Introduction to Computational Practice

Initial Ideas

I wanted to create something which combined both of my interests, which are music, music videos and computation.

So, I decided to create a project which was a collection of visualisers made with code for various songs, or a very stylised music video in which the coded art served as the background for the musician in a music video. However, during the process, I realised that this would take longer than anticipated, so I decided to use one song and a collection of similar coded visuals.

I got the inspiration from the visualisers I saw for music on major streaming platforms, such as Spotify (although I do not see this feature any more). These are frequently characterised by moving visual art, or small snippets playing in a loop of the actual music video.

While I initially really wanted to get a friend or family member to act as the musician in a full blown music video where the background visuals were my coded art, this proved to be unfeasible with the time I had. So I went with the simpler idea of just making visuals.

Examples

1. MICHAEL JACKSON – LEAVE ME ALONE

This is a highly stylised video in which the main musician plays a role in the video but the video is characterised by the bizarre visuals in the background. The background visuals are what I will be attempting to code.

1. RIHANNA – RUDE BOY

This is also a very stylised video but makes more sense to create as I am able to make it fit the brief. The background visuals are mainly a myriad of colours and shapes which I am able to take inspiration from and code through p5.js.

Initial Planning

After this idea solidified, I decided to start thinking of how to execute this. I did not want to overextend myself with learning new software and coding languages when I am already in the midst of learning others, so I decided to use:

1. P5.JS – since I am already familiar with P5.JS, I decided to start creating visuals with this, including many variations of one pattern e.g.

I also wanted to use other Visual Coding Programmes as there were many options available, but I knew that I wanted to build on skills I already had rather than confusing myself by spreading my attention thin over many options.

2. ADOBE PREMIERE PRO

It is a good all round software with which colour can be adjusted, the designs can be cropped into the desired shape, it can be animated and then made into a full video.

Music Aspect

For the music, I wanted to use song covers that I did using Adobe Audition, while of course giving due credit to the respective musician as well as the creator of the backing instrumentals. I was not too sure if this would be a copyright issue but I decided to make the visuals first and keep the video unlisted on YouTube. The song I planned to do was Sway – Dean Martin (1964) which is a song that has been covered numerous times by various artists. I wanted to make coded designs for this 1 minute cover according to its genre, style and mood and then edit it into a video.

To Consider

• To meet the brief ‘LANGUAGE’ as efficiently as possible without overdoing or confusing with personal projects.
• To execute it so that it is visually appropriate to the song and has a unique touch – artistically and technologically sound.

Moodboards

• the running theme will be that the end format will be a video but in which there is a spinning circle where the art and visuals appear. This is to give the visual effect of a spinning vinyl but with a unique touch.
• I feel that this is a unique idea in which I will be able to incorporate computational practices as well as a retro aesthetic, shown by the two video examples below.

The picture shown on the left is the music video for ‘Hula Hoops’. The video is a vivid, colourful music video that has all elements of a traditional music video in the pop genre, except that the video is cropped into a circle, giving it the effect of a CD/Vinyl record.

The picture on the right is the record for Takako Mamiya’s 1982 album ‘Love Trip’. One of the comments under the video state ‘it would be good if the record was spinning’, which led me to develop ideas for simple music video visuals that incorporate the vintage aesthetic of a record as well as modern computational practices for the visuals displayed.

Sway

For the cover of Sway, I took a lot of inspiration from the colour schemes and imagery in Alice in Wonderland. There is a lot of very vivid colours and imagery that I though would go well with a song like this. I initially wanted to do a larger variety of items e.g. dice, cards, mushrooms, but did not want a big barrage of different imagery for just a minute long snippet of music. I then decided to just code a few variations of checkerboard designs moving in different directions as I was not even confident with this one concept yet.

P5.js

• using p5.js, I created colourful sketches based on the same checkerboard design, that I thought were apt to the song and the mood boards I created for it.
• I then cropped the videos into a CD shape and made it rotate using Adobe Premiere Pro using the transform, crop and circle options.

