DP Big Letters

Sometimes you just need text as an image - a title card, a word to composite over a render, or the frames for a text reveal where each letter appears one at a time. This node draws big text onto a canvas you size and color, with a real font, and hands it back as an IMAGE. No diffusion involved; it's a straight text-rendering utility.

It's from DesertPixelAi's Desert-Pixel-Nodes pack, a large collection of quality-of-life tools. Filed under the pack's animation nodes because of its cycler mode, but it's just as useful for a single static text image. The default text is even "DESERT PIXEL," which tells you it's the author's own title-card helper, shared.

How it works

Two modes decide the shape of the output:

• Batch_Mode - renders the text and returns it as a batch, useful when you want the letters as a sequence of frames.
• Cycler_Mode - steps through the text one letter at a time, driven by index and cycle_mode. This is what you use to build a per-letter reveal: increment the index across runs and each frame shows the next letter.

Everything else is layout and styling - canvas size, padding on each side, font, weight, and the two colors. The font list is the set installed on the machine (DejaVu, Nimbus, Noto and friends), so there's nothing to download; it renders with system fonts.

The inputs and outputs that matter

The ones you'll actually touch:

• text - what to render (default "DESERT PIXEL").
• mode - Batch_Mode or Cycler_Mode.
• cycle_mode and index - how the cycler advances (increment / decrement / fixed) and where it currently is. Only relevant in Cycler_Mode.
• image_width and image_height - the canvas size (64 to 2048, default 1024 square).
• font_name, font_weight, font_color, background_color - pick from the available system fonts, a weight (bold / light / medium / regular), and named colors for text and background.

There's also per-side padding (padding_top/bottom/left/right) to keep the text off the edges.

Outputs are image (IMAGE - the rendered text) and letter_name (STRING - which letter/character is current, handy in Cycler_Mode for labeling or logging). Wire the image into a compositing node, a save node, or a video assembler.

How to install it

• ComfyUI Manager: search "Desert Pixel", install, restart. No models needed.
• Manual: cd ComfyUI/custom_nodes && git clone https://github.com/DesertPixelAi/ComfyUI-Desert-Pixel-Nodes, pip install -r requirements.txt if present, restart.

Find it under the DP category.

Common issues

The most likely surprise is fonts. The font_name dropdown lists the fonts available on the system running ComfyUI, and that list differs between a local install, a Docker image, and a cloud host - so a font you saw on one machine may not exist on another, and your carefully chosen typeface can fall back to a default. If the rendered font isn't what you picked, that's the cause; stick to broadly-available families like DejaVu if you move workflows between machines.

The other thing is text overflowing the canvas. Long text in a small image_width with generous padding will clip or shrink awkwardly. Give big words a bigger canvas, or trim the padding, until the text sits comfortably. And in Cycler_Mode, remember index advances according to cycle_mode - if you expected the same letter twice and got the next one, set cycle_mode to fixed while you dial in the look.