In lieu of an abstract, here is a brief excerpt of the content:

Leonardo, Vol. 11, pp. 217-218. 0Pergamon Press Ltd. 1978. Printed in Great Britain. , + OENERATOR + PRODUCER -t> TESTER b 0024--094X/78/070 1-0217902.00/0 COMPUTER ART: AN ATTEMPT TO MAKING PROCESS AUTOMATE ASPECTS OF THE PICTURESteve Scrivener* 1. Introduction For sometimeI have beenconcerned with various aspects of the picture-making process. This stems, on the one hand, from the influence of a general concern in art with art-making processes(forexample,a significantfeatureof Surrealism is the emphasis placed on the process of making pictures) and, on the other hand, from problems arising out of my own work in the visual arts. I have been unwilling to accept the significance attributed to artistic intuition. To my mind, too much of the artistic process isdescribedas intuitive and, therefore, not analysed or questioned. I believe that much of the mental phase of the picture-making process credited to intuition may be brought within the province of reason. In this respect, the reader’s attention is drawn to Ref. 1. 2. Picture generation The mental processesthat generate imagesare not easy to describe. Like a number of artists (but perhaps for different reasons), I have adopted an approach that aims to separatethe mental process.Essentiallythis means that the stepsnecessaryfor the generation of a picture are prespecified in terms of some description or program. A picture, or a set of pictures, can be specifiedby a program that is then executed by the artist, another person or a machine such as a digital computer. The uses of computers in visual art aredescribed in numerous articles in Leonardoand are reviewed in the articles by Franke [2] and by Thompson [3]. The conventional picture-making process may be characterised as a feedback system that comprises a generator. aproducer and a tester (Fig. I.) The generator specifies the image, for example the image of a square, which is passed to the producer who, obeying the instructions, draws the image. The tester decideswhether the image is acceptable by testing it against some criteria and reports back to the generator. This scheme is not meant to represent a complete model of the picturemaking process, but it does help to illustrate my discussion. 3. The usefulness of isolating the image generator My interest has been centered on isolating the.image generator. If decisions relevant to the outcome of a picture are made by the generator, in principal some properties of the picture may be understood and explained in terms of the characteristics of the generator. Isolating the image generator permits the following: (1) *Artist, Dept. of Mathematics, Statistics and Computing, Leicester Polytechnic, Leicester, England. (Received 11 Oct. 1977) The generator isobjectifiedandcan, therefore, be studied. (2) Relationships between the generator and the pictorial properties of the images it produces may be determined. (3) Knowledgederived from a study of the generator may be used to predict the types of images that will be produced. (4) Decisionsmay be understood as changes in the parameter of the generator. 4. Incremental automation Since I have been concerned mainly with automating the generator in the form of the computer programs, it was necessary to find a way to integrate the rest of the picture-making process: the producer (a computer) and, more significantly,the tester (myself).This I have done by writing computer programs that allow the entire picturemaking process to take place interactively at the computer. During the generator and tester steps of the process, some decisions are taken automatically by the computer whilst others I introduce at the appropriate time. Figure 2 is an example of a drawing produced by a computer program named SIXTEEN. (I have used a Honeywell 516 digital computer and a Tetronix 4002-A storage screen and CIL No. 314 plotter.) The drawing comprisesa20 x 20square matrix of squares, or cells,that Fig. 1. 100 100 0 . 0 100 0 . 0 Mo 0 . 0 ow aa 100 100 aa 100 100 wo ow Fig. 2. Computer drawing made using computer program SIXTEEN, 20 x 20 cm, 1976. 217 have been filled with graphic elements. In all there are 16 graphicelements, includingthe null element (shownat the right of Fig. 2).The distribution of graphicelementsin the matrix is controlled by a prohahilisfic...


Additional Information

Print ISSN
pp. 217-218
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.