Yvetox

  • May 10, 2024
  • Joined Oct 16, 2023
  • I compiled some info back in the day on mbp 16 2019 5500m, if you want you can dig through

    Helpful links to disable ditherings:

    macos disable dithering

    https://ledstrain.org/d/268-osx-dithering-help-wanted/171

    If you are using intel based laptops with windows you can try ditherig.exe to disable dithering in windows

    https://kawamoto.no-ip.org/henteko/myapp_en.html#application

    If you are using windows with AMD videocards you can disable dithering by changing following registry keywords on the path of 2d driver(can be found inside amd catalyst or adrenalin drivers):

    "DP_DisableDither"=dword:00000001

    "Embedded_DisableDither"=dword:00000001

    "HDMI_DisableDither"=dword:00000001

    "TMDS_DisableDither"=dword:00000001

    You can find instructions here:

    https://ledstrain.org/d/1093-temporal-dithering-issues-with-laptops/5

    • simplex The benq monitor is bought new from reputable seller. The suggestion regarding exchanged backlight might hold some truth to it.

    • simplex The backlight ON time differs because this monitor has a sensor named ECO Sensor which disables the screen if human is absent from it for 15 seconds. Acutally very useful feature, I would love it on the monitor which doesn't burn my eyes out

      jordan I have tested on all modes of Response times, there are no BFI possible at all with this model as far as I know.

    • Greetings Everyone, let me share my findings here.

      TLDR: Found that Benq monitor with supposed DC dimming is actually pulsating with 47 Hz bright flashes which are not detectable on Iphone slo-mo.

      Let me preface that I started having issues with strain caused by monitors not so long ago - the first monitor that caused it with 100 percent is HP z24i g2
      I have myopia (-6 dioptries) and little astigmatism. The list of issues and usable electronics might depend on your specific eye disorders, so read them with caution.

      The symptoms it caused -

      • blood vessels popping, in terms of direction there was a visible path from the center of the iris to the nose but it also was everywhere on the eye, just generally eyes were horribly red.
      • feeling of tension in the eye muscles.
      • feeling of soreness, dry eyes, like I had sand in the eyes.
      • Just general eye fatigue which was preventing me from working

      Symptoms stayed for 2 to 3 days until I recuperated by limiting eye fatigue and moving to the countryside.
      My list of useable electronics from the easiest to use to hardest
      Iphone XR/11 & CRT monitor with 85 Hz > MBP 16 2019 > Steamdeck IPS > ps vita > New Nintendo 3ds XL > Dell P2421D > LG C2 (on the lowest brightness setting) > Benq BL2420PT > HP z24i G2(the absolute worst)

      Now, I started noticing the same issues and same feeling in the eyes right when I got this HP. It was a trigger point of sorts, from which I started researching this question. I have tried to find myself "good" monitor after that, and spent quite a lot of time and effort digging the internet until I found good things regarding Benq BL2420PT. Considering that out of ALL the options it was the only monitor that I could buy as new I stopped my search on it. After getting it I noticed some strain, but it appeared better than other screens so I tried to push forward with it - it also was relatively gentle in the sense that it did not cause immediate discomfort. The effect of its work is cumulative as a lot of folks on this forum know.

      Fast forward few months of usage and I decided to use it for prolonged periods of time few days in a row and got a horrible symptoms compared to those that HP monitor gave me. It caused me to investigate this issue further, until I found a definitive answer for the cause, and I want to laid it out for everyone to see.

      When I bought the monitor I attempted to weed out a good model using multiple tests that would eliminate possible causes for eyestrain. You can find test methodology here:
      https://ledstrain.org/d/2503-personal-compilation-of-information-from-this-forum-and-some-others/3

      This Benq BL2420PT passed most of these things, which lulled into false sense of security about it, especially considering that folks around internet were positive about this specific model.

      My initial search for the causes led me to check the panel used in the monitor. According to this website
      https://www.displayspecifications.com/en/model/140a738

      this was the panel that was supposed to be used (and which was praised by the people)
      AOU M238DAN01.3
      https://www.panelook.com/M238DAN01.3_AUO_23.8_LCM_overview_24260.html

      And I found that my monitor used other panel instead going to the service menu:
      BENQ-BL2420PT-that-i-bought

      LG LM238WQ1-SLA1
      https://www.panelook.com/LM238WQ1-SLA1_LG%20Display_23.8_LCM_overview_27444.html

      So I already established that the panel that monitor used is different from what it should have been.

