Mouse dropping effect (phantom array) still exists at 144Hz [and up]

Advanced display talk, display hackers, advanced game programmers, scientists, display researchers, display manufacturers, vision researchers & Advanced Display Articles on Blur Busters. The masters on Blur Busters.
Post Reply
HOLECHIEN87
Posts: 7
Joined: 07 Jul 2017, 22:28

Mouse dropping effect (phantom array) still exists at 144Hz [and up]

Post by HOLECHIEN87 » 10 Jul 2017, 21:27

Image

see the cursor. i took this pic with a camera. i thought this phenome should not appear.. what the purpose of having 144hz if i see that?

can someone explain me if it's normal, if you'r tool improve it. i'm really confused right now i bought this monitor (xl2411z) for gaming and especially to avoid this kind of stuffs????

edit : i moved my cursor fast to have this result

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: Isn't this kind of things i should not have with a benq

Post by RealNC » 11 Jul 2017, 06:04

If you mean the gaps between each cursor position, then you need to compare it to 60Hz. Just taking a picture in 144Hz doesn't tell you anything.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

Haste
Posts: 326
Joined: 22 Dec 2013, 09:03

Re: Isn't this kind of things i should not have with a benq

Post by Haste » 11 Jul 2017, 06:20

Hi,

This phenomenon is called the phantom array effect.

Each cursor that you see correspond to one frame/refresh.

The faster the frame rate/refresh rate, the higher the number of cursors and the closer they appear together.

144Hz (while better than 60Hz) is still significantly too slow to prevent that issue when things are moving fast on the screen.

I'm afraid, until we get monitors that refresh at 4 digits refresh rates, we are stuck with that artifact.

This video illustrates it very well:

phpBB [video]

[Jump to 2m35s]

Notice how it gets better as the refresh rate gets higher. But even at 180Hz, it's still very noticeable.
Monitor: Gigabyte M27Q X

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Mouse dropping effect (phantom array) still exists at 144Hz

Post by Chief Blur Buster » 11 Jul 2017, 16:15

Unfortunately, the phantom array effect (stroboscope style stepping effect, also known as the "wagon wheel effect") is visible all the way to approximately 10000 Hz (ten thousand Hertz), depending on circumstances.

Image
(From lighting industry paper about flicker of fluorescent ballasts and ultra-high-frequency PWM)

While this is a lighting industry paper, it illustrates the unobtainium refresh rates needed to truly eliminate certain kinds of display artifacts (without things like artificially-added motion blur). Finite refresh rates (even 240Hz, 480Hz, etc) create display artifacts indirectly, such as http://www.testufo.com/eyetracking -- that improve gradually but doesn't disappear until unobtainium refresh rates -- diminishing points of return. But certain artifacts like mouse phantom arraying is still easy to see even on 240Hz displays, even if the distance between arrows at 240Hz would be 1/4th as much as at 60Hz.

The higher refresh rates you go, the better for the mouse effect. At 144Hz versus 60Hz, it's only 60/144ths the distance between mouse cursor position (in two consecutive refresh cycles) at 144Hz versus 60Hz. However, if you move your mouse pointer at 8000 pixels per second (a common flick speed -- a screen width in a fraction of a second), you would need 8000Hz to fill each pixel position with a mouse arrow, for a continuous blur rather than phantom array.