Coding Process

1. Coding different checkerboard designs.

I ended up with 2 patchwork checkerboard designs, 2 scrolling designs and 4 still designs (which would end up animated anyway by virtue of a spinning record).

Final Pieces

The colours are exactly as I envisioned them, and I planned to do the slight adjustments such as exposure and contrast on Adobe Premiere Pro.

Animating Process

Splicing them into a video with Premiere Pro

Using the saveCanvas() function on p5.js, I downloaded all the work I did. For the animated ones, I screen recorded them and the cropped them. I imported all the pieces into Adobe Premiere Pro and set them all to the same size and position. I added the audio.

I then cropped them into identical circles and colour corrected them slightly so that one was not brighter, more exposed, or cooler than the rest. This was difficult to do in p5.js as RGB values have to be inputted and replaced in a ‘trial and error’ method to find a satisfactory colour combination, which would be an arduous process.

Lastly, I added a small black circle in the middle to make it look like a CD and animated them all to rotate at the same speed to give it the effect of a rotating vinyl. (Had I made the central circle bigger it would look more like a vinyl but the designs would not be as visible, so I kept it smaller).

Final Piece

Reflection

While I believe this is a strong start for executing creative thinking using computational practices, I think there are several areas that can be improved on was I to continue this project in the future. For example, using code to create background visuals for a music video (or a music video itself) might be a tentative next step as here I have just used code in order to create a spinning vinyl animation to accompany a song. In addition to this, incorporating code more seamlessly into every step of the process might be a useful way to further my creative computational practice. In this project, I used code to design the patterns but other software to edit it into a video. However, this is something that can be done using code as well, so delving more deeply into hard coding skills might be useful in this pursuit. Lastly, the fact that only one coding language was used is something that can be improved upon. A myriad of visual programming languages exist to create such designs and making use of some of these, like Hydra and TouchDesigner would benefit my practice, and I would definitely utilise these next time. These languages require extensive knowledge to produce anything satisfactory for the project I imagined however, so I kept it aside for the time being.

While making this project, I was able to greatly improve my technological skills, regarding p5.js and Adobe software, which allowed me to work with increased efficiency. This allowed me to create what I envisioned more accurately. The process was equally challenging and rewarding and I was able to take a step towards creating something I enjoy using newly acquired skillsets. Through trial and error, I learned many new methods in code and how best to create the design I wanted. I also felt that that end product was a satisfactory culmination of everything I newly learned and already knew.

Evaluation

Since I drew plenty of inspiration from the work of other creators and artists, my work is one among many that exist like it, but displayed through my personal artistic vision. Music videos such Rude Boy by Rihanna incorporate similar vivid visuals, and laid the base for the project I created. As well as this, cropping videos into different shapes, and the designs I created are not new practices or inventions but rather ones that many have utilised across various artistic practices and areas, and can be seen in the music video for Hula Hoops – DPR Live and the game-like visuals portrayed in the fantasy genre. Visual Coding is also a process that has been around for years, and many pieces of media are created through this.

However, while it is true that my project exists courtesy of the work of artists before me, the piece stands as something made through a creative practice that I developed in my own unique way. I think that the way in which one codes, why they create, the designs and colours they choose and how they express their creativity through technology is unique to them. The computational practices and methods I used are quite common, but to use them in an innovative way to exhibit my art was illuminating and gratifying.

Initial Idea Development

Since I am interested in music and had previously researched an algorithmic composer in a previous unit, I wanted to attempt using code to compose a simple song which lasts a minute maximum. There are several Audio Programming Languages, but research into several led to the conclusion that Sonic Pi seemed to be the easiest to learn and seemed to have a beginner friendly interface. As well as this, the live coding performances I found online were very engaging to me.

The execution of this project would involve learning a new language but I planned to learn the very basics and create the simplest tune I could; as long as I could create a minute long tune then that was satisfactory. I also had no background in music theory, so everything from a sensible chord progression or a note would have to be googled. I also found a previous lecture and task on ‘performing code’ to be incredibly intriguing, so I wanted to do something which would apply code performance to this project.