      Now to the juicy part. Since I started to reinvestigate the causes I bought myself an OPPLE Light Master Pro/ III tester. Using it inside a black box with the screen shining at it I was able to measure a lot of screens used daily, as well as light bulbs (but thats another question). Long story short, here are the results:

      As you can see I measured strong brightness pulses roughly every 22 milliseconds. If you divide 1000 / 21,27 = 47,01 Pulses per second = 47 Hz.
      These pulses supposedly have stroboscopic effect on humans which are harmful as per multiple research papers. This bright pulse persists on all the modes, and all the levels of brightness, but the dip is less severe on a higher brightness.
      Also, my assumption is that a bright environment lowers the effect of these pulses compared to dark room, therefore people recommend to use screens at well lit environment.

      Now I need a help from the community to understand that deeper, but using Iphone slo-mo camera I cannot see the flickering, and my suggestion is that Iphone slo-mo shutterspeed of 240 hz does not cover 47 hz splashes as it is a prime number and out of all the 240 "snapshots" that camera takes it doesn't see the dip in brightness. Its also possible that iphone camera adjusts brightness sensitivity rapidly enough that you cannot see the flickering - the monitor backlight does not shut down completely, it merely lowers the brightness to roughly 25 lux from the highest point of 200 lux. Again, my assumption is that this trick is good enough to fool most of slo-mo cameras that people have on hand (iphones duh), and it might be used in multiple new monitors out there, mimicking flicker free expirience.

      To close the topic - I think that lot of you who had the eye strain issues could now potentially target very specific cause. I was very relieved when I realized that the possible causes for my issues are not hypochondriacal in nature, and could be measured by the equipment.

      • GBowler

        Boot within bootcamp. I am using official blue (corporate) drivers, not adrenalin one. If you go to Software & Driver details you will see 2d driver filepath that will look something like LOCAL_MACHINE\SYSTEM\ControlSet001\Control\Video{2D2A7876-F8BF-11E8-A4F6-848B33DFC0AC}\0000
        it means that you need to open regedit.exe and go to the path listed there, and add 32-bit DWORD keys with names like "DP_DisableDither", and set the value of these keys to "1". This is what the finished table will look like:

        "DP_DisableDither"=dword:00000001

        "Embedded_DisableDither"=dword:00000001

        "HDMI_DisableDither"=dword:00000001

        "TMDS_DisableDither"=dword:00000001

        If you did everything right you reboot and set background image that can be used to determine dithering like red backgound in this post.
        https://ledstrain.org/d/1093-temporal-dithering-issues-with-laptops/16

        If you see banding it means that dithering is disabled. MBP screen "passes" the pixel enversion tests in this setup without flickering much, but still flickers slightly on one of the modes.

        • The following text is a compilation of themes found in both english/russian/ukranian speaking sources related to the display strain issue, as well as themes started on the ledstrain.org forum that I formed for personal use but adapted to be like an article.

          Symptoms that I personally felt:
          Twitching discomfort sensations in the eyes
          Red eyes
          Dry eyes
          Blood vessels popping from the center of the iris towards point of the nose.
          Troubles with focusing on some distance after looking at the screen for prolonged period of time (might be just related to temporary accommodation)

          Devices that mitigated the strain:

          * CRT monitor (NEC with diamondtron tube)

          * Iphone XR

          * Steamdeck

          * Macbook pro 16 2019 with 5500m in bootcamp mode with dithering disabled for amd drivers in the registry.

          Things that can cause strain:

          * PWM - can be checked using pen waving method. Alternatively can be filmed with Iphone 240fps mode. Instructions can be found in the internet.

          * Blue Light (not really a concern compared to other factors, but seems to be most prominent on cheap IPS panels, can use color filters, described in other threads)

          * Dithering algorithms (2 realisations - FSR dithering and temporal dithering. Out of them 2 temporal dithering is much more prominent)

          * Bit depth of GPU (can play a role because if you wanna display true 8bit and your monitor supports it but GPU for some reason don't you will not see the difference)

          * Display brightness - if its too high you can have issues, try to use automatic regulation if possible, and lower display brightness to the minimum if possible. 120 nits is enough in the dark room.