You do get many benefits of 144Hz, over 60Hz, however.
  • Reduced amount of motion blur, see http://www.testufo.com
    -
  • The gaps between mouse arrow pointers will still shrink, proportionally to the refresh rate.
    -
  • Higher Hz do have reduced input lag at higher refresh rates, see our input lag tests of different refresh rates at GSYNC 101: Input Lag and Optimal Settings. Although this is a GSYNC 101, it also tests the lag of non-GSYNC refresh rates too.
    -
  • In addition, your monitor might also include a Blur Reduction backlight for clearer motion, see Motion Blur Reduction Backlights for more info -- but that won't solve the phantom array effect. Most 144Hz and better monitors have a blur reduction feature.
NOTE ABOUT MOTION BLUR MATH: Going ever higher refresh rates, while has points of diminishing returns, has effects on motion artifacts go far beyond what most humans expect. Ignoring GtG additions to blur (very small factor on 1ms TN 144Hz monitors with 6.9ms refresh cycle visibility times) and ignoring strobe-based blur reduction -- on continuously-illuminated (non impulse/non strobed/non CRT) sample-and-hold displays -- the display motion blur scales exactly linearly with frame visibility time. Assuming the perfect zero-GtG sample-and-hold display. Even 0ms instant pixel response still creates motion blur due to the sample-and-hold effect. Modern TN gaming monitors (at least with good balanced overdrive) tends to roughly resemble the ideal Blur Busters Law of "1ms of frame visibility time equals 1 pixel of motion blur for every 1000 pixels/second of motion" -- found in 60Hz vs 120Hz vs LightBoost. As a result, 120 frames per second has half the motion blur of 60 frames per second. And 240 frames per second has one-quarter the motion blur of 60 frame per second. Display motion blur is mathematically proportional to length of frame visibility time. At low frame rates, this is visible stutter/judder. At high frame rates, the stutter/judder blends into motion blur. The higher the Hz you go, the thinner the motion blur becomes. For educational animations I've created about this phenomenon, see http://www.testufo.com/eyetracking and http://www.testufo.com/blackframes ... Frame visibility duration time is also why non-strobed OLED displays still have motion blur, even if their pixel response is more instantaneous than LCD. The only way to reduce motion blur during eye tracking situations, is to reduce frame visibility time -- either by increasing frame rates or refresh cycles (more frames) -- or by adding black periods between frames or refresh cycles (such as strobing). Now, however, for the phantom array effect -- aka mouse droppings -- the only solution to truly solve that, is to add WAY more fully visible frames (thousands of refresh cycles per second needed) or adding artificial motion blurring to the mouse arrow pointer (which isn't desirable).

TL;DR: The mouse dropping effect becomes less at higher refresh rates, but still exists even at extremely high refresh rates, even at several thousands of Hertz. That said, you get other important advantages of high refresh rates.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Mouse dropping effect (phantom array) still exists at 14

Post by Chief Blur Buster » 06 Nov 2017, 10:26

UPDATE

(I forgot to update this thread more than a month ago, might as well do it now)

Here is photographic proof.

We now have a 480Hz test that shows how refresh rate affects phantom array effects. We are the world's first website to photograph 480 Hertz for the mainstream. These are actual 240 Hz and 480 Hz photographs below:

Image

Image

We'll need kilohertz displays (>1000Hz) to make this effect disappear, unless we want to add intentional one-refresh-cycle worth of motion-blurring to mouse arrow pointers (not a good idea until around ~500-1000Hz).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

John2
Posts: 8
Joined: 14 Jun 2020, 17:56

Re: Mouse dropping effect (phantom array) still exists at 14

Post by John2 » 25 Feb 2021, 12:56

Chief Blur Buster wrote:
06 Nov 2017, 10:26
UPDATE

(I forgot to update this thread more than a month ago, might as well do it now)

Here is photographic proof.

We now have a 480Hz test that shows how refresh rate affects phantom array effects. We are the world's first website to photograph 480 Hertz for the mainstream. These are actual 240 Hz and 480 Hz photographs below:

Image

Image

We'll need kilohertz displays (>1000Hz) to make this effect disappear, unless we want to add intentional one-refresh-cycle worth of motion-blurring to mouse arrow pointers (not a good idea until around ~500-1000Hz).
Hey Mark thanks for taking the time to respond on Neogaf but I left a couple more questions for you so when you get the time.

Man I swear out of all the subject I have ever tried to learn in my life this subject here of diplay motion blur is one of the hardest I have ever tried to wrap my head around (apart from math of course I'd say math was the harderst), the more I look into it the more questions I keep having. So about this mouse stepping effect, I definitely never noticed it but after it was pointed out to me thanks to blurbusters I did what you said and stared at a fixed point on the screen and moved the mouse about and I guess I can barely see the stepping effect just barely. But why oh why does the stepping effect only happen to just the mouse? Shouldn't it happen to any single object on the screen that moves fast such as a car or just when the camera pans really fast, I mean you say the stepping effect happens when the cursor is moved really fast but by that logic the stepping effect should to be happening to anything that moves fast, and so one of the things I did was start scrolling the webpage up and down really fast while I looked at a fixed point and could not notice a stepping effect on anything on the screen, the only thing I noticed is what I notice all the time and thats the sample and hold effect AKA motion blur. But I did notice it on the mouse so whats going on here man why is this stuff so infinitely complicated? And so I know you said it goes away at 10,000Hz, when do you think civilization will have screens like that, 50 years from now?

