image will also include all of my Adjustment layers and Image layers that were used to creat the final result. Taking this approach means that I'm always starting from the same original image. It also means that if I've made any changes to that original image they'll be reflected in any new output. I prefer to take a bit of a cautious approach when I'm preparing an image for output in large part.
Because that process is by it's nature destructive to the image. I'll generally be flattening the image and likely resizing the image depending on how I plan to share the photo. And those changes alter the original pixels in the photo, so I always work with a copy of my image when I'm preparing it for output. To do that, the first step is to create a duplicate of my photo. Not a duplicate of my background Image layer, but an entire copy of the image itself. To do so, I'll choose Image, and then Duplicate from the menu. That will bring up the Duplicate Image dialog.
The default name of the image is the existing filename, plus the word copy. So that if you were to save this image, it would have that filename. That's perfectly fine for my purposes. I just plan to print this image and I don't think I'm going to need to save that derivative image. But if you were going to save the final result, then you might, for example, change the name here to reflect the output size. For example, Elephant 8 by 10, in this case. If I were going to print an 8 by 10 inch print. I want to flatten the image in the process of duplicating it, because that will ensure I'm working with a single layer.
That reduces the memory footprint, which can make a little bit of a difference in terms of the time it takes to process the file. But more importantly it ensures that the image no longer contains multiple layers. If I had performed some image cleanup work on a seperate layer, for example, or applied some sort of special effect on another layer. I want all of those layers to be compressed into a single layer, so that my sharpening will affect all pixels evenly. So I'll go ahead and turn on the Duplicate Merged Layers Only checkbox. This is a little bit of a confusing name. It really just means flatten the image when making a copy of it. So with that option turned on, I'll go ahead and click the OK button. And now you can see that I have two copies of my image. I have the original image with all of my layers intact, and I also have my duplicate image, which has been flattened.
At this point, just to make sure I don't get confused, I'm going to go ahead and close my original. I'll just click the X on the tab or that image. So now I have just my flattened version open. Next I need to re size this image to it's final output size. I'll go ahead and choose Image. Image size from the menu. And here I can specify the perimeters by which I'll re size the image. It can be helpful to turn off the re-sample image check box initially. So that you can get a better sense of the potential output size for the photo.
In this case, I'm working with a very low resolution sample image, just for demonstration purposes. So it's not going to print very large. But I can see exactly how big a print I can get from this image without applying any interpolation. By turning off resample. And them setting the output resolution to my desired value. For a photo inkjet printing, I typically use 360 pixels per inch. And that's because most photo inkjet printers render the data at 360 pixels per inch. You could also use 300 pixels per inch for offset press, for example. But keep in mind that this resolution really only applies for printed images. For images that are being displayed on a website or via digital projection, this pixel per inch resolution has absolutely no impact on your results. This resolution simply determines how the pixels included in the photo are going to be spread out on the paper to produce your final print. I'll go ahead and change the unit of measure to inches. And you can see that with no interpolation at all, and at 360 pixels per inch I'm able to produce a very small image.
About a 3 by 4 inch print. But again, that's just because I'm working with a low res sample image here. If I want to change the actual output size of this image, then I need to turn on the Resample image check box. If I were preparing the image for online display, I could then work exclusively in the pixel dimensions area. Specifying the maximum width that I want the image to appear for example. But in this case I'm planning on printing the image, and I plan to print at about a 4 inch by 6 inch size, essentially a snapshot size. And so I'm going to change the document size width to 6 inches. You'll see that I now have an image size that 4 by 6 inches at 360 pixels per inch.
Taking a look at the top of the dialog you can see that this image was at a little over 4 megabytes and now it's almost over 9 megabytes. Normally producing a 4 by 6 inch print from an image would not cause the image to be enlarged, but rather to be reduced in size. But since I'm working with a low res image, in this case I'm increasing the size of the photo. I also encourage you to leave the Constraint Proportions check box turned on. That will ensure that the aspect ratio of the image is not changed. For example, you can't stretch the image in one direction or the other.
Which can obviously cause a little bit of distortion in your photo. At the bottom of the image size dialog you'll find the option for Re-sampling. You'll never want to use Nearest neighbor or Bilinear. I recommend using by Bicubic for most situations. If you're enlarging the image significantly, then I would suggest using Bicubic smoother. In theory, if you're reducing the size of the photo you could use Bicubic sharper. But I recommend applying sharpening as a separate stage of your output workflow. And finally the bicubic automatic will apply an automatic interpolation based on the resizing you're applying and the nature of the photo. I'll go ahead and leave the option set to Bicubic in this case which is what I use for most photos and then I can click OK and the image will be resized. At this point, I'm ready to Apply Sharpening to the image. If I decide that's necessary.
And then I can Save the final result if I'd like. And of course, send the image to the printer if I want to produce a print.
- Getting started with color management
- Early-workflow sharpening
- Output preparation
- When not to sharpen
- Advanced sharpening techniques
- Printing a photo