Azornes/Comfyui-Resolution-Master

What is ResolutionMaster?

ResolutionMaster is a custom node for ComfyUI, the visual workflow tool used for AI image‑generation pipelines. The node’s purpose is only to calculate and set image dimensions – it does not perform any image processing itself. It gives you a graphical “canvas” where you can drag a rectangle to pick a width and height, and then outputs those numbers (and a derived rescale factor) for the rest of your workflow.

---

Core capabilities

Feature	What it does	Why it matters
2‑D canvas control	Drag a dot on a grid to set width × height; modifiers (Shift, Ctrl) let you keep the aspect ratio or disable snapping.	Quickly pick exact resolutions without typing numbers.
Smart rescaling	Calculates a scale factor (e.g., 2.0×) that you can feed to up‑sampling nodes.	Keeps the math consistent across the workflow.
Snap button & snap‑step slider	Rounds the current dimensions to the nearest multiple (16‑256 px).	Guarantees sizes that are friendly to most diffusion models (they often require dimensions divisible by 8, 16, or 32).
Real‑time info panel	Shows the current pixel count, megapixels, p‑value and aspect ratio as you edit.	Lets you see at a glance whether the size fits your target (e.g., 4 MP for a given model).
Preset library	Hundreds of ready‑made resolutions grouped by use‑case (Standard, SDXL, Flux, Social Media, Print, Cinema, etc.).	No need to remember exact numbers – pick a preset like 1080p or 16:9 and the node fills in the values.
Model‑specific optimizations	When a preset category is selected, the node can automatically enforce constraints required by that model (e.g., SDXL only accepts certain fixed sizes, Flux uses 32‑px steps and caps at 4 MP).	Prevents “invalid resolution” errors that would otherwise stop generation.
Auto‑detect	If you connect an image to the node, it can read that image’s dimensions every second and update the outputs automatically.	Useful for workflows that resize an existing picture or want to keep the original aspect ratio.
Preset manager	UI for creating, editing, renaming, and bulk‑deleting your own custom presets.	Lets you build a personal library (e.g., company‑specific banner sizes).
Extensive property panel	All numeric limits (canvas size, snap step, scaling ranges, etc.) are exposed as node properties you can tweak in ComfyUI.	Adapt the node for high‑resolution work such as 8K images.

---

How you would use it in a typical ComfyUI workflow

1. Add the node – “Resolution Master” appears in the node palette.
2. Connect outputs – width and height feed into any image‑generation or resize node; rescale_factor can be wired to an up‑sampler.
3. Choose a preset or drag on the canvas – Pick a category like SDXL and a preset such as 1024×1024, or manually drag the blue rectangle to the exact size you need.
4. (Optional) Auto‑detect – Connect a “Load Image” node; enable the Auto‑Detect toggle so the node reads the source image’s size and suggests the closest preset.
5. Resize – Add a “Resize Image” node, wire the dimensions from ResolutionMaster, and you now have an image that matches the exact resolution required by your downstream model.

---

Who benefits?

• Artists & hobbyists building stable‑diffusion pipelines who need to meet strict size constraints.
• Developers creating reusable ComfyUI templates – the node guarantees that generated images always conform to model limits.
• Content creators who frequently output to specific platforms (Instagram, YouTube, print) and want a one‑click way to hit the right dimensions.

---

Installation

• Via ComfyUI‑Manager – search for Comfyui‑Resolution‑Master and click Install.
• Manual – clone the repo into ComfyUI/custom_nodes/ and restart ComfyUI.

---

Quick‑start checklist

1. Install the node.
2. Drop it into your graph.
3. Pick a preset or drag on the canvas.
4. Wire width/height to your generator or resize node.
5. (Optional) Enable Auto‑Detect for image‑based workflows.

---

Bottom line

ResolutionMaster is a precision‑control helper for ComfyUI that removes the guesswork from setting image sizes, enforces model‑specific limits, and streamlines the creation of reproducible AI‑image pipelines.