For advanced readers, I insert my famous pandora boxes here and there.
I explain the double-image effects during stutters on sample-and-hold (non-ULMB)
Skip of behind-the-scenes tech stuff bores you.
Read if you're a fan.
<Blur Busters Technology Pandora Box>
Explanation Of Double Image Effects On Non-Strobed Displays During Erratic Stutter/Judder
Normally, non-blur-reduction modes don't create double-edge effects in sample-hold displays, but --
Certain framerate divisors -- like a cadence of 2/3 framerate (e.g. 160fps at 240Hz) -- can create a double-edge effect without using motion blur reduction.
The double image effect has long been a fixture of 3:2 judder by home theater videophiles. Many have speculated why there's a double image even on LCD (non-strobed). The explanation is surprisingly simple for those who already understand display persistence (ala
http://www.testufo.com/eyetracking) --
it's simply two simultaneously overlapping motion blurs (a thinner motion blur and a thicker motion blur).
This is because two very clear stutter harmonics are overlapping each other. 160fps at 240Hz generates a 2:1 pulldown effect creating a judder (erratic stutter). Long stutter, short stutter, long stutter, short stutter, alternating continuously.
144fps at 240Hz is more erratic but is not too far off the mark. So this can create prolonged screen persistence at one edge of the display motion blur, creating an impression of a duplicate image, thanks to increased persistence at one end of the stutter amplitude.
And you know how high-frequency stutter blends into motion blur?
Just like when you stare at
http://www.testufo.com#count=4 .... the fastest UFOs are just motion blur and the slowest UFOs are just stutter. The faster an edge vibrates, the vibrates is beyond human flicker threshold, and is just persistence display motion blur. Yep persistence = stutter. It's a continuum. Low-Hz guitar strings vibrate noticeably. High-Hz guitar strings vibrate so fast it's blurry. Stutter on high-Hz displays is _exactly_ this.
Persistence is stutter. Stutter is persistence. It's a continuum. Don't believe me? Look at
http://www.testufo.com/vrr then -- watch a slowly ramping frame rate on a variable refresh rate display and it's just Eureka like E=mc^2. Which is easily seen in variable refresh rate ramping animations like
http://www.testufo.com/vrr ... Once you see those animations, it's that simple!
Yep, both of the stutter harmonic frequencies blends into their two separate thicknesses of motion blurs.
High-frequency edge stutter vibrates so fast, the edge of moving objects simply blends to motion blur (like a high-Hz guitar string that vibrates so fast, it is blurry) as seen in
Blur Busters Law And The Amazing Journey To Future 1000Hz Displays. At high frequencies (framerates far beyond human flicker fusion threshold), stutter creates persistence (motion blur).
Erratic high-frequency stutter can create two different intensities of motion blur (e.g. 320fps at 480Hz) -- stronger motion blur at leading edge of motion. And fainter motion blur at trailing edge of motion. So even the erraticness can blend differently.
That's why you sometimes see an increased double-image effect during erratic stutter (e.g. 160fps at 240Hz) on non-impulsed displays. You're creating a pulldown, e.g. 2:1:2:1:2:1 -- where every other frame is repeated twice if you don't have VRR with perfect gametime=frametime=refreshtime sync for ideal VRR. Especially with ultra-accurate frame rate capping (like RTSS microsecond-accurate) to keep the pulldown consistent, can intensify the double-image effect for odd divisors of frame rate on fixed-Hz
When you enable VRR for 160fps@240Hz, the double-image effect diminishes or disappears. When you disable VRR for 160fps@240Hz and use VSYNC ON instead, the double-image effect reappears (2:1:2:1:2:1 pulldown...)
The 2:1 pulldown has two different stutter amplitudes. One stutter amplitude is twice as wide as the other stutter amplitude. If both stutter frequencies are high enough, both of them blends into motion blur. You now have a thinner motion blur overlapping a thicker motion blur. Voila. Another cause of double image effect that doesn't need strobing. See?
I should create new TestUFO motion tests that demonstrates simultaneous stutter harmonic frequencies / beat frequencies. It's a rather interesting science for ultra-Hz researchers. Often unimportant to many, but it very nicely explains the "double image effect" very nicely for non-impulsed displays (e.g. not using ULMB).
Once we're using 1000fps+ at 1000Hz+ all of this practically doesn't really matter, but we're still at the brunt of beat-frequency and harmonic-frequency motion artifacts even at 240Hz and 480Hz, in this current Refresh Rate Race to future Retina Refresh Rates.
</Blur Busters Technology Pandora Box>
That concludes a behind-the-scenes explanation. Enjoy.