Let's discuss strictly technical details of icon design
1. Lines, slopes
One of the vital elements of the image is the outline of the object, i.e. solid linear marking, which separates the object from the background. When designing normal-sized images, the designer rarely thinks about highlighting of objects with additional lines. This is unnecessary because of the size: even non-contrasted objects do not blur into a one. Pixel graphics is other story. Two variants are available: either foreground and background colors have to be from different sides of the color chart, or the foreground must be divided from the background by visible outlines and shadows. I would like to discuss lines in more detail.
If we think again to the large scale graphics, we can note that in order to define contours, we can use any (including the most complicated) angles, Bezier curves and edges. Anyway, the line will look perfect due to anti-aliasing. When decreasing the graphic dimensions to the small icon size, situation changes dramatically. When every particular pixel is equally valuable and can ruin the overall appearance of the icon, anti-aliasing is just unreasonable. It comes that you should think about the possible line angles.
The slope you choose for the line, specifies the step of this line. Because each line consists of basic elements, the union of which defines its neatness and visual appeal.
For instance, a 18-degree slope contains many small 3-pixel lines connected using a 1-pixel downshift:
This is how nearly each line looks like. Primary element followed by joining followed by second primary element. Unfortunately, not every angle makes the line look attractive and not annoying.
For example, look at 25 and 20-degree slopes:
It can be easily known which line is more attractive. The 25-degree angle produses a line consisting of equal 2-pixel basic elements. The 20-degree slope has a basic element containing of three lines: 3-pixel, 3-pixel and 2-pixel. The slope is more complex, thus, if not supported by additional visual effects, it brings a feeling of messiness: the primary elements have contrast which is too strong to hide from the human eye.
Here we can see the first conclusion. The most "appropriate" slopes have to make as simple basic elements as they can. therefore, the ideal slopes are 0 and 90 degrees. In them, 1-pixel primary elements are joined without downshift and form smooth lines. Less perfect are slopes which form primary elements of 1, 2, 3 pixel with 1-pixel downshift. Of course, the lines can not be even, but the human sight will process the image and provide to the user what you meant to show to him. Also, the mentioned angles can be considered acceptable (but at a stretch): angles with basic elements consisting of different lines shifted by more than 1 pixel.
It is the time to move to another problem. In the last paragraph, I purposely defined angles with basic elements of 1, 2 and 3 pixels as correct. Such tiny lines, especially when a great amount of them is joined, look like solid. But what do we see if we increase the primary elements to 10 or 20 pixels?
Here is an example:
It can be obviously noticed that the line doesn't have integrity. It is not a continuous line anymore; it is a set of several lines, located near one another. Hardly any designer wants its creation to look that way. So, we come to the second conclusion: if you use minimal angles, which make long primary elements, it has to be justified and employed with maximum caution.
And, finally, the third aspect I have to tell about lines. I did write that the primary element should be as plain as it can.
For example, a 25-degree slope can be produced in the two (sure more of them exists) following ways:
In the first picture the primary element is a 2-pixel line. In the second picture, it consists of three lines: 3-pixel, 2-pixel and 1-pixel. The same slope but the image is different. The complex primary element made the line untidy.
This examples can be produced for about any angle, so in your works, try to simplify everything.
It can be stated without an exaggeration that the color is the main aspect of your work; improper or poorly matched colors can ruin even the best idea. What can we write about the color in miniature graphics?
First, there is a clear rule: if the image looks poor when converted to monochrome, then the colors that were selected are wrong. This rule is valid not exclusively for icons but for the entire graphic design.
Second, you will be unable to apply safe RGB. Sure, you may use internet colors only, but this limits you greatly. Gradients, shadows, dithering and many other tricks will not be available to you.
Third, try to enlarge the portion of "plain" color. The more consistently colored areas without blurring and jumps your picture has, the more neat it looks. Icons are too small and the over usage of special filters makes the picture look dirty.
Fourth, gradients. This is a great instrument which can improve the look of any composition. Other than visuals, gradients are a perfect way to abolish the "broken line" look (lines with too long primary elements). But be careful when using them since too much gradients can ruin the plain color, and be not suitable for the gif palette.
Fifth, shadows and highlights. All aspects of using shades and flares in icon making are completely the same to the aspects of the general graphic design. The only thing I would like to note: draw everything yourself. Don't apply effects, make all shadows and flares in a separate layer and then edit the opacity. If applying filters, you almost never can predict what the result will be. It is unfavorable when you don't control the process of creating of a project. It is worse if you can't control it when drawing icons.
And, lastly, sixth. Tinting is the key aspect of icon creation, which greatly results the esthetic look of the image. If the blue human eye in the image is sized 1 pixel, increase the color saturation. For instance, instead of 0, 131, 159 make 71, 195, 242. It is indifferent that the human eye can't have this color; in the first example it will look like a gray dot, but in the second case it will really be a green eye.
Thoroughly control the picture, do not allow the optical illusions to spoil your work. It is OK if something is not perfect from the technical point of view, but the whole picture must be flawless.
As an example, I drew the packet of juice standing before me. The image has flat color, gradients with broken lines, shades and highlights and tints.
If I talked about something other than icon creation, this chapter wouldn't fit into a whole book, but the size of compositions ;limits fonts too.
The resolution of symbols is the main aspect as opposed to their fineness. Nearly in any case, only if not the letters are the primary part of the image, the font size has to be reduced to the largest extent possible.
In principle, almost any design company has its unique font with small letters. Such a font can be made in several hours. You can browse the web and make your own collection of very interesting fonts. First of all, I advice you to get the DS Pixel and Seventen 7Vedi webpage and the whole series of fonts from the Lakmus Lab website.
If you don't need a new font, little typesizes of Arial and Verdana may do. As a last option, you can create the necessary characters by hand.
There are quite a few rules there. First, letters cannot be less than 5 pixels high or less than 3 pixels wide.
Second, letters cannot use anti-aliasing. It transforms the font into an unrecognizable mixture of pixels.
However, there are no rules without exceptions and the 5-point Arial appears rather understandable.