NewsProTeam | Jan 15, 2021 | 0
Is Generative Art for You?. About the ups and downs of creating… | by Eugen Lindwurm | Jan, 2021
Visual generative art is an intriguing niche of the visual arts that makes use of an autonomous system, typically defined through code, to generate pieces of art. Although it has its roots at least in the 1950s, generative art gained traction with the technological advances of the last decades. Among hobby artists, generative art is perhaps of particular interest to creative-minded hackers who want to express themselves artistically in a structured way or just want to create pleasant imagery, using the skills they excel at.
Knowing some programming myself, I finally started dabbling in generative art during the 2020 lockdown. It is undoubtedly fun to mix the necessary conceptual phase with experimentation through code and iterative improvement. Nonetheless, there are some aspects of generative art that can be off-putting.
In this short article, I will go through the differences in the process of creating generative art to more traditional art. I hope it can serve as guidance to other creatives contemplating getting into generative art. Some of these features will be more appealing to some artists rather than others. Every tool excels at doing certain things and is not as good at others. This “no free lunch” type of effect of course also applies to art artistic media. I believe there are two criteria for choosing an artistic medium:
- Does it allow you to create what you want to create? The implications for this question are the main part of this article, but of course what is can be done in any medium also depends on the artist.
- Does it fun for you to use? I will cover my own perspective, but perhaps you can make an inference to your own.
So let’s get right into the ups and downs of generative art!
Coding Skills Needed
For me and perhaps most readers of this article this is not a problem. If you don’t know how to code you can either learn the basics or try to follow the processing tutorials, but be warned that you’ll be off to a slow start in either case.
One of the big cons for me is that you can’t really sketch out your piece before getting into the nitty-gritty of actually programming. Of course, you can jot down some ideas on paper and play around with them, but due to the amount of complexity and randomness you’re probably going to pack into your artwork, it will probably not at all resemble the final piece.
With generative art there is always some upfront investment until you can see something. It will get smaller the better you know your tools but it will never be as little as a quick pencil sketch, which is definitely a con.
More Room for Exploration Later On
The exploration ability missing in the early stages of conception and sketching comes in later in the generative process. Since your artwork is defined by code and you can change and rerun it at any time, it becomes very easy to make changes — even large ones — later on, and to try multiple versions. You want to see whether your abstract trees would look better with purple rhombs instead of green ovals as the basis of their foliage? Nothing easier than that! This absolute freedom to mix and match and try changes without permanently erasing previous versions is something I appreciate about the process of defining an artwork in code.
Confined to the Abstract
Since we’re talking about trees already: generative art is usually confined to abstract art. You can try to get around this by using generative machine learning models, but without them, generative art is likely to fail at representing the real world, simply because the real world is too complex and varied to be captured by a reasonable amount of code. (Also a reason why image recognition didn’t work well before deep learning.)
Many artists are happy with the abstract and if you are among them, generative art might be just for you. Concrete objects, characters, storytelling, and perhaps imitating aspects of classical art are harder to achieve in the generative medium, which can make it a little less appealing to some.
Develop and Reuse Tools
This one is a mixed bag. It can be fun and challenging to develop your own tools (shaders, geometric objects, subdivision algorithms, …) and people who code for fun would appreciate developing tools just for the heck of it. Also, it is really rewarding to reuse the tools you created in subsequent pieces and it definitely enhances your process. (And perhaps you can even build your own unique style by means of a toolbox?)
On the downside, the need to develop or find tools that go beyond the basics can be a frustrating experience if you just want to get on with creating what you had in mind or experiment with some specific way of shading or drawing.
Complexity, Randomness, and Surprise
This is where generative art shines. Through programming, we can in a fraction of a second create patterns of a complexity that would be unthinkable to create by hand. Just look at fractals! With the powerful tool of random number generators, we can create virtually infinite variations of such patterns and also add some noise to textures and geometries that help make the piece look more ‘natural’, if we so desire.
What’s best about it is that you can never be certain about the outcome, and more often than not, you will be surprised. (On top of the surprise arising from randomness, your own coding mistakes can produce interesting results.) A downside of using randomness — and for many generative artists it is an irreplaceable staple — is the loss of total control. You won’t be able to precisely define your artwork and also enjoy the advantages of randomness.
If you want to create intricately complex pieces and like going through an arbitrary number of variations to find the one that really stands out, then wait no more!
This is perhaps the biggest downturn for me: there is always a delay between your work and the result. This aspect is not a limitation on what you can create, but it influences a lot how it feels to create. When you paint a picture using a more traditional medium you get immediate feedback. As soon as your pencil touches the paper, you see what is happening, you know where you’re going, and that makes the process in itself enjoyable and sometimes almost meditative. If you program your artwork you will only ever see the result once you implemented the change and ran the code — and as we all know this can sometimes take longer than expected.