Sigmas Chaos

Sigmas Chaos is the smaller cousin of Sigmas Attractor in RES4LYF's sigma toolbox - instead of a full three-dimensional differential-equation system like Lorenz, it iterates one of five textbook chaotic maps against your schedule: logistic, henon, tent, sine, or cubic. These are the maps you'd meet in an introductory chaos theory course, each one simple to write down and each one famous for producing wildly unpredictable output from a deterministic rule.

How it works

The chosen system gets iterated iterations times, driven by a control constant, parameter. One detail worth noting: the default parameter value is 3.9, and that's not an arbitrary placeholder - the logistic map (x → r·x·(1−x)) is only chaotic once its control constant r passes roughly 3.57, so 3.9 sits deliberately inside the chaotic regime rather than the map's tamer, periodic behavior at lower values. That's a small sign this node was built with actual knowledge of the underlying math, even without any documentation to go with it. normalize_output rescales the result into a usable sigma range, and use_as_seed switches whether your input sigmas seed the chaotic process or the map runs independently of them.

The inputs and outputs that matter

• sigmas (SIGMAS, required) - the input schedule.
• system (enum: logistic, henon, tent, sine, cubic; default logistic) - which chaotic map to iterate.
• parameter (default 3.9, range 0.1–5) - the map's control constant.
• iterations (default 10, up to 100) - how many times to apply the map.
• normalize_output (default true) - rescale into a usable range.
• use_as_seed (default false) - whether the input schedule seeds the process.

Output is a single SIGMAS list.

Should you use this?

Same honest answer as Sigmas Attractor: probably not for anything you care about the outcome of. I checked whether there's any real community discussion of RES4LYF's chaos-based sigma generators and found nothing - not one thread, not even in places where people are actively hand-tuning and sharing custom sigma schedules for specific models. This is squarely in the category the pack's own README describes as "experimental," and it reads as a mathematical curiosity more than a recommended technique. Worth playing with if you're the kind of person who enjoys that; not something to reach for if you just want a better image.

How to install it

• ComfyUI Manager - search "RES4LYF", install, restart.
• Manual - activate your venv, cd ComfyUI/custom_nodes && git clone https://github.com/ClownsharkBatwing/RES4LYF, cd RES4LYF, pip install -r requirements.txt (portable builds: use the embedded pip.exe). Restart.

Common issues & troubleshooting

Non-monotonic, possibly-negative output is expected, not a bug. Chaotic maps don't care that a sigma schedule is supposed to decrease smoothly. Run the result through Sigmas Abs and Sigmas Cleanup, and check the actual values with Sigmas Count or a preview, before it goes anywhere near a real render.

Near the edges of parameter's range, behavior can flip from chaotic to periodic without warning - that's just how these maps work (the logistic map, for instance, is well-behaved below ~3.57 and only chaotic above it). If results look suspiciously smooth or repetitive, you may have wandered out of the chaotic regime for whichever map you picked.