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Mouse dropping effect (phantom array) still exists at 14

Post by Chief Blur Buster » 25 Feb 2021, 22:54

John2 wrote:
25 Feb 2021, 12:56
Man I swear out of all the subject I have ever tried to learn in my life this subject here of diplay motion blur is one of the hardest I have ever tried to wrap my head around (apart from math of course I'd say math was the harderst), the more I look into it the more questions I keep having.
Display science is much easier to understand when you have a 240Hz display because it allows you to see for yourself (60 vs 120 vs 240) of various artifacts by running specific TestUFO tests. Once you realize the behaviour, it becomes easy to mentally extrapolate.

Do you have any access to 120Hz or 240Hz display, with or without an optional strobing mode?
John2 wrote:
25 Feb 2021, 12:56
So about this mouse stepping effect, I definitely never noticed it but after it was pointed out to me thanks to blurbusters I did what you said and stared at a fixed point on the screen and moved the mouse about and I guess I can barely see the stepping effect just barely. But why oh why does the stepping effect only happen to just the mouse?
That's actually an assumption that is a myth -- you should read The Stroboscopic Effect of Finite Frame Rates.

Image

If you read the Stroboscopics article, you'll realize the effect occurs with all high-contrast motion, not just with the mouse arrow.

Testing The Stroboscopic Effect in a Game
  1. Load a game with high contrast objects
    ......Cyberpunk 2077 bright neon in black background
    ......Vertical dark-vs-bright boundaries during horizontal pannings.
    .
  2. Go to some place in FPS game with high contrast (superbright stuff next to superdark stuff like neons/streetlamps/etc) (or a bright sunlit-vs-dark shadow vertical wall edge in CS:GO)
    .
  3. Stationary-gaze at crosshairs while turning left/right. Force your eyes to stare ONLY at the crosshairs permanently for the duration of the test. Turn left/right very fast by swiping mouse left and right.
    .
  4. You will see the mousedropping effect whenever your eyes are at a different motion vector than the game's motion. For example, stationary gaze at crosshairs while background scrolls past.
Questions:
1. Do you have a recent VR headset? (any OLED or new 2020 LCD)
2. Do you have a 240Hz display?
3. Do you have a small, medium, and huge display that your computer can connect to?
4. Can any of your displays do a high-quality strobe mode?

If you don't have 1/2/3, it becomes more difficult to teach certain elements of display science, because the "aha, 2+2=4" easy teachable moments is seeing-for-yourself "how refresh rate doubling" scales (60Hz vs 120Hz vs 240Hz). Swich to 60 Hz, do a TestUFO, switch to 120Hz, do a TestUFO, switch to 240Hz, do a TestUFO, then it becomes easy to extrapolate what happens at 480Hz or 960Hz (and it's also experimentally verified that the guesses are correct, in the laboratory high-Hz displays) --

It is also precisely why 60fps YouTube videos are almost impossible to correctly teach some display-concepts. I need 120fps HFR (or ideally 240fps HFR) to properly teach some Blur Busters concepts in a proper quicker ELI5 format. When 240Hz iPhones becomes widespread, I can broadcast 240fps HFR to them using a special video player (YouTube doesn't support 240fps), so we'll see how the world improves in high-Hz education.

Once I show off a real 240Hz display to a classroom, some of the motion blur concepts become simple high school mathematics. It is actually hard to grasp head around it if you've been only staring at 60Hz LCDs or low-quality 120Hz LCDs (where GtG is too big to make 120Hz exactly half the motion blur of 60Hz). I also do paid classroom services from time to time.
John2 wrote:
25 Feb 2021, 12:56
Shouldn't it happen to any single object on the screen that moves fast such as a car or just when the camera pans really fast, I mean you say the stepping effect happens when the cursor is moved really fast but by that logic the stepping effect should to be happening to anything that moves fast, and so one of the things I did was start scrolling the webpage up and down really fast while I looked at a fixed point and could not notice a stepping effect on anything on the screen
Your executed very limited scientific experiment test variables.

