This chapter gives an overview of how layers work in &chalk;. Extensive use of &chalk; will almost require you to have some knowledge of layers. Using layers, you can work on one part of the image without touching the rest of it, and most effects are best applied on a layer, instead of on the whole image. Of course, if you do want to apply an effect to an entire image, &chalk; does offer you that possibility, and there is nothing against it. The idea behind layers is quite simple. As the name suggests, layers lie on top of each other, and together form the layer stack. The final resulting image is that what you see when looking through the stack from top to bottom. This means that usually the upper layers of your image will have more or less transparency, since you cannot look through a layer which has no transparency. (&chalk; works with opaqueness instead of transparency. A layer that is 100 percent opaque is 0 percent transparent, and vice versa.) A layer higher in the stack gets applied later than one lower in the stack. For example, if your image contains four layers, numbered from 1 (lowest) to 4 (highest), the effect that layer number 4 adds to the image, is applied to the result from applying layers 1 through 3. Every image you edit in &chalk; contains layers. When you create a new image, the layer box (usually shown at the bottom right of your screen, see this section) will contain one layer. The painting and editing you do is then applied to that layer. Once you add more layers, you can choose on which part of the image you want to work, by selecting the respective layer. All further painting is then applied to that layer, until you select another one. Layers are also an excellent way to check whether adding certain effects (or applying certain image modifications) come out right. Add a layer which contains what you want to try out, and show or hide it with the eye icon in the layer box. You can especially profit from this method if you have multiple effects to check out: show and hide them in any combination, and decide which you like best. And since you can move the layers around, you can also experiment with the order in which the effects are applied. See the Selections and layers tutorial for a small hands-on introduction. The layer box is the instrument you will use most to work with layers. It gives an overview of the layers that are present in your image, and using it you can manage layers by adding, removing, reordering or modifying them. The layer box consists of three parts. The middle part gives an overview of the layers in the image. At the top, you can set some properties for the current layer. At the bottom, a couple of layer management options can be found. The next sections describe these three parts in more detail. This part shows you which layers are present in your image. In a tree-like structure, the layer group hierarchy is shown: layers that are contained within a layer group are displayed a bit to the right to indicate their belonging to that group. For each layer, a thumbnail preview and its name are shown. The layer name is preceded by a folder icon if it is a group layer. Furthermore, two indicators are present: the eye icon shows whether the layer is currently visible (an open eye indicates that the layer is visible, a closed eye indicates that it is not), and the lock icon shows whether the layer is locked. No changes can be made to a locked layer. When you click on a layer's eye icon, its visibility is switched from on to off or vice versa. Clicking on the lock icon enables or disables editing of that layer. You can click on the name of the current layer to rename it. Note that to rename a layer, it has to be the current one. You do not need to activate a layer in order to make it (in)visible or (un)locked via the eye and lock icons, respectively: these work directly. Doubleclick on a layer entry in the list to open the Layer Properties dialog. This dialog shows a layer's colorspace and profile. You can also change its name, opacity and composite mode here. The top of the layer box contains two controls for setting properties of the currently selected layer. The list box at the left allows you to quickly set the layer's composite mode. The spin field and slider at the right can be used to change the layer's opacity. At the bottom of the layer box, there are five buttons. From left to right, these are as follows. The New Layer icon brings up a submenu from which you can choose which type of layer you want to add. This menu can also be opened by clicking with the &RMB; on the layer box. The Move Layer Down and Move Layer Up buttons move the current layer one level down and up, respectively, within the current layer group. If the layer is already the last or first within the layer group, trying to move it further will move it out of the layer group. The Layer Properties button opens the Layer Properties dialog, just as when you would have doubleclicked on the layer. The Delete Layer button deletes the current layer. Because layers are quite important when extensively using &chalk;, you can perform a lot of operations on them. These are all available via the Layer menu. Some of the possibilities: Add, remove, and duplicate layers; Create and edit layer masks; Flip, rotate, scale and shear layers; Convert layers between colorspaces; Save layers as images; View layer histograms. Adjustment layers are layers that consist of a filter and an optional selection. The filter effect is applied to the composite image of all layers under the adjustment layer in the current layer group. The big thing is, adjustment layers apply these effects non-destructively. The original image data is not modified. Almost all &chalk; filters are suitable for use in adjustment layers -- even filters that would downgrade the image quality. For instance, the raindrops filter converts to 8-bit RGB before working its magic. If you would try to use this filter directly on a 16-bit L*a*b* layer, &chalk; would warn you about the conversion to RGB and back again this filter would cause. Not so with adjustment layers: the original data isn't touched, so applying the filter is safe. What about the colorspace of an adjustment layer then? In order to examine this issue, you need to know what happens when &chalk; renders an adjustment layer. If the currently active layer has an active selection, then that selection will be copied and used as a mask for the adjustment layer. If there is no active selection, then there will be no mask and the adjustment will apply to the entire extent of the layers under the adjustment layer in the current group. There is no way of adding a mask to an existing adjustment layer. If there is a mask in the adjustment layer, you can edit the mask using the ordinary painting tools and painting