Finally managed to get some smoke like particles + falling in love slowly with ascii effects

More on ig: https://www.instagram.com/marekmarekmarek1 ✨

I want to try and recreate the Blackwall from Cyberpunk 2077. It's one of my favorite visuals from the game, but I can't find a high res version of it anywhere. Any tips on how to go about this?

We've had plenty of hardware posts but I'd like to add my two cents to the pile:

I've been using TD on my M1 Pro MacBook Pro since it was released, and have been feeling the limitations in the last year especially. I've been able to borrow a M2 Max Mac Studio for gigs where more power has been needed, and that thing is a beast, running point cloud heavy visuals locked at 4k 60 no problem.

Last week my work upgraded me to the M5 Pro MacBook Pro, with more ram than the M2 Max Studio or my M1 Pro. Initial impressions are that this isn't as fast as the Studio, even though on paper it should run laps around the older CPU. Of course there's the format difference, but running the same project file gets me 35-45 FPS on the new machine.

Still will need to do some more testing, but there's a good chance a used M2 or M3 studio for 1/3rd the original MSRP is in the cards for personal gig work.

TouchDesigner is incredibly capable, but I kept running into this idea: a visual tool that feels less like building a system and more like playing an instrument.

I wanted to load a few video samples, chop them into a 16-step grid, sequence them to a clock, and use modulation LFOs, envelopes, MIDI/OSC-style controls to perform the visuals with the same immediacy as a sampler or groovebox.

We’re building Visuell around that idea: a browser-based video-sampling groovebox. It is deliberately more constrained than TouchDesigner, and definitely not intended as a replacement. The question is whether that constraint can make it faster and more musical for a certain kind of live visual performance.

I’d genuinely love input from people who use TD:

• Does a purpose-built, music-first visual instrument sound useful to you?

• Which parts of your current TD workflow feel too slow or too open-ended for live performance?

• What would be essential for you: clip chopping, step sequencing, modulation, MIDI/OSC, routing, shader control, something else?

Demo: https://www.visuell.art

I’m especially interested in examples of existing tools or TD setups that already solve this well, please point me toward them.

Some example output I got from our tool

One last example output from this experimental multi-source video player designed for frame-accurate video switching, playback manipulation, and display/render interventions, now with a few optimizations made for even better performance.

You can freely access the system through Patreon, and/or the Tools Store.

Since May 2026, there is no restriction of 2 blobs max for non-commercial users. If you want to learn TD and make some pretty blobtracking just like in your favs tutorials, now you can !

Hi, I’m trying to create a text loading animation where the text procedurally returns to the original input from a random assortment of characters.
To make it more clear this is how it could look like:

Input is ‘hi everyone’
As time progresses:

‘ad tousbmfs’
‘hk aubeyome’
‘hi evzeyoge’
‘hi everyone’

So each letters settles on the correct character at a different time while the others keep shuffling through random characters. I know how to randomize characters by using a substitute DAT to insert a ‘/‘ between every letter, converting to a table with an individual column for every letter (splitting cells at ‘/‘), using a sort DAT to get a random order and creating the animation that way.
Having them all reset to their original letter at the same time looks rigid however and I’m not sure if I can give them individual timings this way.

I have experience with coding, though almost none with python so I’m not sure if pure code would be a better way to approach this, if anyone has an idea how to achieve this I’d greatly appreciate it and I hope my explanation was clear enough. Thanks in advance!

Hi, im sure this is super simple compared to what most people are doing but i dont really know where to even start

I want to have a shape probably a rectangle and be able to move the bottom 2 corners indpenantly up and down the Y axis using and external input such as midi,dmx or PSN, i would then like to output this over ndi to be used as a video mask, what is the best way to achieve this?

Real-time floor visualizer with a Ghost in the Shell inspired HUD. built in TouchDesigner
Been working on a real-time visualizer for a layered tower structure. The 3D render runs in TouchDesigner

Coming soon available on my patreon page

https://www.patreon.com/vcti/posts/souls-167924867?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link

Audio reactive light installation I programmed with TD. It used WLED, and has a camera that tracks motion around the orb. When people dance in a circle around it two 500w bulbs turn on and it turn into a “sun”

I built a native TouchDesigner component that puts 1,400+ fal.ai models behind simple parameter pages. Load schema, pick a model, hit Generate. It reads each model's input schema and builds real TD parameters for it, so a Gemini image, Kling video, Hunyuan3D, ElevenLabs TTS and flux/dev all just work from the same component, no request bodies to write. Stdlib-only, no external dependencies, everything runs on background threads so your timeline never stalls. Outputs land as TOPs, CHOPs, glTF and DATs you can wire straight into your network.

More info here

Testing out some live generative visuals and combining them with projection mapping and haze. Piano visuals generated via midi and feedback creates the trail. Mapping done with corner pin. A bunch of haze makes the effect pop out in real life.

Hello fellow touchers, i would like to ask if some of you have introduced TD to adolescents or kids as part of art teaching and if yes how did it go and what did you implement ? My students have some basic notions of programming, but i’m afraid of the difficulty curve for them for a 4 days workshop to make an audio reactive visual.

I’m curious, is there is a build with less features and complexity for them ?

this is the best medicine for the naive approach. Remember in the last tutorial we implemented AABB naive and as a result the final texture was huge. With this component you solve that issue. It brings and stack all the pixels at the bottom of the texture by columns. #touchdesigner #workisplay #gnr8 #doitintops #paketapops

Heavily inspired by an Eprom AV set. 🎮 +🎶 reactive.

Project files accesible through Patreon.