DUANE M. PALYKA
Computer memory is an electronic surface—a thin sheet consisting of millions of small electronic elements called bits*, each with its own electronic field which can be measured plus or minus, one or zero, white or black. As these bits are set by a computer program, one can stand away from this surface and see clusters of bits, patterns of bits. These bit settings can form different gray levels, shapes, and visual images.
This is not totally imaginary discourse. A device called a frame-buffer was invented for just this purpose: to allow bits of a computer memory to reflect themselves on a visual device—the principal one being a color TV monitor. The computer memory is continuously scanned and displayed on the monitor; the way the bits are set reflects in the way the three color guns excite the phosphors on the tube surface. Programs can be written to change these bits dynamically and therefore give instantaneous color feedback on the TV monitor.
A computer Painting, 1975. Duane M. Palyka
When a computer lends itself so easily to visual artistic exploration, why should it be restricted to engineering uses? Why should any separation be made between this and other artistic devices? The computer artist need not know about his medium below the plastic level any more than the painter has to know about the chemistry of his paints. The plastic level of the computer art medium, however, includes programming; but programming is just as plastic a medium in its own right as paint and brush, and can be thought of in a direct visual sense when coupled with an accommodating hardware device like a framebuffer.
Both the creation of paintings and the creation of computer programs are the creation of objects—objects constructed out of ideas, concepts, and craftsmanship. The aim is a finished work with strong structure, patterns, imagery, and textures. A painting can be considered 'clean' if devoid of meaningless shapes and forms. A program can be considered 'clean' when the code has no meaningless instructions. Good structure in a program can bring as much esthetic satisfaction as good structure in a painting. The code and structure of a program reflects the personality of the person generating it. Both programming and painting are problem-solving processes to which each person has his own approach.
Actually, in painting more emphasis may be placed upon the subconscious as the source of images and ideas. Not to exclude the fact that leaps of the imagination are needed to generate creative ideas in both media, programming is basically a logical process which utilizes an individual's conscious mind. Because of this difference, painting can reach a more mystical level of awareness—a depth of consciousness which we can experience but not explain. This does not mean that works of art are not logical, far from it. It is because the most creative works of art have such a logical basis that the computer fits in so well with the creative artistic process.
Take as an example Lewis Carroll's Alice in Wonderland. Few people will deny this to be a highly-creative work with imagery which reaches into unknown areas of our psyche, but this work is based heavily on strong mathematical logic taken to absurd levels. Lewis Carroll had a background in mathematics. On the art level, Cezanne's paintings have provided art historians with a source of analytical material for years, but so have Mondrian, Da Vinci, and many other painters.
An emphasis on logically conscious thinking is required with this medium to a certain extent since the computer works this way, but the computer artist's thought processes need not rest solely on this level. Programming is a step-by-step process where the programmer must understand what is happening at every step, but the images one makes using this process need not be totally 'understood.' The images themselves do not have a strict logical tie to the images which follow. The creation of images must have some logical visual flow from one to another to add order to the composition but, unlike the programming which generates it, the system of images will not collapse from the slightest deviation from logical conscious order. In fact, the deviations are what give emotion and tension to the piece. As in painting, with the computer art medium subconscious 'not-understood' constructions can co-exist with conscious 'well-understood' constructions. It is this aspect which makes art reflect everyday human thought processes and hence reflect life. Taking a computer scientist or a programmer who thinks continually on this linearly conscious level, it is interesting to see what works of art he prefers and what his view of representing the real world is. His thinking demands that he view the real world on a solely conscious 'understood' level. He makes strict comparisons between the images he generates and the world in front of him. In this attempt to render the world in this strictest sense he may even find it disturbing to look through two eyes when rendering a flat image because that means he has to choose which view to select, the one from the right eye or the one from the left.
It is interesting to note that as far as artists go, besides renaissance painters, Escher is one of the favorites. This is because Escher renders a reality on a very conscious level with deviations made on a strictly logical basis to keep it interesting. There is little leakage from the unconscious into Escher's images.
What this thought process lacks is the realization that leakage from the subconscious can make the images even more real in spite of the fact that they now deviate from the strictly-observable real world. This deviation personalizes the images by allowing the viewer's imagination to get in sync with the artist's on a 'non-understood' subconscious level.
Conversely, the artist could benefit from getting involved in linear conscious thought processes since it will give him more power in 'cleaning-up' his images and structuring his pieces. Through these processes he will have the tools to analyse his processes as well as the processes of others and help understand himself a little better. This experience can also assist in removing some of the irrational fears laymen have about computers.