There were several video tutorials for Sonic Pi online, but they seemed to be teaching a version that was no longer available as it had updated since. So, I decided to use the manual Sonic Pi itself provides in order to familiarise myself with the basics.

Inspirations

The below are my initial inspiration for the tunes. These seemed especially doable as they did not incorporate any instruments but piano, which would be perfect to attempt as I do not have a background in music production and was just starting out with coding music.

Moodboard

I took a lot of inspiration from retro sci fi art works that revolve around space exploration, as well as comics I read as a child.

Human Machine Interactions

For a previous task, I was prompted to research human machine interactions, how they can be thought about differently and how it might change in the future. In this project, since I am going to code music, I thought about how creativity is becoming increasingly technology assisted, whether that be making art, architecture or music. My diving into this field makes me believe that this technology utilisation in the creation of art is not restricted to the incredibly smart or tech minded people, but will become ever more available to the common person. Not just this, I feel that it might be likely used as commonly as automatic cars.

Process and Development

Using Sonic Pi

I first played around with chords and chord progressions, which created appealing but generic sounds. I also learned about the sleep function which creates the gap between two notes, and so played around with that until I was familiar with how timing worked within Sonic Pi.

Deciding that I wanted to do something more original, I decided to do a bit of exploration since I have no knowledge in music theory. I first hummed the tune I wanted into voice memos, and then matched the tune with a virtual online piano through by ear, as I do not know pitch names nor piano keys. This required a great deal of trial and error as matching a note to one piano key among many was tiring work.

I played the resulting melody through again and again, making changes in the timing, adding effects such as reverb or wobble, and playing around with the synth in order to create one that I liked. I also had to keep in mind that I was going for a spooky or ethereal space themes tune, so I did not have too much leeway when playing around on the software.

Using Adobe Audition

After I created a satisfactory tune, I recorded it and imported it into Adobe Audition, where I played around with extra effects such as adding some delays, compression and putting it through the parametric equaliser to get the effects that I wanted. While I had made the base tune in Sonic Pi, I wanted it to have a little extra reverb to create a more immersive quality.

I also downloaded free open source sounds from Zapsplat such as white noise, and the sound of an asteroid passing by in order to make a more immersive soundscape.

Final Piece

https://youtu.be/ZGcsdCWpIZI

Reflection

This project allowed me to play around and learn the rudiments of new software, and I was able to create a solid piece, although I was not able to utilise the software every step of the way. Any extra edits, effects, and anything beyond the basic tune had to be done on Adobe Audition. As I continue my journey with learning Sonic Pi, I would use the program to create music without help from external software that would incorporate many different instruments and chords progressions so as to work towards a more professional sounding piece. The work I created is very basic and only contains a few key elements of the bigger project I envisioned. While I successfully executed the idea of a soundscape and music amalgamation, it lacks key elements such as length, immersion and better production quality. This is unmistakably due to the fact that I am amateur at both music theory and Sonic Pi, which is easily remedied for future projects by constant studying and practice.

While I was in the process of creating this, I was able to dive into a completely new Audio Programming Language and learn the basics, which was incredibly engaging. The constant use of trial and error, going back and redoing, and scrapping old code in favour of new code solidified my knowledge base and I am now confident in my ability to build upon that solid foundation. I also greatly enjoyed coming up with a theme and cover art and using my mood boards and sketches to guide my coded composition.

Evaluation

Algorithmic composition has an extensive history and dabbling in making music through code was an enlightening process. I was introduced to new ways of implementing my creativity through computational practice. Of the many algorithmic compositions, AI-assisted music pieces and tunes created by Sonic Pi, I would say mine is very simple and does not include many of what they do. However, my end goal was not to create an incredibly advanced, code heavy professional piece, but rather a simple musical soundscape, which I was able to achieve. The soundscapes I found online, created by others, were similarly simple, but of more professionalism. Thus, I ended up satisfied with what I made using the little I learnt, and believe that it expresses what I planned for it to express.