- Also, most people don't pay attention until they're "trained" to this effect.
- Also, not everyone can see it (i.e. not picky about it; some people are more sensitive to it)
- Or there might be motion medical issues such as Conjugate gaze palsy, or Akinetopsia, or whatever.

For a beginner tester doing this:

To Amplify Experimential Observations With Web Browser:
  1. You need a longer observation time of multiple seconds.
  2. Try a super long webpage with 100 pageful of scrolling for a longer scroll observation.
  3. Try a webpage without images, to prevent webpage scroll stuttering.
  4. Try a lower refresh rate to begin with, like 60 Hz or 75 Hz (lower Hz = easier to see)
  5. Try a bigger display like desktop LCD, not a phone or tablet (bigger FOV = easier to see)
  6. Make sure your browser is GPU accelerated (don't use a 10-year-old laptop) for smooth scrolling
  7. Try automatic scrolling so you don't have to touch the mouse (let go of mouse) after you begin scrolling. This decouples eye-hand coordination-effect that can distract some people's brain from seeing the artifact.
Steps To See Stroboscopic Text Effect In Browser Scroll
  1. View this big web page www.xmpp.org/extensions/xep-0301.html (an old spec I made)
    It's about 100 pages tall with no images, perfect for this educational exercise.
    .
  2. Click the middle mouse button (not scrollbar) to switch your mouse into smooth autoscroll mode.
    .
  3. Your mouse cursor should look like this on both PC and Mac
    autoscroll.png
    autoscroll.png (2.96 KiB) Viewed 15322 times
    .
  4. Now move your mouse downwards. You'll see the page automatically scroll superfast smoothly.
    .
  5. Stare at center of webpage while it scrolls very fast for a few seconds.
    .
  6. You should see stroboscopic text effect.
    .
  7. If you still can't see it, repeat steps 1-6 with faster autoscroll speed.
    You can autoscroll faster by moving mouse further away from middle click location to autoscroll faster.
    You can reverse scroll by moving mouse above middle-click location.. Go back-and-fourth at different speeds.
    Additionally, try Dark Mode temporarily. Stroboscopic text is easier to see in dark mode. Windows Search Box->"Dark"
    dark-mode.png
    dark-mode.png (3.52 KiB) Viewed 15322 times
This is why effects are often easier to see on bigger displays at higher resolutions, thanks to Vicious Cycle Effect. What you sometimes don't see on a small phone can be amplified on a big display.

That is why maybe ~240Hz to ~480Hz-ish (random numbers, don't take my word) might be a final frontier for phones, while ~10,000Hz-ish (random number, don't take my word) might be a final frontier for 180-degree FOV VR headsets (IMAX screen strapped to your face). The bigger, the higher refresh rate is needed to retina-it-out. The wider-FOV, the higher the refresh rate is needed to retina-it-out. (distant tiny TV versus jumbotron VR strapped to your eyes). Etc. etc.

Should be obvious/easy to understand once you read the other reply -- hopefully, especially if you have a 240Hz display to run the educational animations to properly test the versus-animations to help your brain extrapolate the line beyond 60-120-240 (because it becomes obvious what happens at 480Hz or 960Hz). Once you've seen something at 60Hz then 120Hz then 240Hz, it becomes super-easy to guess what happens at 480Hz for certain motion tests. That's why my TestUFO PowerPoints (during travel) brings my 240Hz laptop for teaching purposes.

The bigger the FOV, the higher the refresh rate you need. This post is another good explanation why higher resolutions and bigger displays amplify the need for higher refresh rates.

Conclusion:

Eye tracking = eye movement EQUAL display motion speed = persistence blur (equalling frame visibility time)
Eye stationary = eye movement UNEQUAL display motion speed = stroboscopic effect
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Mouse dropping effect (phantom array) still exists at 14

Post by Chief Blur Buster » 27 Feb 2021, 22:11

John2 wrote:
25 Feb 2021, 12:56
But I did notice it on the mouse so whats going on here man why is this stuff so infinitely complicated?
Just wanted to check with you if you did the above test (scroll-testing that 100 page webpage).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply