Kuycon G32P - 31.5″ 6K (6144×3456) IPS Black Monitor

[EugW]

I can only assume it does.

Just sharing my experience that my M4 iPP pumps 3840x2160/60Hz/HDR via a TB5 cable to the G32P (while the M4 MBA does not) 🤷‍♂️

Weird compatibility issue? I still don't understand though. Is your Mac rendering internally at 3840x2160 or at 7680x4320? IOW, when you say scaled, is it 2X scaled on the Mac and then downsampled for output at 6144x3456, or is it output at 3840x2160 and then scaled by the monitor?

From my M4 Mac mini, I can run 3840x2560 (3:2 monitor) / 60 Hz / HDR via a TB4 cable to my Huawei monitor, but that is the native resolution, so not scaled on the Mac.

I normally run at 2560x1707 though, so the resolution rendered is 5120x3414.



Also, you mentioned that you are getting 80 Gbps DSC, but if you have DSC, you don't need 80 Gbps for 6K / 60 Hz / HDR. Conversely, if you have 80 Gbps, you don't need DSC for 6K / 60 Hz / HDR.

 

[splifingate]

True HDR, right? verified with HDR video.

Looks like i am going to return lg 32 6k, and buy G32p instead. G32p is clearly the winner now.

The G32P says "HDR/On", and the iPP has the HDR switch (which is on).

I'm on the Spark right now (connected to the G32P via DP Alt Mode @ 6144x3456/60Hz) , , , I'll re-visit stats-for-nerds in a HDR video on the iPP tmrw.

 

[splifingate]

Weird compatibility issue? I still don't understand though. Is your Mac rendering internally at 3840x2160 or at 7680x4320? IOW, when you say scaled, is it 2X scaled on the Mac and then downsampled for output at 6144x3456, or is it output at 3840x2160 and then scaled by the monitor?

This was when I had the iPP connected.

I don't have a mouse/kb connected to the iPP, so all I was able to see was a 2nd screen (a screen to which I could move a window, yet not be able to interact-with it).

It was the G32P Menu which stated "3840x2160/60Hz" and "HDR/On".

My M4 MBA does not want to do HDR. I have not tried HDMI, as the Air has only USB-C ports. After I get the Office re-arranged, I'll move the G32P there, and (and, at least, try to) connect it to my M2 Max via HDMI.

From my M4 Mac mini, I can run 3840x2560 (3:2 monitor) / 60 Hz / HDR via a TB4 cable to my Huawei monitor, but that is the native resolution, so not scaled on the Mac.

View attachment 2569166

I normally run at 2560x1707 though, so the resolution rendered is 5120x3414.

View attachment 2569169

Also, you mentioned that you are getting 80 Gbps DSC, but if you have DSC, you don't need 80 Gbps for 6K / 60 Hz / HDR. Conversely, if you have 80 Gbps, you don't need DSC for 6K / 60 Hz / HDR.

I dunno Man . . . there are too many variables at-play for me to clearly know what's-what.

 

[EugW]

This was when I had the iPP connected.

I don't have a mouse/kb connected to the iPP, so all I was able to see was a 2nd screen (a screen to which I could move a window, yet not be able to interact-with it).

It was the G32P Menu which stated "3840x2160/60Hz" and "HDR/On".

My M4 MBA does not want to do HDR. I have not tried HDMI, as the Air has only USB-C ports. After I get the Office re-arranged, I'll move the G32P there, and (and, at least, try to) connect it to my M2 Max via HDMI.

Just out of curiosity, I tried my M4 MacBook Air and my M4 iPad Pro with a Plugable USB-C 3.1 —> HDMI adapter (which came free with my Plugable TB4 hub), connected to my LG C8 4K TV, so no Thunderbolt involved at all.

I get 3840x2160 (clear text) out of both of them, but only the iPad Pro has HDR. No HDR option on the MacBook Air.

My TV and HDMI dongle are both HDMI 2.0, not 2.1. HDMI 2.0 is limited to 18 Gbps.

It appears I’m getting 4:4:4 colour too, so 4K 3840x2160 / 10-bit HDR / 4:4:4 through USB 3 / HDMI 2.0.
I would have thought it is 60 Hz, but I can’t actually check that. It doesn’t feel choppy like I thought 30 Hz would, but I didn’t think 4K 60 Hz 4:4:4 HDR was possible over HDMI 2.0. However, the TV clearly indicates I’m using HDR, so maybe it’s 30 Hz after all.

