Video Glitch Art Enhanced

This is the sibling of Video Glitch Art Transition, and "Enhanced" is not marketing: it's the version where the glitch stops being a transition and starts being an aesthetic event. Where the regular node gives you signal interference and pixel sorting, this one offers compression_nightmare, digital_meltdown, quantum_decay, and neural_collapse - shader effects tuned to look less like a damaged file and more like the video is having a medical emergency. If you want an intro that looks like AI art is destroying itself in real time, this is the pack's most distinctive node. It's also its most system-hungry.

How it works

Same machinery as the base glitch node: headless Chromium, frames in as base64, effects running as WebGL/GLSL shaders, screenshots back to an IMAGE tensor. The difference is the tuning surface - the effects are built to be pushed hard, and the inputs read like chaos dials rather than cleanup knobs. The README's flavor text ("compression nightmare," "neural collapse") tells you the intent: these are deliberately aggressive, AI-culture-flavored distortions. use_gpu defaults to true here too, and these four styles are the heaviest shaders in the pack.

Inputs and outputs that matter

• video1, video2 - the two clips.
• glitch_style - compression_nightmare, digital_meltdown, quantum_decay, neural_collapse. Each is a distinct look; worth rendering a couple frames per style to see which fits.
• visual_intensity (1–5, default 3) - the master violence dial. 1 is still glitchy; 5 is almost abstract.
• chaos_level (0.5–5, default 2) - how unpredictable the corruption is frame to frame.
• corruption_rate (0.2–1, default 0.6) - fraction of the frame corrupted at peak.
• color_madness (1–4, default 2.5) - how far the RGB channels fly apart.
• screen_tear and flash_effects - toggles for horizontal tearing and strobing flashes. Note the flashes: flash_effects on is genuinely strobe-y - worth keeping in mind for light-sensitive viewers.
• total_frames (4–120, default 36), fps (default 24).
• use_gpu (default on), batch_size (default 6), background_color, width/height, quality.

Output is video - an IMAGE batch; combine it with VHS_VideoCombine or ImageBatch to export.

Installing it

Same pack, same install as the base glitch node:

cd ComfyUI/custom_nodes
git clone https://github.com/yichengup/ComfyUI-VideoTransition
cd ComfyUI-VideoTransition
pip install -r requirements.txt
playwright install chromium

Or just ComfyUI Manager → "VideoTransition". Chromium is required.

Common issues

• This one chugs harder than the base glitch - expected, the shaders are heavier. Drop visual_intensity toward 1–2 for speed, or lower fps/total_frames per the README's own advice.
• Flashy to the point of strobing - that's flash_effects. Off if you want something a human can watch.
• Too much, too fast - chaos_level and corruption_rate at defaults (2 / 0.6) are already aggressive. For a usable transition, start at intensity 1–2 and short total_frames; reserve the defaults for full-screen destruction.
• Missing Chromium crash - playwright install chromium (mirror flag behind the GFW).