LCD Motion Artifacts 101: Introduction

First published March 12th, 2013

Illustrations of different motion artifacts that affect clarity of moving objects on LCD displays, using the TestUFO Ghosting Test. The UFO objects were moving horizontally at 960 pixels per second on a 60 Hz LCD, and captured using a pursuit camera.

Ghosting

pursuitcam_ghostingGhosting is a trailing motion artifact. It is caused by asymmetric pixel transitions: Transitions from between two colors can be faster in one direction than the other direction. This motion artifact appears only on trailing edges; observe the yellow dome.

Coronas / Inverse Ghosting

pursuitcam_coronasCoronas are trailing artifacts caused by response time acceleration (RTC, overdrive, ASUS Trace-Free, BENQ AMA). Pixels can overshoot their final color value before bouncing back, causing bright inverse ghosting. See TFTCentral Overdrive.

Motion Blur

pursuitcam_motionblurMotion blur is symmetric for both trailing and leading edges. It is the blurring caused by eye tracking on continuously-displayed refreshes (sample-and-hold). This is not pixel persistence; but a different cause of display motion blur. See Science & References. Higher refresh rates (120Hz) or flicker (CRT and LightBoost) reduce this type of motion blur, see 60Hz vs 120Hz vs LightBoost.

PWM Artifacts

pursuitcam_pwmPWM artifacts look like repeated images, and can affect motion fludity. PWM represents Pulse-Width Modulation, a technique that many LCD backlights use for dimming screen brightness. Motion artifacts can appear at dim brightness settings. The above photograph was taken during Brightness setting of 0%. See TFTCentral PWM.

How Were These Images Captured?

Stationary Camera: Capture of Pixel Transitions

A stationary camera is good for photographing pixel transitions statically. However, it is not a very accurate representation of perceived display motion blur and motion artifacts:

stationarycam_persistence
Example: Stationary camera photo of a moving object on a display.

Pursuit Camera: Accurate Capture of LCD Motion Artifacts

Pursuit camera are used by display manufacturers for testing (e.g. MotionMaster, and other MPRT pursuit cameras). This is simply a camera that follows on-screen motion. These expensive cameras are extremely accurate at measuring motion blur and other artifacts, since they simulate the eye tracking motion of moving eyes.

Blur Busters has developed an inexpensive pursuit camera method which operates in conjunction with the Blur Busters UFO Motion Tests. It makes possible accurate photography of motion artifacts. The test at www.testufo.com/ghosting was used to take the pictures on this page. Blur Busters is the world’s first blog to utilize a pursuit camera for accurate capture of motion artifacts, in WYSIWYG format, as seen by the human eye.

Can All LCD Motion Artifacts Be Eliminated?

Nearly all LCD motion artifacts can be reduced or eliminated by scanning backlights and newer strobe backlight technologies including the popular LightBoost tweak used by high-end computer enthusiasts. Such displays successfully bypass pixel persistence by turning off the backlight during LCD pixel transitions, and precisely strobing the whole backlight only on fully-refreshed frames (see high-speed video). These displays produces CRT motion quality, with all perceptible motion blur completely eliminated.

Also See

See the next article in this series, LCD Motion Artifacts: Overdrive.

12 Responses to LCD Motion Artifacts 101: Introduction

  1. Pingback: Article on LCD Motion Artifacts! | The Blur Busters

  2. Pingback: Inexpensive motion-blur tracking camera being obtained | The Blur Busters

  3. haar says:

    I bought the vg278h because of your web site and the benefits of removing motion blur…

    a great program for showing motion blur on your monitor (if you have a strong graphics card) is the gimark benchmark from

    http://www.geeks3d.com/20121113/gputest-0-2-0-cross-platform-opengl-benchmark-furmark-lands-on-linux-and-os-x/

    a strong graphics card such as the gtx680 at 1241/660o on the GiMark will produce a frame rate of 120 fps (117 to 124 on that card)

    on the vg278h, lightboost on low(10%) it looks great, no motion blur.
    it is also great for showing how the lightboost 10% is “the setting”… the others add a bit to a lot of “sandyness” to the picture (look Closely and you can see the difference, between off, 10%, and 20%-100%)

    • Interesting stuff! Too bad LightBoost 10% is pretty dim (1.4ms strobes); One of us should mod a LightBoost monitor with 5 LED edgelight bars (even wide-gamut LED’s) going through a prism focussed through a panel’s edge. Bulky edgelight but 5x brighter for the same strobe lengths. Maybe add non-PWM dimming too. blurbusters.com/category/homebrew
      And maybe use even shorter strobe lengths for even clearer motion than LightBoost=10% while keeping a bright image. Electronically triggered from the pulse from the original LightBoost edgelight, could cause a supercharged LED strobe of 0.5ms; allow clearer than CRT motion.

  4. borf says:

    The pursuit camera appears to be working well. I wasn’t aware of the PWM artifacts. Post some pics soon.

    • The good news: Any standard off-the-shelf digital camera with manual exposure control! I use my existing Casio EX-FC200S camera (which I also use for my high speed videos) as my pursuit camera.

      I invented a cheap method of sufficent blogger-league accuracy. Following 960 pixels/sec motion on 24″ monitors at an accuracy of less than 1 pixel error margin. You can even see the LCD screendoor effect isn’t vertically blurred. It is very blogger/magazine/homebrew friendly! In fact, an improved rig could do even better. More details to be announced after I launch the new Blur Busters Motion Tests.

      EDIT: The tracking error margin of the images on this page is about +/- 1 pixel (0.25mm). However, I’ve since made a few tweaks and captured some new pursuit camera photographs, that have less than a 100 micrometer motion-object-tracking error! (camera moving at a speed +/- 0.1mm of the moving object on-screen). This is more accurate than expected.

  5. Pingback: Blur Busters: A great website about HFR displays - HFR Movies

  6. Black Octagon says:

    Interesting article. I was always under the impression that motion blur, ghosting and ‘streaking’ were one and the same thing.

    • Yes, they are different, with different causes. They can all overlap each other (e.g. ghosting combined with eye-tracking-based motion blur), but they all have different causes.

      Except, streaking is the same thing as ghosting — but the word is usually used to describes extreme cases of ghosting where the ghosting trail is very, very long (several refreshes long). Eventually, will need to obtain a very old LCD, or an overclockable LCD, and capture the streaking effect.

  7. Pingback: Anonymous

  8. Pingback: G-SYNC Monitors are PWM-Free | Blur Busters

  9. Pingback: RTings.com adopts Mark Rejhon Invention: Pursuit Camera Blur Test | Blur Busters

Add Comment Register



Leave a Reply