would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Talk about overclocking displays at a higher refresh rate. This includes homebrew, 165Hz, QNIX, Catleap, Overlord Tempest, SEIKI displays, certain HDTVs, and other overclockable displays.
rexey64
Posts: 1
Joined: 22 May 2020, 00:10

would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by rexey64 » 22 May 2020, 00:15

so 'im problably dreaming, but i got this crazy idea.

i have a 4k 144hz monitor. i was wondering if it was possible to run it at 1080p, but "offsetting" each pixels in every group of 4 pixels by a 1/4 frame.

basically every pixel would still refresh at 144hz, but they would do so with a 1/4 of a frame delay between each other, making it a 1080p
576 hz.

i felt like this place was the right one to ask this question so... why would this be possible / not possible ?

overaller
Posts: 12
Joined: 18 Nov 2020, 22:53

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by overaller » 25 Nov 2020, 23:42

Even at 1080p the display refreshes all pixels, so it's limited by its design. So you can't force it to only refresh certain pixels. Search this forum and the blurbusters site for "scanout" if you're interested how the display is refreshed.

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

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by Chief Blur Buster » 26 Nov 2020, 07:47

rexey64 wrote:
22 May 2020, 00:15
so 'im problably dreaming, but i got this crazy idea.

i have a 4k 144hz monitor. i was wondering if it was possible to run it at 1080p, but "offsetting" each pixels in every group of 4 pixels by a 1/4 frame.

basically every pixel would still refresh at 144hz, but they would do so with a 1/4 of a frame delay between each other, making it a 1080p
576 hz.

i felt like this place was the right one to ask this question so... why would this be possible / not possible ?
While not possible in the horizontal dimension, it's possible to do in the vertical dimension.

Halve vertical resolution to double the refresh rate (up to law-of-physics limitations that prevents you from going any further). That's how the Zisworks 4K 120Hz display achieve 480Hz 540p, as seen at www.blurbusters.com/480hz

First, you have to purchase an appropriate FPGA, create a custom scaler/TCON, and write a custom computer program in very advanced languages. You are going to have to hardware-replace your monitor's scaler/TCON (the panel's motherboard) with your custom electronics board compatible with the existing LVDS ribbon cables connecting to the compatible LCD panel.

Read Making Of 480Hz to see how you can do this.

It's over the head of most people, but fascinating;
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!

BeamCoder
Posts: 3
Joined: 09 Jun 2021, 04:27

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by BeamCoder » 10 Jun 2021, 18:27

Chief Blur Buster wrote:
26 Nov 2020, 07:47
rexey64 wrote:
22 May 2020, 00:15
so 'im problably dreaming, but i got this crazy idea.

i have a 4k 144hz monitor. i was wondering if it was possible to run it at 1080p, but "offsetting" each pixels in every group of 4 pixels by a 1/4 frame.

basically every pixel would still refresh at 144hz, but they would do so with a 1/4 of a frame delay between each other, making it a 1080p
576 hz.

i felt like this place was the right one to ask this question so... why would this be possible / not possible ?
While not possible in the horizontal dimension, it's possible to do in the vertical dimension.

Halve vertical resolution to double the refresh rate (up to law-of-physics limitations that prevents you from going any further). That's how the Zisworks 4K 120Hz display achieve 480Hz 540p, as seen at www.blurbusters.com/480hz

First, you have to purchase an appropriate FPGA, create a custom scaler/TCON, and write a custom computer program in very advanced languages. You are going to have to hardware-replace your monitor's scaler/TCON (the panel's motherboard) with your custom electronics board compatible with the existing LVDS ribbon cables connecting to the compatible LCD panel.

Read Making Of 480Hz to see how you can do this.

It's over the head of most people, but fascinating;
If I use this method on a Asus 360hz monitor, can it reach 1000+Hz at 360p?

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

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by Chief Blur Buster » 10 Jun 2021, 22:39

BeamCoder wrote:
10 Jun 2021, 18:27
If I use this method on a Asus 360hz monitor, can it reach 1000+Hz at 360p?
Unknown. It depends on the panel's architecture.

I'd start simple and attempt 720p at half vertical resolution.

There are other refresh bottlenecks that are reached, plus LCD GtG will be smearing over multiple refresh cycles like a 33-50ms 60Hz LCD. Improvements 360Hz vs 720Hz might be excessively marginal without sufficient LCD GtG improvements. Even a 1000Hz mouse hurts the difference between 240Hz vs 360Hz, a 2000Hz+ mouse helps significantly.
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: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by elexor » 11 Jun 2021, 01:48

I really hope we get a 4k 240hz display that can also do 1080p 480hz in a similar method to the zisworks display. integer 1080p on a 4k panel looks just fine. especially if it's a small monitor.

BeamCoder
Posts: 3
Joined: 09 Jun 2021, 04:27

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by BeamCoder » 11 Jun 2021, 01:50

Chief Blur Buster wrote:
10 Jun 2021, 22:39
BeamCoder wrote:
10 Jun 2021, 18:27
If I use this method on a Asus 360hz monitor, can it reach 1000+Hz at 360p?
Unknown. It depends on the panel's architecture.

