[Monitor Electronics Hack] World first zowie XL2540 240hz 60hz singlestrobe Experiment with VRR-DyAc!

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.
elexor
Posts: 169
Joined: 07 Jan 2020, 04:53

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by elexor » 24 Nov 2020, 00:56

I would not recommend this setup it's very jank and a pain to manage 2 different monitor osd's and inputs. I'm just using it as proof that high refreshrate monitors 240hz+ are capable of really high quality 60hz strobing. Since no company makes such a monitor. I don't know of any lcd that does single strobe without any crosstalk. Soon we will have commercial monitors with single strobe support thanks to the chief of blurbusters.

User avatar
AddictFPS
Posts: 314
Joined: 15 Jan 2020, 14:13

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by AddictFPS » 09 Dec 2020, 09:38

Sadly new XL2546K also without 60Hz single strobe. Rtings review confirm range 100-240. This is a nightmare.

User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by Chief Blur Buster » 09 Dec 2020, 20:40

AddictFPS wrote:
09 Dec 2020, 09:38
Sadly new XL2546K also without 60Hz single strobe. Rtings review confirm range 100-240. This is a nightmare.
All 240Hz BenQ supports single strobe at 75-240Hz.

RTINGS currently only tests EDID reported settings, but if you create 75 Hz vis CRU, it supports 75 Hz single strobe.
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
AddictFPS
Posts: 314
Joined: 15 Jan 2020, 14:13

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by AddictFPS » 10 Dec 2020, 02:54

Thanks you, very useful !

alapsu
Posts: 23
Joined: 18 Jul 2019, 17:33

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by alapsu » 11 Dec 2020, 13:26

Chief Blur Buster wrote:
09 Dec 2020, 20:40
AddictFPS wrote:
09 Dec 2020, 09:38
Sadly new XL2546K also without 60Hz single strobe. Rtings review confirm range 100-240. This is a nightmare.
All 240Hz BenQ supports single strobe at 75-240Hz.

RTINGS currently only tests EDID reported settings, but if you create 75 Hz vis CRU, it supports 75 Hz single strobe.
Hey Chief,

If you're able, could you post an example (maybe a screenshot) of how one might do this in CRU?

I've successfully used CRU to achieve lower-crosstalk sub-240Hz strobing on my XL2546 (following advice found on this website), but I've never managed to get sub-100Hz strobing to work.

Is it possible that the Blur Busters Strobe Utility only works for >=100 Hz, but the monitor itself can actually strobe at >=75Hz?

elexor
Posts: 169
Joined: 07 Jan 2020, 04:53

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by elexor » 17 Dec 2020, 07:57

Update* I have tried a different route with different pro's and cons I can't implement a perfect solution. literally nowhere on the board I can find a sync signal. it's all high speed signaling beyond my knowledge. DP > realtek > lvds

My new method that only requires a dirt cheap arduino pro micro.

Hijack the original benq sync pulse it will only output that pulse at over 100hz+ if you change your refreshrate below that the pulse disappears.

I tried to generate my own pwm pulse on the arduino but alas with no reference signal the phase slowly drifts out of sync and requires constant adjustment to work.

What I can do is set my monitor to a refreshrate eg 120hz and strobe at half of that 60hz, can even do 60hz strobing at 240hz. taking full advantage of the monitors scanout speed. pixel overdrive is optimized for 240hz. (can this decrease inputlag?)

120hz to 60hz
Image

240hz to 60hz
Image


Pros:
full control over strobe length can do some crazy stuff like 0.250ms strobes :O
can still control phase
crosstalk is even better then my orginal hack. my original had some small crosstalk drift because i tiny difference in hdmi timing.
should work on any monitor with backlight strobing.

less crosstalk then a oculus quest oled. I tried some nes games with virtualdesktop at 60hz.

Cons:
sometimes strobe lands on the newer frame when you minimize the game which will add some crosstalk. It can be fixed by minimizing again. and stays perfect aslong as you don't minimize.
requires careful setup in games. to lock the framerate to stroberate
can't be used with devices that will only output a 60hz signal

I do want to figure out a way to strobe at native refreshrates but it I don't know how to keep phase from drifting with no reference.

