Algorithmic Fine Artist
Sudfah
The process behind the Curated Artblocks Generative Art NFT Drop.
.jpg)
From the moment I learned about Artblocks - around November 2021 - I was SUPER excited about the idea of applying with a project. I quickly became obsessed with the idea of dropping on Artblocks and thought about it day and night.
One day I was really jet-lagged from an overseas flight and trying to adjust my sleep schedule. I went to bed early only to wake up at 2 am with a clear art idea in my head - a single drawn line that went vertically from top to bottom on a canvas, wiggling around in the middle, and then gets messed up with digital liquid. I got up, went straight to the computer, and spent about 2 hours coding up the very first version of this project. What you see on the left is pretty much the result of that initial inspiration.
At first, I played around a lot with the liquid spill itself trying to make it look as realistic as I could. I experimented with the darkness of the bleed and also with how much of the original line to leave in place.
Notice I hadn't yet thought of making the line look calligraphic.
.jpg)
.jpg)
At this point, I felt really stuck. I did like what I had come with so far, but couldn't think of any creative approaches toward developing enough interesting variation for an Artblocks drop.
I made a few mockups to help myself see the art from a new perspective and was quite pleased with how it looked in a contemporary setting. I decided to take a break and let this project breathe a bit, whilst I continued to work on other generative art.
A few days later it came to me that I could try experimenting with different types of drawn line styles. I tried making a sort of 'brushed' effect but didn't really like it all that much.
.png)
.png)
I then had two major breakthroughs in one day. I tried a calligraphic line and LOVED the result. But the particular way I chose to do it at that time, made the entire project run slowly. I now had to think of ways to optimize my code.
I realized the blur for the bleed was probably taking up most of the computation, so I decreased the 'quality' of the blur all the way down from 100 to 3.
What emerged amazed me, as all of a sudden the spill had SO much more character and texture! It was really a big ’aha' moment, honestly.
Still trying to think of how to add variation, I played with line width and different coloring. I even tried adding a rainbow option for a bit, but it didn't make the final cut.
.png)
.png)
Having come this far, I was still struggling with knowing that the variation that was developed so far wasn't enough for me. It was too simplistic in output but had so much potential, so I continued to ponder a more varied range of options.
In the next few days, I got the idea to turn the whole thing sideways and make it landscape format to emphasize the feeling of it resembling a written text format. I also decided to make a 'baseline' from which there would be shapes rising up. I was thinking that those shapes could also resemble things like mountains, trees, people, etc. The first image I saved from that experiment is on the right.
Also at this point, I added different 'spill types' and some black circles which eventually became the 'blots'.
I still had MONTHS of tweaking ahead of me to get everything to look just the way I liked it. The circles had to be warped, and the different spills all had so many different things about them that had to be adjusted. But one could say that the base of everything was present at this stage of the development.
%20copy.jpg)
I tried several other things that didn't make the final cut. For example, layering different spills on top of each other in a way that they didn't interact with each other. I actually did really like the results but I had to cut it for performance reasons.
My daughter absolutely loves this piece which she has titled 'the' and insists that I show you.
At one point I even tried animating the piece.
While I do enjoy the animation, I found it to be problematic for consistent performance and I didn't feel it added enough to the overall piece to be worthwhile, so it also didn't make the final cut.
.png)
.png)
.png)
.png)
.png)
.png)
The end of the project consisted of months of editing, trying things, backing up, redoing this thing or that thing, and polishing, polishing, polishing. Deciding on how much creative mess to make, how much line to leave under the blotchy surface, and how to develop a pleasing color palette that suited the mood of the project took up most of the time.
The absolute hardest part of the entire project for me was testing in the Artblocks renderer and making sure the outputs were consistent across all browsers and devices and were performant enough to get a good render without crashing.
I had finally arrived at a harmonious balance where I could celebrate the beauty found in an accidental mess and simply loved the arising details in a drop of spilled liquid. I do find it ironic how much work I put into making sure the accidental mess was just right.
What is interesting to note is that I actually thought about the concept of trying to simulate an ink bleed with code for years before it came to me as a complete idea for an NFT drop. Having thought that it would be really hard to code it successfully, I now just had to prove to myself that it could be done.
I also wanted to make sure that it was really fast and didn't make the computer struggle to render it - that was sort of a geeky obsession of mine throughout the whole process.
Through this piece, I hope that I have pushed a few boundaries. I do feel that we are JUST NOW starting to see what this art form is capable of achieving and I hope to be one generative artist that helps the world really see what an amazing creative medium this is.
Thank you all for following along on this journey. I hope you love Sudfah as much as I do!