Programming itself is filled with interesting visual imagery which, in turn, makes one think with new visual insight. Take, for instance, the idea of the 'bubble sort.' The bubble sort is a programming device for organizing a list of disorganized numbers into a list of organized numbers which increases (or decreases) sequentially. It does this by forcing the smallest numbers to bubble up to the top by means of a one-on-one comparison with the number above it. If it is smaller than the one above it, they change positions; if it is greater than or equal to, no change occurs. Imagining this process visually, it resembles the motion of bubbles. If one thinks in Paul Klee's visual terms, one can 'see' an implementation of this process on canvas. For the socially-conscious artist this concept can be expanded into the 'bubble-sort principle' to illustrate individuals rising in the business world according to their moral I.Q.
What is the overhead in learning the computer art medium so an artist can start thinking in terms of forms, shapes, and colors through numbers and programs? With paint, the first stroke one makes yields visual results. With programming, a person may expend many hours learning a programming language before ever really seeing a visual image produced with it. Unless the artist can get interested in the programming medium for its own sake, he can get discouraged and put an end to this business before ever getting started.
On the other hand, one can have the computer simulate a traditional art medium with which the artist is familiar and leave it to the artist to make the transition from the medium he knows to the new medium on his own terms. The computer is very good at simulation and, coupled with a frame-buffer, this particular simulation is simple to implement. Using an electronic pen and tablet for input, the artist can be provided with a medium similar to acrylic painting. He can watch on a TV monitor a flow of color reflecting his hand and pen movements on the tablet. In fact, he can even select the brush sizes he wishes to use.
An advantage that this medium has over acrylic paint is that the artist can change the medium to suit his own personal artistic needs through programming. As the artist learns to program he can see the development and change of his medium and, hence, of his images.
What kind of artist can get involved with the computer art medium? At this stage of the game, it takes an artist who can cope with dualities since he has to straddle two fields. He must have a flexible enough identity to accept the interflow of ideas from one discipline to another. He must be able to pursue what is interesting in spite of the labels that have been attached to it. He must be interested in developing both hemispheres of his brain—the half dealing with algebraic logic as well as the half dealing with esthetics. An artist who comes to mind who seems to have the flexibility to deal creatively with this medium is Marcel Duchamp. An artist capable in playing chess can become capable in programming computers. An artist involved in Dada or Zen can cope with the dualities. An artist whose ideas are conceptual as well as sensual can fit in well here.
The computer art medium can also help with the artistic risk-taking process. When an artist makes radical changes to his painting based on new thoughts and ideas, he risks destroying the painting to produce a highly-creative work. Giacometti and Matisse had the guts to wipe out hours of work to start afresh on the same painting. This can be a difficult thing for an artist to do. Giacometti developed a process of taking a painting to completion, wiping it clean, painting it again, wiping it again, etc. until he reached a state where he liked it or just gave up. A problem occurs when a previous stage of the painting had the best result but there is no way of retrieving it. Picasso found the desire to have two paintings developed to a certain identical state then each taken to completion in different directions. These tasks can be made easier with the computer art medium.
A digital picture, by its very nature, can be saved, transferred, and restored without any loss of information. To make a copy of a digitized image is to make another original. A bit is a bit and is transferred exactly. This means that an artist can save a state of his image, explore new directions in changing that image—perhaps a risky new direction that he was afraid to approach before, and retrieve his saved image to continue if he did not like the results of that exploration. This means that anyone with a compatible computer system can have the artist's work to do with as he wishes. In fact, the rip-off problems could cause some legal difficulties in the future.
The computer is also very good at repetitive tasks. A hypothetical program which would utilize this feature to the artist's advantage is one which would generate an infinite series of pictures, each created according to programmed rules of design coupled with algorithms or input devices which generate pseudo-random numbers. With a program like this one could actually satisfy the needs of the art world without boring oneself to death—creating enough similar works so the artist can easily be identified with a certain style.
Incidentally, because of the mathematical base of the medium, works produced with it could force the pseudo-mathematical approach to painting to assert itself in new and less ambiguous ways or die a natural death. An example of this approach is where the artist points to a triangle on his canvas and tries to imply a mystical mathematical base to the work by stating that one point of his triangle is 'A,' another is 'B,' etc. without reaching any definitive depth on the matter. To be ambiguous is a natural part of the art process for reasons explained above. No one is really expected to explain a work of art. However, to be ambiguous when making a correlation between an art object and mathematics is placing oneself on dangerous grounds.