 

[thenewepic]

I didn’t think 4K 60 Hz 4:4:4 HDR was possible over HDMI 2.0

yes, 4K60 4:4:4 requires 15.68 Gbps but HDMI 2.0 can provide only 14.40 Gbps so it's not possible.

 

[splifingate]

So, do you mean M4 Macbook Air doesn't show HDR option when connected to the monitor in 4K mode?

No HDR at all (so-far). I have tried both a USB-C -> DP, and a Cable Matters TB5 cable.

 

[EugW]

yes, 4K60 4:4:4 requires 15.68 Gbps but HDMI 2.0 can provide only 14.40 Gbps so it's not possible.

I assume you mean 4K60 4:4:4 HDR, because without HDR it’s only 12.5 Gbps.

Anyhow, assuming @splifingate’s M4 iPad Pro is putting out 4K60 4:4:4 HDR, that’s no big surprise for its Thunderbolt port, since it only needs 15.7 Gbps. No TB5 or DSC required.

I wonder if the 80 Gbps DSC being reported is just a weird glitch, and it’s the monitor doing the upscaling to 6K from the 4K input. In that context, a Thunderbolt 4 cable would also give the same result, but perhaps the monitor would report differently because TB4 provides different authentication.

 

[thenewepic]

I assume you mean 4K60 4:4:4 HDR, because without HDR it’s only 12.5 Gbps

yes, I mean 10 bit color which is required for HDR.
4K60 4:4:4 8 bit color requires 12.54Gbps.

M4 iPad Pro is putting out 4K60 4:4:4 HDR, that’s no big surprise for its Thunderbolt port, since it only needs 15.7 Gbps. No TB5 or DSC required

that's true, 4K60 4:4:4 10 bit works even with 4 lanes of HBR2 (introduced back then in DP1.2).

No HDR at all (so-far). I have tried both a USB-C -> DP, and a Cable Matters TB5 cable

did you try 1080p hiDPI resolution? Or 4K lowDPI?

 

[EugW]

thank you very much for this test!

Unfortunately, my worst fear came true. Your result proves that DSC doesn't work with this exact resolution - 6144×3456 in macOS when connected via DP1.4 so it means that users get 4:2:2 chroma subsampling instead of full RGB 4:4:4 on the monitor.
Personally I cannot call 4:2:2 chroma subsampling image retina quality.

Hmmm... @splifingate's picture actually looks to me like 4:4:4 but with motion blur or incorrect focus. Another shot may help.

 

[splifingate]

Just snapped this photo (with my iPhone 13-Mini @ 1x) of the rtings ref. image in Firefox (DGX OS 7 (basically Ubuntu w/perks); DP Alt Mode @ 6144x3456/60Hz/100% scaling):

 

[splifingate]

did you try 1080p hiDPI resolution? Or 4K lowDPI?

hmmm . . . not that I know. I'm deep into my studies on Autism Spectrum Disorder today, and now questioning whether this may be one of the many reasons why I can't seem to account-for all the variables in the equation

 

[thenewepic]

@splifingate's picture actually looks to me like 4:4:4 but with motion blur or incorrect focus. Another shot may help

what picture are you talking about? That's in your message? It looks like 4:2:2 to me.
If you mean post https://forums.macrumors.com/thread...x3456-ips-black-monitor.2459111/post-34232165 it's obviously 4:4:4 but it was taken on Linux OS which doesn't have the bug anyway.

hmmm . . . not that I know.

in the "Displays" section of macOS settings you should select "1920x1080" (this is "1080p hiDPI") from the list for the monitor. Please check if HDR option is visible with this resolution.

 

[EugW]

what picture are you talking about? That's in your message? It looks like 4:2:2 to me.
If you mean post https://forums.macrumors.com/thread...x3456-ips-black-monitor.2459111/post-34232165 it's obviously 4:4:4 but it was taken on Linux OS which doesn't have the bug anyway.

I mean the photo that I had included in my post, which came from this post:

https://forums.macrumors.com/threads/kuycon-g32p-31-5-6k-6144x3456-ips-black-monitor.2459111/page-19?post=34164502#post-34164502

It looks like 4:4:4 that is blurred, either by camera motion or lack of focus, or both. If you look at the white text, you can see just how blurred it is.

