Rate This Tutorial!  2.712   Log in to rate this! 0 = Not Helpful up to 3 = Very Helpful. NR = No Rating. Questions? Feedback? Join the Discussion on this Tutorial

Workflow, Lossy Formats & Intro to Layers in Photoshop

by Bear_Music

Lossy Formats

We are basically talking about jpg here. It is also possible for TIFF to be "lossy", if you use the compression algorithms when you save to TIFF.

A "lossy" format is any format that uses image compression; each time you save the image, it runs a new compression routine and some of the original information is lost. Repeated saves while you are working on a compressed image will result in serious degradation of image quality, which will show up as clumpy "jpg artefacts" (blocky areas of flat tonality with no detail) in the image.

Here's your correct workflow:

1. Shoot picture at as low an ISO ("film speed") as possible and at as fine a resolution as possible, given the space you have available on your chip.
2. Download the images to a folder where you save original, unaltered images. It will have some sort of a number assigned to it by your camera software. Let's assume the number/filename is 123456.jpg.
3. Browse the folder and decide upon an image on which you will be working. Open that image the usual way.
4. IMMEDIATELY do a "save as" on that image, in the native Photoshop file format (.psd) in a separate folder that fits with your filing system. Rename the image as follows:
5. Let's assume the image is of your cat "Mehitabel": when you save-as, rename the image "Mehitabel_yawning_123456.psd".
6. You now have a working copy of your image exactly as shot, saved in a lossless format in which you can tinker to your heart's content. I usually save various versions of the image as well with further, descriptive names. ("Mehitabel_yawning_Misty_Duotone_123456.psd") The key here is to have some way of sorting all this, and to include the original number in all "working" saves so you can easily refer back to the original sometime in the future.

Procedure for submitting an image to DPC:

1. When you have a finished image you wish to upload to your portfolio in DPC or to a Challenge, you need to shrink it in size and compress it so it is less than 150kb.
2. First, working with the finished .psd file, go to layers/flatten image and flatten the image to a single layer.
3. Next, go to image/resize image and set the lpi value to 72 (this is the maximum resolution of a computer screen; anything more detailed is wasted)
4. Still in the resize dialogue box, go to the pixel values at the top and make your longest dimension 640 pixels (this is DPC's maximum size; if you submit anything larger they will downsize it)
5. Click OK
6. The image will now be much smaller on your screen. Go the "view" menu bar and select "actual pixels"; this will display the image in Photoshop at exactly the size DPC displays it.
7. Now look at this image critically and, if needed, apply some USM (unsharp mask, under "sharpen" in the filters list) to pop the image a little bit. Be careful not to overuse this filter; once you think it looks alright, blow up a smaller portion of the image with the magnifier and check to make sure you're not getting haloes where different values are juxtaposed; these will always result in lower scores.
8. Finally, save this image as a jpg file, and adjust the quality slider to as clsoe to 150 kb as you can get without exceeding it. You may use any name you choose for this file, no need to include the original image number here; the file is in the same folder as the "original" psd from which it was cloned, and that one references the image number.
9. Now close the image window. It will ask you if you want to save changes to your file; IT IS CRITICAL THAT YOU NOT DO THIS. Why? Because the file you are closing is the .psd file, and it has been flattened and resampled. If you save those changes, you will lose all your layers and most of your fine detail.

Photoshop Layers Primer

One of the greatest things about working on native PS images in Photoshop is that you can use unlimited numbers of "layers" to separate the different effects you are experimenting with. Here's your basic workflow, for an image that requires only minor tuneup, nothing extreme:

1. As soon as you save your original as a .psd file, go to layers/duplicate layer and create a new layer from background. Use the default layer name (background copy 1)
2. The original layer is now locked; you cannot work on it. It's your baseline.
3. Now create another layer from background and name it "out sharp"
4. On this layer do your "output sharpening"; whatever global sharpening the image seems to need.
5. Now go to layers/new adjustment layer and choose curves or levels, whichever you are more comfortable with, and adjust your tonalities and contrast. Curves works better, but it takes more practice.
6. Click OK when it looks good, then go to layers dialogue box and click on the "out sharp" layer to make it the active layer.
7. Go back to layers/new adjustment layer and choose hue/saturation. Mess with the master slider to see what happens at the extremes of the range, then return it to zero. Click the down arrow at top and, one by one, adjust saturation and brightness on each of the color channels individually. When it looks good click ok and then go back to the "out sharp" layer and make it active again.
8. Click on the eyeball beside any layer to hide it, and its effects. Do this on all layers but the original, to see how you started out.
9. On any adjustment layer, click on the little "picture" in the layers dialogue box to bring that set of adjustments back up again.
10. When it's as you like it, prepare for posting using above guidelines.

Now, layers can be used in much more sophisticated ways. If you right click within a layer box and choose "layer blending options" you get a very interesting dialogue box that allows you to alter the way the layer interacts with the ones beneath it. Try choosing different blend options and see how they effect the image; couple these effects with the layer transparency options to get some very subtle effects.

When you want to use filters, always clone a new layer off the topmost layer you have made from background, and then apply the filter to that layer. This way you can easily turn the effect on and off, or discard the layer entirely if it does not work.

If you are applying filters or effects onto selected portions of your image, do that also on a new layer cloned from background, and be sure to SAVE all your selections in select/save selection field. Then at any time you wish you can reload that selection, and even invert it, on another layer or the same layer.

Think of your layers as a series of clear, mylar overlays on the original image, on each of which you have "painted" some sort image effect. They can be "shuffled" by dragging them up or down the stack, to see them in a different order. They can be made more or less transparent, they can be blended in in different ways, and you can apply effects like drop shadows, glows, bevels, and so forth from the "layer blending options" dialogue box.

Always bear in mind that only certain sorts of layering are allowed in basic rules shallenges. Read the rules for more information.

Rate This Tutorial!  2.712   Log in to rate this! 0 = Not Helpful up to 3 = Very Helpful. NR = No Rating. Questions? Feedback? Join the Discussion on this Tutorial