liveBook · Manning

Resolution

Resolution is the fineness with which values, measurements, or spatial details can be distinguished or represented.

How 6 Manning books explain it

Build a Text-to-Image Generator (from Scratch)

Image resolution is the spatial dimensions and corresponding level of discrete visual detail represented by an image.

How it is used
Frames image-generation output as a trade-off between spatial detail, computational cost, and latent-space efficiency.
How deep it goes
Traces 4 × 64 × 64 latent tensors through VAE decoding to 3 × 512 × 512 outputs and demonstrates 768 × 512 generation.
Examples
Uses banana rider, lion and tiger outputs, astronaut-unicorn prompts, and a 768 × 512 beach scene.
What you can do after
Operate/configure Stable Diffusion through diffusers, selecting width and height and inspecting latent-to-output conversion.

Read how Build a Text-to-Image Generator (from Scratch) explains resolution →

How Computers Make Books

Resolution is the density of individual image dots within a specified physical area.

How it is used
Connects physical printing and pixel screens to the problem of placing marks that form readable images.
How deep it goes
Defines dpi, calculates page dot counts, and works through how resolution changes rasterized-line quality.
Examples
Uses a capital A at 60, 30, and 15 dpi, a 1,200 dpi letter page, and lines of differing slopes.
What you can do after
Evaluate/understand why a denser dot grid improves approximations and why output devices require different renderings.

Read how How Computers Make Books explains resolution →

Quantization and Fast Inference

Resolution is the spacing between adjacent representable numerical values.

How it is used
Guides quantization choices by balancing representable range against the spacing between values.
How deep it goes
Defines representable-value spacing and compares fixed-point steps with near-zero density in low-precision formats.
Examples
Compares Q3.4 steps of 0.0625 with E4M3's 49 and symmetric INT8's 32 representable values in [0, 0.5].
What you can do after
Evaluate number formats by comparing their step sizes and representable-value density across the ranges that matter.

Read how Quantization and Fast Inference explains resolution →

WordPress Power Toolkit: Harness AI to build next-level websites

Resolution is the amount of detail represented in a digital image or display, commonly expressed through pixel dimensions or density.

How it is used
Guides image preparation for websites by balancing visible detail, cropping flexibility, storage, and loading speed.
How deep it goes
Defines pixel density, dimensions, and file size, then applies them to web image recommendations and cropping consequences.
Examples
Uses cropped low-resolution imagery, a 1,920 × 1,080 background, and a 1,280 × 720 hero image.
What you can do after
Apply suitable image dimensions and resolution when cropping or preparing images for website display.

Read how WordPress Power Toolkit: Harness AI to build next-level websites explains resolution →

Learn Generative AI with PyTorch

The pixel dimensions of an image, used here to compare lower- and higher-resolution generated images.

How it is used
Compares generated-image sizes while connecting pixel dimensions to preprocessing, model inputs, and training memory.
How deep it goes
Defines resolution through 64 × 64 and 256 × 256 examples, then traces tensor shape and batch-size consequences.
Examples
Uses 64 × 64 and 256 × 256 images, 3 × 256 × 256 tensors, and resized anime-face or glasses-image datasets.
What you can do after
Apply resolution-aware preprocessing via PyTorch and adjust batch size when larger image tensors exceed available memory.

Read how Learn Generative AI with PyTorch explains resolution →

Local Image Generation

The amount of spatial detail in an image, commonly expressed through its pixel dimensions.

How it is used
Guides Stable Diffusion image generation toward sharper displays, complex editing, printing, wallpapers, and higher-quality training data.
How deep it goes
Explains pixel dimensions, fixed information, DPI, and latent upscaling workflows, but stops short of detailed upscaler internals.
Examples
Uses 512×512 images, 300-DPI printing, a 3440×1440 wallpaper, and a ComfyUI two-times latent upscale.
What you can do after
Apply it via A1111 or ComfyUI to choose targets, estimate print size, and upscale generated images.

Read how Local Image Generation explains resolution →

Also covered in

What to know first

Related concepts

Questions about resolution

What is resolution?

Resolution is the fineness with which values, measurements, or spatial details can be distinguished or represented.

Which Manning books explain resolution?

Build a Text-to-Image Generator (from Scratch), How Computers Make Books, Quantization and Fast Inference, WordPress Power Toolkit: Harness AI to build next-level websites, Learn Generative AI with PyTorch and Local Image Generation.

What should I know before learning resolution?

Manning books that teach resolution usually introduce digital image, pixel, diffusion model and image generation model first.

What is the difference between resolution and range?

Resolution is the fineness with which values, measurements, or spatial details can be distinguished or represented. The set or interval of values that a function produces or a representation can encode.

What kinds of resolution are there?

Kinds of resolution include screen resolution.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.