> In my seven years of programming I have had the most fun in the past few months after discovering generative art.
Careful! It's totally addictive. Next thing you know, you'll be spending thousands on high res prints, or even worse, making it your day job and your hobby. ;)
Nice work! It looks great, and I'm looking forward to more.
I feel myself getting this addiction. Can you say more about what kind of day jobs use these techniques?
I realise that to achieve good results requires time, and the easiest way to find time is to spend working hours on it, so I'm considering pushing my consultancy in a direction that includes this kind of work---though I'm not actually sure what this is!
Design teams at large companies that have a strong design culture, ad agencies, and design studios working on contract work (particularly installations for music events etc) are the kind of day jobs where you can work on these kind of things.
It definitely takes some time (and a lot of personal work) to break in in this field though. Like most artistic practices sold to the Market, deadlines are usually tight, pay isn't the highest, and there are more people who would like to make a full time living from it than there is available paid work.
Get in a routine (make things with regularity, many artists make something daily), make your work public (post it on Instagram/Twitter/art subreddits/etc.), make it clear that you're available for paid work, and after some time you might make it happen. Good luck!
Vfx & animation is one place people do this, and what I’ve done. Advertising is another one, but probably more hit & miss, I think there are only a few people who figure out how to spend much time doing the fun stuff. Another avenue is selling your generative art, but having tried a little bit, breaking into the art market can be pretty tough. Some design work can let you practice generative techniques too.
Back in a college graphics class I had a lot of fun, and some lucky success, with roughly simulating watercolors with GL shaders. I kept several textures that represented the paper, it's wetness, and the paint. The shader would calculate new states for each based on the previous state, time and user input.
The cool thing was that you could watch the paint disperse in real time and could add things like gravity to cause it to drip down the page.
It wasn't physically that accurate, but it looked pretty good and was super fun to hack on. If only I could find the source :(
It's too bad that e-ink never really went anywhere. I'd love to have low-power/no-power, random, dynamic, full-color paintings hung up around my house.
Nice idea, here's a little WebGL version inspired by your post. I just used a noise function rather than the polygons though. https://www.shadertoy.com/view/lt2BRm
Ah, I made something similar back when that original article came out. Should have maybe fleshed it out and posted on HN? https://github.com/akx/watercolour
https://hi.stamen.com/watercolor-process-3dd5135861fe