I'd start simple and attempt 720p at half vertical resolution.

There are other refresh bottlenecks that are reached, plus LCD GtG will be smearing over multiple refresh cycles like a 33-50ms 60Hz LCD. Improvements 360Hz vs 720Hz might be excessively marginal without sufficient LCD GtG improvements. Even a 1000Hz mouse hurts the difference between 240Hz vs 360Hz, a 2000Hz+ mouse helps significantly.
Hmmm interesting... I've started to be interested on FPGA's and this could be an interesting project to do.

What do you think is the theoretical minimum an LCD GtG time should be to not get smearing on a 1000Hz LCD? Also I found that the panel use by Asus is M250HAN03.0 which is around 5ms response time according to Panelook.

If LCD GtG is one of the bottlenecks, then with a 360Hz OLED monitor, it would be probably be possible to achieve 1000Hz if the panel architecture allows it.

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

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by Chief Blur Buster » 11 Jun 2021, 17:45

BeamCoder wrote:
11 Jun 2021, 01:50
Hmmm interesting... I've started to be interested on FPGA's and this could be an interesting project to do.

What do you think is the theoretical minimum an LCD GtG time should be to not get smearing on a 1000Hz LCD? Also I found that the panel use by Asus is M250HAN03.0 which is around 5ms response time according to Panelook.

If LCD GtG is one of the bottlenecks, then with a 360Hz OLED monitor, it would be probably be possible to achieve 1000Hz if the panel architecture allows it.
I notice your nickname BeamCoder. Does that have anything to do with racing the beam (raster interrupts)? If so, you might be interested in reading about my Tearline Jedi work...
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: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by Chief Blur Buster » 11 Jun 2021, 17:53

BeamCoder wrote:
11 Jun 2021, 01:50
What do you think is the theoretical minimum an LCD GtG time should be to not get smearing on a 1000Hz LCD? Also I found that the panel use by Asus is M250HAN03.0 which is around 5ms response time according to Panelook.
Real world LCD GtG (100%) needs to be less than a refresh cycle.

Manufacturer LCD GtG is measured on an oscilloscope from the 10% to 90%, as seen in Pixel Response FAQ: GtG versus MPRT[/url]. For that metric, that needs to be less than half a refresh cycle to prevent significantly adding motion blur (MPRT) beyond Blur Busters Law.

A perfect 0ms GtG display with squarewave pixel transition always follows Blur Buster's Law, where 1ms of visible frame time translates to 1 pixel of motion blur per 1000 pixels/sec motion. Any LCD GtG "retards" that a bit, but not noticeably until real world measured LCD GtG starts approaching a refresh cycle. You can read more Blur Busters white paper style articles in Blur Busters Area 51.

Also, different colors are slower pixel response. On the 360Hz monitor, certain colors take 2 refresh cycles to complete. You can see these in GtG heatmaps at places like RTINGS and HardwareUnboxed. That's why a 360Hz monitor is only about ~1.3x to ~1.4x clearer than a 240Hz monitor instead of the Blur Busters Law theoretical (1.5x) in motion blur (at framerate=Hz), it's already pushing the edge of LCD GtG limitations. In fact, using a 1000Hz mouse instead of an 8000Hz mouse also turns mouseturn motion clarity 240Hz-vs-360Hz difference to as small as ~1.1x because of high-frequency microstuttering blending into some additional 1-pixel motion blurring (like a fast-vibrating guitar string).

I have a suggestion to send to you, please email me at mark [at] blurbusters.com
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!

BeamCoder
Posts: 3
Joined: 09 Jun 2021, 04:27

Re: would it be possible to lit alternatively 1/4 pixels of a 4k monitor to quadruple the refresh rate at 1080p ?

Post by BeamCoder » 13 Jun 2021, 09:09

Chief Blur Buster wrote:
11 Jun 2021, 17:45
BeamCoder wrote:
11 Jun 2021, 01:50
Hmmm interesting... I've started to be interested on FPGA's and this could be an interesting project to do.

What do you think is the theoretical minimum an LCD GtG time should be to not get smearing on a 1000Hz LCD? Also I found that the panel use by Asus is M250HAN03.0 which is around 5ms response time according to Panelook.

If LCD GtG is one of the bottlenecks, then with a 360Hz OLED monitor, it would be probably be possible to achieve 1000Hz if the panel architecture allows it.
I notice your nickname BeamCoder. Does that have anything to do with racing the beam (raster interrupts)? If so, you might be interested in reading about my Tearline Jedi work...
Close enough, the name is about the Beam Synchronized Processor of the Amiga called "Copper". I'm aware of your work since I use WinUAE and currently learning how to program on the Amiga :D . I use the BeamRaced feature sometimes however it jitters on my 2010ish mid range laptop even at 4 slices...

Not to take away from your tearline algorithm, but I kind of thought it would be better to have a beam synchronized hardware like the Amiga and Atari 800 has instead of using a CPU thread to busy wait. Also since PC has many different configurations, it probably won't be time-exact, just like with my laptop for example so with a dedicated hardware it would be expected to be time-exact if you have one. But since there's no demand for this, your algorithm is the best option though.

Post Reply