Code: Select all

int on = B00100000;
int off = B00000000;
int strobePulseTime = 2000; //strobe duration in uS
volatile byte countPulse;
const byte interruptPin = 3;

void setup() {
  DDRB = on;
  pinMode(interruptPin, INPUT_PULLUP);
  attachInterrupt(digitalPinToInterrupt(interruptPin), blink, RISING);
} 

void loop()
{
  while(1) {
   if (countPulse == 2) //change to modify stroberate
    PORTB = on; //Turn backlight on.
    delayMicroseconds(strobePulseTime);
    PORTB = off; //turn off the backlight
    countPulse = 0;
   }
  }
}
void blink() {
  countPulse++;
}

elexor
Posts: 169
Joined: 07 Jan 2020, 04:53

Re: World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by elexor » 17 Dec 2020, 09:36

60hz strobing 60fps 240hz refreshrate
it's sharper in person my camera seems to struggle with focus while doing the pursuit freehand.
full size
Image


User avatar
Chief Blur Buster
Site Admin
Posts: 11653
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: [Monitor Electronics Hack] World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by Chief Blur Buster » 22 Dec 2020, 17:58

Fantastic stuff! Area51-worthy work, so this thread is now visible in Area51.

In theory, you could transmit the VSYNC signal using a PC app over USB cable and use heavy filtering to slew the VSYNC signal in a lag compensated manner (with phase adjustment feature).

The jitter can be filtered out. Teesny 4.0 supports 8000 Hz micropolling (0.125us) which is accurate enough for VSYNC monitoring to as resident system tray PC app (raised task priority + raised thread priority) if you make a tweak and you can timestamp the PC-side VSYNC signal.

You may be aware of my Terarline Jedi experiments where I used Windows API D3DKMTGetScanLine() which works at all times (because it works off the desktop handle of the primary monitor), it can tell you enter/exit VBlank status and there's a waitable call (one of those D3DKMT... calls) too, that you can run in a high priority process + high priority thread for precision. It would use almost no CPU but give a mostly jitter-free tick tock that can be processed / dejittered / timestamped / data transmitted to your Arduino (like strobe frequency and strobe phase)

You would need heavy filtering though and about a 1-second to sync to a new Hz, and ignore skipped signals.

In Arduino, you simply autosync on a fixed microsecond-exact interval, but use the received data over USB to slowly slew the phase back to the correct location. As long as phase slews only slowly, it won't be noticeable -- a 1% slow slew in phase is a 1% slow upward/downward shifts in strobe crosstalk which is not noticeable, so you can simply keep flywheeling in the Arduino and use the USB-transmitted VSYNC information as a guide to slew the VSYNC tick tock back into the approximate region again.

Both myself and ad8e have written VSYNC monitoring + filtering code (jitter filtering + missed VSYNC filtering) that uses a software equivalent of a flywheel to compute a microsecond accurate exact Hz tick-tock from a series of jittery VSYNC timestamps even with 10% missing timestamps. It takes about 1 second to get results accurate enough for strobing, so should be fast enough for your needs.

You will need a phase adjustment, to compensate for the latency, but you probably already have one -- as a strobe phase adjustment (like Blur Busters Strobe Utility's Crosstalk adjustment) -- which will suit this purpose just fine. You will need to readjust every refresh rate or every Vertical Total change.

You're doing something that helped start-off Blur Busters almost a decade ago in an Arduino scanning backlight :)
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!

elexor
Posts: 169
Joined: 07 Jan 2020, 04:53

Re: [Monitor Electronics Hack] World first zowie XL2540 240hz 60hz singlestrobe? Experiment.

Post by elexor » 22 Dec 2020, 18:55

Chief Blur Buster wrote:
22 Dec 2020, 17:58
In theory, you could transmit the VSYNC signal using a PC app over USB cable and use heavy filtering to slew the VSYNC signal in a lag compensated manner (with phase adjustment feature).

Both myself and ad8e have written VSYNC monitoring + filtering code (jitter filtering + missed VSYNC filtering) that uses a software equivalent of a flywheel to compute a microsecond accurate exact Hz tick-tock from a series of jittery VSYNC timestamps even with 10% missing timestamps. It takes about 1 second to get results accurate enough for strobing, so should be fast enough for your needs.
If you have any example code you could share It would be very much appreciated

Post Reply