Very nice @LH99, and cool that you’ve made this in one week. Made me realise i’ve been making some OPENRNDR visuals for a VJ set recently — Made my 25 minute loop available on youtube for anyone to use for fun.
Haha true, that’s why I published it without music…
Nice! How did you do it? Did you first create many videos and put them together? Or is it one program creating everything in one take?
Yes because of time constraints I ended up making separate videos and stitched them together. This also allowed me to do some blending. Ofc this could have also been done in OPENRNDR. Next time 
I took me around 3 weeks to figure out the graphics pipeline.
I tried a lot of ways to make convincing LEDs.
I realized i hit gold when even a “hello world” circle in the center on key press felt juicy and snappy.
Sketch+Setup → 1 day (8h+)
Tweaking parameters to my liking → 1 evening, 3h
Implementing and following the inspiration of the shapes → 2 evenings, 3h each
The final puzzle piece came with the blending for colorBuffers.
Therefore, much love to @abe <3:
Sweet 
I like many of those “thousand cuts” fine-line visuals, but feel that the blocky volumes are a bit too synthetic.
Would imagine it depends on the audience, the music, the industrial/organic ratio that people expect 
But especially if I see your variation, I know that was the tip of the iceberg. Even the LED setup is quite fruitful. I mean, I used bars, circles, lissajous an bezier curves - the CG beginner visuals.
Would like too see more (even without music 
)
@LH99 Nice! I like the dreamy effect it has .
Can I ask if it is audio reactive? I can see there are some cues connected to the music, but I can’t appreciate if it reacts to it.
Not yet. Just raw inputs on beat.
I want to use minim, and maybe go for the bass/mid/treble abstraction that milkdrop uses.
For example, if you use Resolume (stage design software) and hook up parameters to volume …
they usually don’t look like you expect them to.
Perceived loudness (or intensity) is not raw volume afaik.
For non real-time purposes I’ve used https://www.sonicvisualiser.org/ and their command line version Sonic Annotator. They have many plugins to get all kinds of data out of audio files. Some of them can map well to how we perceive sound, but I think different analyses are good for different songs or kinds of music.
Got Minim with range detection to work.
The large bars with the strip indicate the sampled acoustic ranges (x = f from 20Hz to 20kHz, log scaled).
The green bars mimic old school VFD sound displays. They are “pushed up” by the detected volume, and take a bit of time to fall to the ground again.
The theory came from here:
I left Minim behind (out of poor performance) and now use TarsosDSP with pleasent results.
I made a sample application today to perform a few types of audio analysis.
Nice! Thank you for sharing that. It’s good to know about an alternative to Minim, and that TarsosDSP was updated recently:
Version 2.5
2023-01-09
Changes to the build system: moved from ant to gradle. Better documentation, testing and CI with github actions. Release of maven packages.
Here a few of my recent results using recursive texture lookups (inspired by the “Milkdrop” music visualizers)
Something I did recently, maybe fits with your visuals @LH99
Instancing a bunch of half-discarded (GLSL discard) spheres.
What have you been experimenting with? I want to see
If anyone is ever in Berlin the 1st Friday or the 3rd Saturday of the month, we have creative coding meetups going on . Sometimes, like in this case, I create the poster image using OPENRNDR.
