Almost every graphic design student has been told: “Always export web images at 72 DPI, and print images at 300 DPI.”
Here is the blunt reality: For digital web graphics, DPI is completely meaningless. A 1920×1080 image saved at 72 DPI, 300 DPI, or 10,000 DPI will look 100% identical on every computer monitor and smartphone in existence. In this guide, we break down why this myth started and what actually dictates sharpness on modern Retina screens.
📐 Resize Images by Exact Pixel Dimensions
Crate Resize Image lets you scale images by true pixel dimensions or percentages with aspect ratio lock and one-click ZIP download.
Launch Free Image Resizer→The Fundamental Difference: PPI vs. DPI
Pixels Per Inch (PPI)
Measures the physical pixel density of an electronic display screen. An iPhone 15 Pro screen has a hardware density of ~460 PPI. You cannot “save an image as 460 PPI”—PPI is a fixed property of the physical display hardware.
Dots Per Inch (DPI)
Measures the density of physical microscopic ink droplets sprayed by a commercial printing press onto physical paper. 300 DPI is the standard threshold where human eyes no longer see printer halftone dots.
Where Did the “72 DPI” Myth Come From?
In 1984, Steve Jobs and Apple launched the original Macintosh with a 9-inch black-and-white display. That specific cathode-ray screen physically measured 512 pixels wide across approximately 7.1 inches—which works out to exactly 72 pixels per inch.
Because Adobe software was originally developed on early Macs, 72 was hardcoded as a legacy metadata tag into file headers. Web browsers completely ignore this tag. A browser cares only about total pixel dimensions: width="1920" height="1080".
Retina Displays & the 2x/3x Asset Rule
Modern smartphones and 4K laptops feature Device Pixel Ratios (DPR) of 2x or 3x. A button styled at 400px wide on an iPhone screen actually requires an 800px or 1200px wide image asset to avoid looking blurry.
To ensure razor-sharp graphics on mobile screens, always export your graphic assets at double resolution (2x) and compress the file size using modern WebP or mozjpeg algorithms.