Test Patterns: The Classic Monitor Test Charts and What Each One Reveals
From SMPTE color bars to the dot matrix and crosshair, the classic test patterns expose the most stubborn display problems in the simplest form: hue errors, banding, overscan and broken pixel mapping. This page explains what to look for in each pattern and how to read it — then ties everything into a five-minute full check.
SMPTE color bars: what each bar checks
The SMPTE color bar pattern is the television engineer’s classic: seven vertical bars at 75% amplitude — white, yellow, cyan, green, magenta, red, blue — with a reversed strip and a stepped gradient at the bottom. Each bar lights exactly two of the three subpixels at roughly 75% while the third stays off, so any hue error maps directly onto a leaking or starving channel. A red bar drifting toward orange betrays the green channel adding light it should not; a blue bar pulled toward violet or a muddy magenta points at another combination. Judge the bars at arm’s length in a dim room: two bars that should look clearly different but blend together are the fastest signal that a channel is off.
Inside each bar the field should be perfectly flat. Visible texture, vertical striping or slow brightness drift betrays dithering or FRC of an 8-bit panel pretending to be 10-bit, or uneven drive from the scaler. The bottom of the pattern carries the fine print: patches of saturation and luminance plus a black-to-white gradient strip — the banding detector. Smooth means clean gray levels; steps mean lost bit depth. Cross-check steps with the grayscale test before blaming the panel. Switch to the full-strength color bars and the pattern becomes a saturation check: cheap panels clip here, red flattening into a solid slab, yellow bleeding toward white. Hues that stay stable between 75% and 100% indicate honest processing; a hue that is wrong in both patterns sends you to the color calibration page.
Grid, dot matrix and crosshair: geometry and pixel mapping
The alignment grid answers what bars cannot: is the image centered, complete and straight? White lines should run perfectly vertical and horizontal with no pincushion or barrel bowing and equal margins on all four sides — cropped edges mean the TV or monitor is applying overscan and rescaling the signal. Color fringes where red, green and blue no longer land on the same spot are convergence error, common on projectors and aging CRTs, and the grid magnifies it line by line.
The dot matrix goes one level finer: one dot per pixel, ideally crisp and separated at normal viewing distance. Dots blurred into a haze mean non-1:1 scaling (the resolution check helps identify the panel’s native timing), while a soft patch in one corner betrays focus or sharpness uniformity problems. The crosshair is the simplest pattern and the one you will use most: projectors square up keystone and centering with it, multi-monitor setups align bezels and heights by matching the center mark across screens. Run the five-minute full check in order: screen test for dead pixels and uniformity → the 64-step gray ramp for banding → both color bar patterns for hue and saturation → grid, dots and crosshair for geometry. Fix settings first, draw conclusions after.
A five-minute full check that covers everything
Warm the display up for ten minutes, set the GPU output to full-range RGB, and put the monitor in its Standard or sRGB picture mode — presets like “Vivid” defeat any pattern. Then work through the sequence: uniformity and dead pixels at full black, white and the primaries; the 64-step gray ramp, where sixty-four distinct, evenly spaced steps from black to white are the goal. Steps that merge signal lost bit depth — often limited RGB range or a dithered 8-bit panel; steps wearing a color cast reveal channel mismatch in the midtones.
Judging order matters: fix settings first, because a wrong setup mimics a hardware problem. A merged step that disappears after disabling dynamic contrast was never the panel. Only after the settings pass should you cross-check with the specialized pages: the contrast test for crush and bloom, the grayscale test for tinted neutrals, and the color calibration patterns if a hue refuses to sit still. Ten minutes, every pattern, no guesswork — and on the dot pattern, rainbow moiré at non-native resolution is itself the diagnostic: scaling is active, not that the screen is faulty.
Frequently Asked Questions
What are SMPTE color bars used for?+
They are the broadcast-standard reference for hue and saturation. Each of the seven 75% bars exercises two of the three channels, so a hue that drifts toward its neighbor directly identifies which channel is over- or under-driving. The lower strip adds saturation and luminance steps plus a gradient for catching banding.
Why 75% amplitude instead of 100%?+
The 75% bars leave headroom so errors stay visible instead of being clipped. The 100% bars show the top of the range — where cheap panels saturate and colors bloom. Comparing both is a quick test of honest signal processing.
What should the 64-step gray ramp show?+
Sixty-four distinct, evenly spaced steps from black to white. Steps that merge signal lost bit depth, often from limited RGB range or a dithered 8-bit panel; steps with a color cast reveal channel mismatch in the midtones.
Why does the dot pattern shimmer or show moiré?+
A one-dot-per-pixel pattern beats against the panel grid when the image is not at native resolution, producing rainbow shimmer. The shimmer is the diagnostic: scaling is active, the screen is not faulty. At native resolution it disappears.
Can I use these patterns on a TV or projector?+
Yes — that is where the grid and crosshair earn their keep. Turn off every “fit to screen” and overscan option, project the alignment grid, then use the crosshair to center and square the image. Multi-monitor users can match the crosshair across displays to align them.