4:2:2 would look even worse.

Below is RTINGS’ 4:2:2 picture. Note that the other text lines above the red and blue are in perfect focus, yet the red and blue lines are still worse than @splifingate's.

 

[thenewepic]

It looks like 4:4:4 out of focus / blurred.

I disagree. I see a clear difference between that image and image in this post. So my conclusion is that the latter is 4:4:4 and the image in your post is 4:2:2.

 

[EugW]

I disagree. I see a clear difference between that image and image in this post. So my conclusion is that the latter is 4:4:4 and the image in your post is 4:2:2.

The new picture you linked is in much better focus. Like I said, look at the white text to compare the focus. Until @splifingate can provide a better picture in macOS, I think it’s a mistake to conclude that earlier blurred picture is 4:2:2, because the whole picture is blurred.

 

[thenewepic]

look at the white text to compare the focus.

The focus isn't great here but even with such non ideal focus I see patterns only on the colored sections where there's no uniformity in lines. I'm already convinced this is 4:2:2 but also agree - if we have better photo it would clear all doubts.

 

[splifingate]

The new picture you linked is in much better focus. Like I said, look at the white text to compare the focus. Until @splifingate can provide a better picture in macOS, I think it’s a mistake to conclude that earlier blurred picture is 4:2:2, because the whole picture is blurred.

I see it now (the new clarity of the white row).

To be sure, the G32P (the *er* focus in my first submission) is not matte.

I disagree. I see a clear difference between that image and image in this post. So my conclusion is that the latter is 4:4:4 and the image in your post is 4:2:2.

These are two photos taken by me; with the same camera (if you can call what my 13-mini houses such); same screen; same browser; different platform/soft.

If I read the room correctly, EugW was just quietly asking me to snap a new photo

 

[EugW]

I see it now (the new clarity of the white row).

To be sure, the G32P (the *er* focus in my first submission) is not matte.

Yeah, I removed that line later. I momentarily forgot I was in the Kuycon thread. 🙃

These are two photos taken by me; with the same camera (if you can call what my 13-mini houses such); same screen; same browser; different platform/soft.

If I read the room correctly, EugW was just quietly asking me to snap a new photo

Yes. I find that taking these shots can be difficult because of the low brightness and the need to maintain some distance from the screen so the phone can focus. A stable base helps though.

Anyhow, thanks for all the pix!

 

[splifingate]

if we have better photo it would clear all doubts

Part of me would enjoy spending the rest of the day gathering objective date assuaging such . . .

. . . but, I'm deep into my weekly coursework; sp. Autism Spectrum Disorder:

Assignment #1: write a detailed (original, and reflective) response for each of the following twelve questions relating to pp. 218-254 in Exceptional Children

Assignment #2: do a deep-dive into the individually-assigned Response Modality (mine is on "Reinforcement"), and in the Class Forum, share your written precis of this with the Community. Additionally, respond to at-least five other individual responses with accuracy, cogency and originality.

As for bugs: Gnome/X11/nVidia leases a warehouse...

 

[PaulD-UK]

@EugW
The YCbCr 4.2.2 encoding shows itself in the RTings test mostly in the vertical line 'pipe' symbols:
The '4' means every pixel gets full luminance information, parts of the image that include some 'whiteness' are very well displayed.
The '2.2' means that the Cb and Cr component information are is transmitted for every second pixel.**

With pure blue and red pixels, there is very little Luminance (Y) component, so you only see one pixel instead of two.
Where there is a brighter pixel that has a Cb or Cr component, one pixel will have colour, and the next will be only white.

This is what @splifingate has posted. The camera shake is irrelevant, the image shows clearly that Cb Cr components are only being shown in every second pixel - when there isn't a bright luminance component.

The 4.4.4 images clearly don't show this artefact,

**Edit: My reply is simplified, because in the RTings image every coloured pixel in the rows of pipes is followed by two black pixels.
But they have constructed their image in such a way so as to clearly display the component compression algorithm - by ensuring some of the pixels that have Cb or Cr information fall on a black area of the design.
The following coloured pixel then has a luminance of zero, so any 'colour' doesn't get displayed. 😉

 

[EugW]

@EugW
The YCbCr 4.2.2 encoding shows itself in the RTings test mostly in the vertical line 'pipe' symbols:
The '4' means every pixel gets full luminance information, parts of the image that include some 'whiteness' are very well displayed.
The '2.2' means that the Cb and Cr component information are is transmitted for every other pixel.