Every case mentioned above has the artist working alone to create his images. Why not have the artist work with a programmer to produce his images? The artist-programmer relationship can be a learning process for both individuals; each finds out more about how the other thinks. However, if we view the situation according to the quality of art work produced, and if we make the assumptions that the programmer is not an artist and that the artist is not a programmer, the following two problems occur.
First, the artist uses the programmer as a tool to produce his art. In the process of providing the artist with a higher level between him and the machine, the programmer becomes an extension of the machine in the sense that his work is also included in the area of the black box. This is fine for having the computer simulate a calculator, but not so good for trying to produce creative work on a more personal level.
A computer painting, September, 1975.
Second, since the artist's ideas about what the computer can do are sketchy and since his own ideas about what he wants the computer to do are not defined well enough in a common language that both understand, a lack of communication occurs which prevents any real depth of image-making to occur. The scientific process and the artistic process differ on one basic point: the artist allows feedback from the process to dictate changes in his goals whereas the scientist will interpret his feedback as defining a state of his progress towards his predefined goals. Consequently, at the start of the process, the artist has a sketchy, rather vague, set of goals whereas the scientist has a well-defined set. An artist starts out with a tentative plan and then feels around until he finds something he likes. The end result may not even resemble what he started out to achieve. There are, of course, grey levels in this black-and-white description—especially in the area of the more creative programmers and artists; but, then again, it is the creative individual who can overcome rigid processes and change his thinking to fit the situation.
When our stereotyped programmer gets an error or unexpected result he straightens out his program to give the 'right' result, but when our stereotyped artist gets an error or unexpected result he evaluates it from an esthetic viewpoint and may incorporate it into his work or allow it to dictate the direction of the work if it 'feels good.' The former philosophy is more authoritarian—greater emphasis is placed on rules—generating and obeying them. The artist likes to break rules if he can. He wants to disprove the seemingly absolute.
The difference in methodology, incidentally, makes it difficult for the artist to write a definite proposal to get funds for the creation of a computer art environment. It is difficult for him to outline goals, needs, and step-by-step procedures in a well-defined scientific manner when his work is not based on these methods.
Now a major problem with achieving access to a flexible computer art medium is the initial cost of setting it up: $50,000 for a computer, $80,000 for a frame-buffer, $5,000 for a tablet, and $5,000 for a color TV monitor. This means that none but the wealthiest artists can afford his own electronic 'paints-and-brushes.' However, as the cost of everything else is on its way up, the cost of computer equipment is plummetting downward. In five years the cost of this equipment should be within the range of an art department budget.
Let us assume that the artist has secured use of this hardware, has developed some software, and has generated some images on the TV monitor. What are the possibilities of him freeing his images from the electronic cage and putting them into some form which he can hold in his hands? Right now all he can do is record the image on photographic paper or film. Putting it on video tape—another possibility—is just the transferral of his images from one electronic cage to another. Both methods involve transferring to a medium other than the one with which the artist is working. This may be more palatable if one makes the analogy with the etching medium—where the artist works on a metal plate and prints his images onto paper. Or one could take solace in the realization that most art is seen through art magazines where the art magazine reader has left to his imagination the sizes, true colors, and the textures of paintings. Actually, a numerical bridge can be constructed in the transferral of images from the TV monitor to film so that the light intensities are adjusted to make the film image match as closely as possible the TV image.
If the artist does not like the above input-output devices, the versatility of this medium is its ability to handle quite varying devices for input or output. All that is required is that each device have a wire or two containing electrical current which varies within a certain prescribed range. The rest is within the imagination of the individual designing the device.
If one considers the question as to which medium is preferable—the computer or paint-and-canvas, the answer ultimately lies with the individual. Exciting works or boring works can be produced in either medium.
Salt Lake City, Utah
Gilot, Francoise, and Lake, Carlton. Life With Picasso. New York, N.Y.: The New American Library, Inc., 1965.
Lebel, Robert. Marcel Duchamp. New York, N.Y.: Grossman Publishers, 1959.
* A bit is the smallest numeric unit in computing. A bit can be either 0 or 1 in the binary number system. Three bits set 1 side-by-side make 111 (binary) which is the number 7 (octal or decimal).
Return to Table of Contents | Previous Section | Next Section