creative systems design, generative mograph, working at the intersection of sausages and ketchup co-founder of https://xitter.cf/t.co/9zgpslmHXV he/him

Plymouth, UK
Joined July 2015
Hey @NotionHQ I just want to take Notes. You know, the thing you were made for. I don’t want to chat to an AI our have AI fill in notes! I want to do it myself. Why can apps not just STOP adding features. You’ve finished building the product years ago, just stop adding things nobody installed you for.
1
163
We're safe.
Today we're launching Motion in Claude. One skill turns it into a full video team: > Drop a product URL: a finished launch video comes back, scripted, voiced, rendered > Say "make it shorter" or "swap the music": it re-cuts on the spot, no timeline > Upload your own footage and logo: it builds around them The video below was made this way, in Claude, in one message. Comment "MOTION" and I'll send you the skill. πŸ‘‡
1
241
For many years I've wanted to have a website that shows a constrant stream of cool motion design inspiration to run on a spare screen in a design studio. So I made the FrameRate ScreenSaver which displays a grid of videos from FrameRate and plays random ones all day long. play.florianthamer.com/frss Thanks to @__T_Williams__ and @justincone for making FrameRate what Vimeo used to be, but better!
1
4
1
30
2,211
This is the third iteration of the setup. I'm using Meta's free and open source DEMUCS for stem extraction, and librosa for analyzing the audio data. Do I know what any of that means? Hell no. But Gemini/ChatGPT/Claude knows and that's good enough for my little Python SOP!
2
3
27
1,421
Why not have the particles move so frantically that the compression on social media fux up the whole thing and then call it artistic intent lol.
2
1
21
841
We've recently gotten into doing lots of audio-visualization RnD at sunmachine.io . So I've built a @sidefx Houdini HDA that extracts stems from music and outputs clean midi for the instruments. I've then piped it into a little POP solver advecting particles with the filamentsample() function only, because it looks cool!
9
11
1
186
6,020
π–‹π–‘π–”π–—π–Žπ–†π–“ π–™π–π–†π–’π–Šπ–— β˜€οΈπŸ€– retweeted
A bit of data extraction from audio. Using #python and #houdini
31
123
9
912
44,657
"What bitrate should I export my mp4s in?" I used to wonder this until I found a formula in an old PDF that an Adobe employee created over 15 years ago. I turned it into a little calculator for everyone to use: bitratecalculator.app
22
90
7
1,159
62,459
π–‹π–‘π–”π–—π–Žπ–†π–“ π–™π–π–†π–’π–Šπ–— β˜€οΈπŸ€– retweeted
A little music visualisation experiment. More info: sunmachine.io/work/light-for… #dataviz #musicvisualization #houdini #ebsynth #suno
1
2
14
706
π–‹π–‘π–”π–—π–Žπ–†π–“ π–™π–π–†π–’π–Šπ–— β˜€οΈπŸ€– retweeted
Human vocals β†’ data β†’ particle simulation. This visualization extracts vocal features and uses them to drive a particle system in Houdini. #houdini #synesthesia #audiovisualizatioin #soundart #mograph sunmachine.io/work/acoustic-…
3
4
47
1,538
π–‹π–‘π–”π–—π–Žπ–†π–“ π–™π–π–†π–’π–Šπ–— β˜€οΈπŸ€– retweeted
Another test experimenting with visualising audio. This one’s a thrush. I analysed the recording in Spear, then rebuilt the spectral data inside Houdini with Python. #houdini #audiovisualization #sonification sunmachine.io/work/acoustic-…
3
4
57
2,471
Simon and I have opened a studio called sunmachine.io recently, we love doing things like this. He's also a clever bastard sometimes!
I’ve been experimenting with visualising different kinds of audio, human voices and animal calls. This one’s a thrush. I analysed the recording in Spear, then rebuilt the spectral data inside Houdini with Python. More here: sunmachine.io/work/acousti... #houdini #audiovis #mograph
1
1
6
776
And we thought AI will make it impossible to know what's real anymore. How we've overestimated the human intelligence.
OMG HE THOUGHT THE PHOTOSHOP KNUCKLES WERE REAL
1
1
405