operations. &chalk; composites the layers bottom to top, within each layer group. The aggregate -- or the projection as it is also called -- is then filtered by the adjustment layer. If there are layers on top of the adjustment layer, those are composited onto the projection. &chalk; converts all layer data before compositing, so if the bottom-most layer in an image is grayscale, all layers are converted to grayscale before compositing -- and that means that the adjustment layer projection will be grayscale, too. With this knowledge you'll understand why &chalk; can often offer better performance working with layers on top of an adjustment layer which is on top of a complex layer structure: &chalk; uses the projection and doesn't even look anymore at the layers under the adjustment layer. Unless, of course, you change one of them. Layers can be composited in various ways, each yielding a different effect. This section describes the available compositing modes. Each description is accompanied by an example: on top of an original image (see below), a rainbow gradient is added. The original image The original image The original image The Normal mode does nothing special. It adds the layer to the image, and if no other special effects like opacity are changed, the underlying layers will only be visible at places where the new layer is itself transparent. The gradient applied with the Normal compositing mode The gradient applied with the Normal compositing mode The gradient applied with the Normal compositing mode The Multiply mode blends the two layers so that the bottom layer gets colorized by the new layer. The resulting image is generally quite dark. The gradient applied with the Multiply compositing mode The gradient applied with the Multiply compositing mode The gradient applied with the Multiply compositing mode The Burn, Dodge, Divide and Screen modes all add an extra burning effect by following contours instead of using straight lines. In addition, Burn and Divide use the inverted colors instead of the actual colors of the composited layer. The gradient applied with the Burn compositing mode The gradient applied with the Burn compositing mode The gradient applied with the Burn compositing mode The gradient applied with the Dodge compositing mode The gradient applied with the Dodge compositing mode The gradient applied with the Dodge compositing mode The gradient applied with the Divide compositing mode The gradient applied with the Divide compositing mode The gradient applied with the Divide compositing mode The gradient applied with the Screen compositing mode The gradient applied with the Screen compositing mode The gradient applied with the Screen compositing mode Like Multiply, the Overlay mode colorizes the underlying layer. The resulting image is about as light as the original layer. The gradient applied with the Overlay compositing mode The gradient applied with the Overlay compositing mode The gradient applied with the Overlay compositing mode The Darken mode darkens the underlying layer while colorizing it to match the colors in the composited layer. The gradient applied with the Darken compositing mode The gradient applied with the Darken compositing mode The gradient applied with the Darken compositing mode The Lighten mode lightens the underlying layer while colorizing it to match the colors in the composited layer. The gradient applied with the Lighten compositing mode The gradient applied with the Lighten compositing mode The gradient applied with the Lighten compositing mode The Hue, Saturation and Value modes respectively apply the hue, saturation and value components of the composited layer to the underlying layer. The gradient applied with the Hue compositing mode The gradient applied with the Hue compositing mode The gradient applied with the Hue compositing mode The gradient applied with the Saturation compositing mode The gradient applied with the Saturation compositing mode The gradient applied with the Saturation compositing mode The gradient applied with the Value compositing mode The gradient applied with the Value compositing mode The gradient applied with the Value compositing mode The Color mode colorizes the underlying layer, yielding very strong colors. The gradient applied with the Color compositing mode The gradient applied with the Color compositing mode The gradient applied with the Color compositing mode Basically, a layer mask is a mask that you place on your paint layer. This will literally mask areas of the layer, so that the content underneath shows through. You can paint on it with greyscale colors: the more black the color, the less the layer under it will shine through, the more white, the less the layer under it will be shown. So complete white will let nothing through, complete black will let everything through. Basically, it is a bit like selecting a piece of your image, and then cutting it, so that the selected bits go away. So what is the use for a mask here? The big advantage is that it is non-destructive: if you decide that you masked out the wrong part of your layer, you can easily remove the mask and start anew, something a lot harder (not to say near impossible, especially in between sessions) with regular selection-cutting. So, how to create a mask? There are 2 ways: Start from scratch. LayerMaskCreate Mask. The mask starts with everything being retained, that is, a complete white mask. Basically you will not see any changes as long as you do not paint on it. Start from the current selection. LayerMaskMask From Selection. The selectedness will be converted to whiteness. This means that fully selected area will be visible, fully unselected areas will be invisible, and the rest will be partially visible, depending on how much the area was selected. First, make sure you are editing the mask, not the layer, by making sure LayerMaskEdit Mask is checked. (This is checked by default.) Then you can paint on the layer just like before, only now you are painting on the mask, instead of on the layer itself. To stop painting on the mask, you can uncheck the Edit Mask checkbox. There's also the option to show the mask, through checking LayerMaskShow Mask. (This is not checked by default). This option will render the entire layer as a visual representation of the mask in greyscale, instead of the actual layer. This can be handy to see where your mask is, but it might be not as handy when you want to edit it, since you cannot look at the actual layer. Other actions: you can also remove the mask if you are not satisfied with it, and want to start over again, or just want to remove it, with LayerMaskRemove Mask. You can also apply the mask, meaning that the mask will be made permanently. This means that the mask is removed, but that its effect of transparency will be committed to the layer.