How an LED TV Handles Motion: Refresh Rate, Blur, and Processing

Two televisions built around the same panel, calibrated to the same brightness, can look like different products the moment something moves fast. A football match, a whip-pan across a stadium, a firefight in a game. That is where they separate. Almost none of the difference is explained by the letters L-E-D. It comes from refresh rate, the physics of the pixel, and a processing chip making dozens of decisions every frame.

Refresh rate isn’t frame rate

Start with the number everyone quotes and most people misread. A 120Hz panel redraws the screen 120 times a second. That is a property of the hardware, not the video. Most film is shot at 24 frames per second; a lot of television and streaming lands at 25, 30, or 60. If the source only carries 24 new images a second, a 120Hz panel cannot invent detail that was never filmed. It can only decide how to spread those 24 frames across its 120 refreshes.

That distribution hides a problem almost no salesperson mentions. Twenty-four does not divide evenly into 60. To fit film onto a 60Hz screen, sets fall back on a trick called 3:2 pulldown: one frame is held for three refreshes, the next for two, alternating forever. The uneven rhythm surfaces as judder: a faint stutter in slow, steady camera pans, the kind of shot where a stationary building drifts across the frame. Feed the same 24fps into a 120Hz panel and the arithmetic is clean: every frame gets exactly five refreshes, the cadence is even, and the judder eases.

That is the quiet reason a genuine 120Hz panel can help with film even though film never reaches 120fps.

Be skeptical of the large motion numbers on the box. Labels like “Motion Rate 240” or “Clear Motion 960” are marketing composites that fold in backlight tricks and processing; they are not the native refresh rate. In practice two native rates cover almost every set worth considering: 60Hz and 120Hz, with a small crop of gaming models now stretching to 144Hz over HDMI. If a panel can actually accept and display 120 frames a second from a games console, that is the specification worth paying for.

Why moving objects blur

Refresh rate feeds straight into the other half of motion: blur. Flat panels are “sample and hold” displays. Each frame is painted and then held, unchanged, until the next one replaces it. Your eyes do not hold still, though: they track the moving object smoothly across the glass. While the eye glides, the picture underneath sits frozen, so the retina smears that stationary frame into a streak. The longer each frame lingers, the longer the smear. At 60Hz a frame is held for roughly 16.7 milliseconds; at 120Hz, about 8.3.

Halving the hold time roughly halves this kind of blur, which is the most honest argument for a faster panel.

Persistence is only one source. The second is pixel response time, how quickly a liquid-crystal cell can physically twist from one shade to the next. If the crystals are slow, a fast edge drags a faint trail behind it no matter how high the refresh rate is. Makers push the transition harder with overdrive, but overshoot the target and you get the opposite artifact: a bright “inverse ghost” fringe. Panel type shapes this too. VA cells often smear dark transitions, the notorious black smear that shows up in shadowy scenes, while IPS usually settles faster but starts from weaker contrast.

Self-emissive panels sidestep this particular problem, because an OLED pixel switches state almost instantly; that is one reason self-emissive displays look cleaner in motion even at the same refresh rate.

Some sets attack persistence head-on by imitating how old tube TVs worked. Black frame insertion and backlight strobing flash the image and then blank it, cutting the hold time so the eye has less frozen frame to smear. It works, and it costs you. Blanking the screen part of the time drops peak brightness, and if the strobe rate is low or mismatched to the content, you can see flicker or a faint double image. It is a real tool with a real tax, and whether the crisper motion justifies a dimmer, flickerier picture depends on your room and your eyes.

The soap-opera effect

The most divisive processing feature tries to erase judder and blur at once by manufacturing frames that were never shot. Motion interpolation (motion estimation and motion compensation, if you want the textbook term) compares two real frames, guesses how each object moved between them, and paints a synthetic in-between frame. On live sport the extra smoothness can look genuinely good. On a 24fps film it strips out the texture that makes film feel like film: the image takes on the flat, hyper-fluid look of a cheaply shot studio drama, which is why people call it the soap-opera effect.

It also fails in visible ways. When two objects cross, or motion is too fast or too chaotic to estimate, the guess breaks and you see halos shimmering around moving edges or a brief warping smear. Every brand ships its own version under its own name (Auto Motion Plus, TruMotion, Motionflow, and the rest) and each is doing the same job with a different appetite for aggressiveness. My blunt advice: switch it off or near-off for movies, leave a light touch for sport if you enjoy it, and use Filmmaker Mode when a title offers it, since that mode disables interpolation and most of the other “enhancements” by design.

Lag, tearing, and game mode

Gaming rewrites the priorities. Here the number that hurts is not blur but input lag, the delay between pressing a button and seeing the result, and it is a different thing from response time, which people constantly conflate. Response time is how fast a pixel changes color. Input lag is how long the whole signal chain, processing included, takes to reach the screen. A television can have quick pixels and still feel sluggish, because its processing pipeline is adding delay before the panel ever gets the frame.

Game mode fixes that by stripping the pipeline down: most interpolation, noise reduction, and heavy scaling are shut off to push the signal to the panel as fast as possible. Pair it with variable refresh rate and you solve a second problem, tearing. When a console renders, say, 47 frames one second and 58 the next while the panel refreshes at a fixed pace, the screen can show pieces of two frames at once, split by a horizontal tear. VRR, sold as FreeSync, G-Sync compatibility, or plain HDMI 2.1 VRR, lets the panel shift its refresh moment to moment to match whatever the console is drawing, so frames arrive whole.

Auto Low Latency Mode is the convenience layer that flips the set into game mode automatically when it detects a console.

The processor decides the rest

Every stage above runs on the same system-on-chip, and this is where two sets with identical panels stop being equal. Almost nothing you watch is native 4K: a 1080p stream, a broadcast channel, an old disc, all of it has to be upscaled, and good upscaling reconstructs a plausible sharper image while bad upscaling just magnifies the softness or smears fine detail into wax. Streaming piles on a second job, because heavily compressed video arrives with banding across gradients, blocking in dark scenes, and mosquito noise clinging to edges; the chip has to clean that up without scrubbing away real texture.

Add deinterlacing for 1080i broadcast, cadence detection to spot film buried inside a video signal, and HDR tone mapping that squeezes a master graded for a brighter display down to what this panel can actually show without crushing the contrast in the shadows, and you have a stack of hard software problems solved either well or poorly.

That is why “which panel is it” is the wrong first question for everyday viewing. Two televisions can share a backlight, a resolution, and a refresh rate and still look a class apart, because one maker’s silicon upscales, denoises, and tone-maps with more care. When you compare sets for movement and real streaming, look past the backlight to the native refresh rate, the measured game-mode lag, and the manufacturer’s track record on processing. That is where real-world performance is actually decided.