I’ve been spending some of my free time sketching in code using Processing and the Hype library. There are tons or resources and inspiration out there, but I’ve learned a lot by joining Joshua Davis’ Patreon page, where he shares the methods to his generative art madness. His lessons helped me generate this video:
Color, texture, and algorithm
One of takeaways I especially liked from this post was the idea that you can generate endless number of patterns by manipulating three variables: (1) color, (2) texture, and (3) the algorithm.
To speed up my process, I’ve started to build a library of colors and textures I can use in my code sketches. When I have time to create a new sketch, I grab a few elements from my library and jump right into code. Like anything else, when you start to look for inspiration, you’ll see it everywhere.
My color inspiration came from the plethora of Apple logos created to promote their October 30th event. I had saved of number of the logos from this gallery, so I picked three of my favorites, created gradient scales in Figma, then output those gradients as PNGs. Note: this color technique is covered in Josh’s posts on working with color.
My texture inspiration came from the case study of Uber’s 2018 rebrand. I spotted a cool tile pattern in one of the photos featured in their photography guidelines, and created several graphics based on those patterns.
Using my colors and textures and the code Josh shared, I started playing with different combinations of elements until I found one I liked. You can find my final source code on GitHub.
Turning a Processing sketch into a video
There are a number of ways to generate a video from a Processing sketch; my current preferred technique is using a snippet of code that exports each frame of the sketch as a high-res image. I wasn’t exactly sure how many frames it would take to create a usable video, so I started by exporting 5,000 jpegs. After ~12 minutes of render time, I looked through the jpegs to find a good stopping point for my video — frame 2,580 was the winner. It’s easily divisible by 30 and 60, which is great for my purposes because 1 frame = 1 second of video. Simple math!
I trashed the jpegs, changed the output format to TIFF (for a crisp, high-res image) and output my images. 2 minutes later, I had 2,580 TIFFs ready to use.
I like to use Adobe Premiere to create my videos because it gives me a lot of control over outputs and effects. I start by importing all the TIFFs into Premiere, add those to the timeline, set the frame duration to 1 second, then export a video. I’m still trying to find a combination of export settings that I’m happy with; so far, starting with Premiere’s default settings for Vimeo and changing VBR to 2-pass has delivered the best results. In this case, the result is this slow-paced video:
To make things more interesting, I import my first video into a new Premiere project. Now, I can play with the playback speed, visual effects (mirroring is always fun to try), and color treatments. In this case, I changed the duration to 10 seconds, which dramatically speeds up the animation, and exported a new video (using the same settings from the first pass), resulting in my final video:
It’s just that easy!
Once you get the steps down, this is a relatively quick process — and the possibilities are endless. If you think it looks interesting, I highly encourage you to jump in and try it for yourself. Josh’s Patreon posts are a great starting point and cover everything you need to get started.