          * On windows DWM.exe can add some processing to the image seen. (Can be seen when setting red image wallpaper. You should see banding that shows that the image is reproduced within colourspace limitation, but if you look at the same image from browser it will appear without banding, which means that browser apply some sort of dithering/smoothing to the image )

          # How to test for display things

          1. EIZO - web test for general image unpleasantry, 1st image pattern
          https://www.eizo.be/monitor-test/
          - if any of squares shimmers heavily you probably can already discard monitor as problematic.

          2. To test image artifacts you need to display pixel perfect test image. For built-in OS display you need to open images with some application that will diplay 1:1 image as it is.

          If inside windows 10/11 - image viewer by default displays 1:1 pixel image. You only need to change to max display resolution and 100 percent scaling in monitor settings.

          If inside linux - Ubuntu with max resolution will open images 1:1 - good solution. SteamOS on steamdeck appear to open images 1:1 with Gwenview as it passed all the checks.

          If inside macos - you need to change resolution to max supported one. On macos Sonoma go to "System settings > Advanced > Check the box 'Show resolutions as list' > return to the list of resolutions > Flip the switch 'Show all resolutions' > Find the resolution that is max supported by your panel. You will need to check which resolution your panel supports first as for me it shows avaliable mode that is higher than what panel can display (probably solution to connect to external TVs/Monitors). You can use webbrowser with tested viewport that match your resolution to open images, if you configured the browser as described lower.


          Images that you want to check:


          Chroma subsampling - you should clearly see 4:4:4 and almost dont see 4:2:2 (You should only see red and blue tint on the top and bottom of numbers, and body of numbers should be somewhat "shifted"against the backdrop but not be visible from the distance) If you see 4:2:2 it means that image is not displayed using full bits of depth, and chroma subsampling is the issue.


          Pixel inversion test - none of the squares should shimmer. For me I found that most troublesome ones are 3rd, 5th, and 7th. On old TFT panels it seems that 4th is causing issues.


          3. Web based tests - you first need to ensure that browser displays pixel perfect image. Use mosilla if possible, as it doesnt apply any dithering to the image (requires confirmation) Just to be sure I performed tests with different browsers and there was a perceptible difference between them. For these tests ensure that screen resolution is the same as viewport when entering fulllscreen for browser.

          https://whatismyviewport.com/

          The screen resolution should match 1:1 your display resolution, minus some pixels taken by the Browser bar at the top. If you enter fullscreen mode (F11 on most browsers) you should see that viewport changed to the same resolution as the display resolution. Next step is performing pixel inversion tests.


          Pixel inversion test

          http://www.lagom.nl/lcd-test/inversion.php

          Go to the fullscreen test and switch between different display modes in fullscreen inside browser. The screen that causes me the least amount of eyestrain (steamdeck) perform well in all tests, shimmering slightly with only one of them. On the other hand displays that I found uncomfortable shimmered/flickered heavily on these.


          Pixel inversion test on blurbusters

          https://www.testufo.com/inversion#pattern=uniformity2&ppf=5&fullscreen=1

          Even the best screens display some colorshift on these for me. Steamdeck screen appears to "lighten up" once per second on pixel uniformity extended edition test but otherwise remains static color. Monitor that I find relatively comfortable shifts its colors to green on the second line and alternate between blue/orange on the third line, so this test is not the end of the world.


          Blur test

          https://www.testufo.com/

          You probably want your monitor to pass blur tests, but pixel response seems to be relatively unimportant for me. Both steamdeck and MBP 16 2019 have some troubles with pixel response but they appear somewhat comfortable.

          Helpful links to disable ditherings:

          macos disable dithering

          https://ledstrain.org/d/268-osx-dithering-help-wanted/171

          If you are using intel based laptops with windows you can try ditherig.exe to disable dithering in windows

          https://kawamoto.no-ip.org/henteko/myapp_en.html#application

          If you are using windows with AMD videocards you can disable dithering by changing following registry keywords on the path of 2d driver(can be found inside amd catalyst or adrenalin drivers):

          "DP_DisableDither"=dword:00000001

          "Embedded_DisableDither"=dword:00000001

          "HDMI_DisableDither"=dword:00000001

          "TMDS_DisableDither"=dword:00000001

          You can find instructions here:

          https://ledstrain.org/d/1093-temporal-dithering-issues-with-laptops/5

          dev