Bitmaps

Bitmap graphics (also known as raster images) owe their name to the way in which the image is generated. Painting and Photo-edition programs create this kind of file assigning a colour to each point of the screen or pixel. This becomes clear when we see a very enlarged bitmap image. It is possible to be realize that the image is built around a series of points, each one with colour. The group of these points creates the illusion of an image of continuous tone on the normal scale of the screen. With the correct resolution, the quality of these images is good; a photography may seem as realistic as its equivalent in paper prints, although the resolution of the screen is sensibly inferior.

Number of colors and resolution.

The bitmap images can be defined in a variable number of colours. The more colours the image has, the greater quality it will have and, if there’s no compression, the bigger the resulting file will be. However, different systems from compression exist and this means that this rule is not so simple: some graphical formats like GIF, PNG or JPEG are very efficient and they can store a digital image with a modest size.

1 bit, 4 bits, 8 bits... many programs classify the images in relation to their number of colours like "1 bit colour", " 24 bit colour", et cetera. What does this mean? 1 bit colour implies the possibility of giving pixel of the screen a value chosen between two values (black and white.) In the 4 bit mode we have 16 colours; 8 bit colour allows 256; most present day graphic cards display colours in the 16 to 32 bits range (which allow to show simultaneously, in this last case, up to 16.7 million different colours.) The 32 bit colour is photorealistic, because, in fact, our eye is unable to differentiate colours with such sensitivity. What interests us here is knowing that certain modifications in the images is only possible in 24 or 32 bit colour; and the fact that converting from onw colour mode to another more limited always implies a loss of information in the picture; for this reason always it is advisable to keep a copy of the images in full colour.

The workspace in paint programs

The simplest program that you may try to begin with, is the Paint utility that comes included in Windows (in Programs|Accessories.) In spite of its limitations, this program shows perfectly which are the basic tools of all painting program. Even programs as the old Deluxe Paint for DOS (does somebody remember this beloved program from the Palaeolithic period of computers?) already had this same workspace and painting tools: a great space in white to paint, equivalent to the paper or canvas where we create the image, surrounded by a menu bar with the different options, a palette to choose colours and one or several toolboxes.

The toolbox, then, is more or less standardized. Here we’ll find the drawing, painting and selection tools which we’ll use most. The toolbox includes a brush, with variable shapes and sizes, tools to draw basic shapes (ellipses, rectangles, polygons...) simply by clicking and dragging —as we stretch the shape it will be drawn; tools to fill with gradients, to add text, draw lines and curves, select parts of the image...

The best programs share, in fact, these same tools, but remarkably improved. For example, there are more kinds of brushes where to choose, it is possible to graduate the transparency of the applied colour and many other effects. With the more sophisticatd programs you can even simulate watercolour brushtrokes, with more intense margins of the typical wash, and glazes or partially transparent washes on top of other parts of the image; charcoal or ink-like drawing, diverse textures...

Natural media programs.

In our blog about watercolour and other techniques there are several articles about this kind of rather sophisticated and powerful applications, notablemente sofisticadas. They offer a great set of brushes and textures to paint and different kinds of textured surfaces, much like real papers and canvases. In some cases the painting with natural media simulates traditional drawing and painting techniques very well: crayon, charcoal, ink, watercolour, oil... You see near this text an example created by the author, in one of these natural media programs (a pseudo-DeKooning, certainly.)

Although the image is created completely in the computer, the brushtrokes remind of of real painting in oil or acrylic, with the different textures from paint and different brushwork. This type of programs is meant for the artistic creation —directly in screen, or to alter images previously elaborated; their best complement is a graphic tablet device, a peripheral hardware which takes advantage of the qualities of programs like Painter, the unquestionable leader in this category of software, now distributed by Corel. The program has an infinity of options —and a quite a complex interface, even if it has been rationalised somehow in recent versions; it can be used in standalone mode or use it together with others. Painter allows surprising effects combining its brushes for realistic techniques and the paper textures, canvas and other surfaces. The program can be downloaded in tryout version from Corel.

Limitations of painting programs

Given the system of image generation of painting and photo-edition programs, there is one major limitation; but at the same time the mosaic-like pixel grid is what lets us create complex and realistic effects. Since a colour is assigned to each point to create the image, they have a fixed resolution; when changing the size it will inevitably lose quality. This means that when exlarging or reducing bitmaps we will end up with a more or less degraded copy of the original, but never with a quality as perfect as in the original with the right resolution. Any modification which we apply in this mosaic will affect neighbouring pixels of the edited area if we’re not careful. It is easy, then, to commit errors, and to rectify them can be time-consuming if not impossible. The programs may have an option to undo the last action (or a number of undo steps, but this implies memory usage.) On the other hand, whatever we do in the image is inlaid in its mosaic of pixels and there’s no way of distinguishing it from the rest of pixels if not for colour differences; most programs allow to select the specific areas that share a common colour, and then apply the wished changes.

Layers, channels and masks.

In order to solve some of the limitations in the painting programs they have been incorporating different systems to protect parts of the image while one works in other areas. These systems are the layers, paths, selections, channels and masks (with so many Channels and masks, it seems that we’re talking about the Venice Carnival...)

Layers

The layers allow to work with a procedure similar to overlaid acetate sheets, like in the classic production of cartoons. They are transparent until something is drawn on them. The palette which controls layers is similar in all the painting and drawing programs: it shows the existing layers, it allows to change their order, to add or to delete layers, to decide if they are visible or not (eye symbol), printable or not, editable or protected (pencil or brush symbol) and you can give each layer a descriptive name. These functions let the designer work with more comfort and security; layers are really an internal work mechanism of the program, because in the end, the result is still a bitmap, and if you save in most formats, in a single, flat layer.

Channels

The channels separate the different chromatic values of the image. For example, in CMYK mode there are four channels, for Cyan, Magenta, Yellow and blacK colour (see our article about colours.) Each channel can be modified separately, and custom, “artificial” channels can be created which preserve a part of the image of any modification of the original colours or lets process part of them. This procedure is very useful to colour comics and other illustrations. The reserve masks get this name from the traditional method of illustrators and photographers to protect some areas of an image while altering the rest: a mask in a pencil drawing can be a bit of scrap paper that covers part of the drawing to avoid that we make marks on it; painting in watercolours, a mask they can be prepared with anything that repels water with pigment, for example, wax or a special rubber. The meaning of masks in painting programs is often the opposite one because it may refer to a selection from the image that is altered while the rest is left aside, protected. In programs such as Photoshop a mask is very easy to create and edit. The program has an excellent tool for it, call quick mask, which essentially temporarily turns the brush or any other tool into a mask-drawing or painting implement. When finished, and getting back to normal painting and drawing mode, everything we have selected or painted upon will become a selected area. This system allows to special effects and adjustments whicha are simplicity itself (although Photoshop itself allegedly has a very steep learning curve.)

As an example, we’ll describe a very common example of mask use:

In step 1) we select a part of the image with a selection rectangle, in quick mask mode. The red colour indicates the selected area. We can make the selection with any drawing or selection tool, and we can even apply gradients or filters. In this case, gradients or filters will be applied to the selection, varying its shape and intensity.

Next (step 2) we apply a filter in the selected area (here, the Crystallize filter) and click in the icon of normal mode (in the bottom of the Photoshop toolbox.) This has the previous area selected, with some irregular contour, the product of the applied filter.

Finally, (step 3) in Selection, we choose Invert. We delete the rest with the suppress key and we’re done: we have an image delimited by an irregular contour.