With pure blue and red pixels, there is very little Luminance (Y) component, so you only see one pixel instead of two.
Where there is a brighter pixel that has a Cb or Cr component, one pixel will have colour, and the next will be only white.

View attachment 2569459

This is what @splifingate has posted. The camera shake is irrelevant. the image shows clearly that Cb Cr components are only being shown in every second pixel - when there isn't a bright luminance component.

The 4.4.4 images clearly don't show this artefact,

That makes sense and you and @thenewepic may be right, but it'd still be nice to see an updated pic when @splifingate isn't swamped with work.

 

[oKUtItyp]

Just snapped this photo (with my iPhone 13-Mini @ 1x) of the rtings ref. image in Firefox (DGX OS 7 (basically Ubuntu w/perks); DP Alt Mode @ 6144x3456/60Hz/100% scaling):

Don't use Firefox. Open the .png in Preview with 100% zoom is the best, as I found Firefox has issue when displaying png in hidpi mode.

 

[splifingate]

I momentarily forgot I was in the Kuycon thread. 🙃

Aye; I, too, dream of the day when we have 15m x 12m wall-displays using sub-mm diodes/'whatevers'....

 

[splifingate]

@EugW
The YCbCr 4.2.2 encoding shows itself in the RTings test mostly in the vertical line 'pipe' symbols:
The '4' means every pixel gets full luminance information, parts of the image that include some 'whiteness' are very well displayed.
The '2.2' means that the Cb and Cr component information are is transmitted for every second pixel.**

With pure blue and red pixels, there is very little Luminance (Y) component, so you only see one pixel instead of two.
Where there is a brighter pixel that has a Cb or Cr component, one pixel will have colour, and the next will be only white.

View attachment 2569459

This is what @splifingate has posted. The camera shake is irrelevant, the image shows clearly that Cb Cr components are only being shown in every second pixel - when there isn't a bright luminance component.

The 4.4.4 images clearly don't show this artefact,

**Edit: My reply is simplified, because in the RTings image every coloured pixel in the rows of pipes is followed by two black pixels.
But they have constructed their image in such a way so as to clearly display the component compression algorithm - by ensuring some of the pixels that have Cb or Cr information fall on a black area of the design.
The following coloured pixel then has a luminance of zero, so any 'colour' doesn't get displayed. 😉

This is a crop from the previously-shared DGX screen:

https://forums.macrumors.com/attachments/g32p-rtings-1x-jpg.2569420/

Just an iPhone 'photo' I took.

This being shared: while the G32P is an exceptionally nice screen, I am not feeling the urge to spend USD1,9K on another. 6Kx32" might be enough for near-term single-screen work!

Next step is to wrench my attention from coursework (primary to my fascination with the Spark) and take oKUtItyp's advice to get the rtings ref. image into Preview for further review on the MBA/M2Max.

 

[EugW]

This being shared: while the G32P is an exceptionally nice screen, I am not feeling the urge to spend USD1,9K on another. 6Kx32" might be enough for near-term single-screen work!

I used to run dual 27" screens (5K 27" iMac plus a 2.5K 27" iMac used as an external monitor). The "break" between the two screens was right in the middle of course. While that was definitely workable, it was also annoying. Plus I didn't need the full width of dual 27" screens.

I also tried using a single 30" 2.5K Cinema HD Display, and the width was OK for me, just enough of an increase in width over a 27" but of course the text quality was not good at 101 ppi. I then tried a 31.5" 4K screen, and while the width was good, again the text was poor at 140 ppi. I eventually went to a single 164 ppi 28" screen and decided to wait for 32" 6K pricing to come down. Unfortunately, that took 3 years. Now I await my pre-order for the 31.5" LG 6K, to be shipped in November.

BTW, if I'm understanding @oKUtItyp correctly, s/he claims to get 4:4:4 HDR 60 Hz on the LG 6K through Thunderbolt 4. However, I haven't seen an RTINGS test image yet confirming or refuting that for the LG 6K in HDR.

I'd probably use HDMI 2.1 anyway, but it'd be nice to have full 6K 4:4:4 60 Hz HDR through Thunderbolt 4 too just in case.

Next step is to wrench my attention from coursework (primary to my fascination with the Spark) and take oKUtItyp's advice to get the rtings ref. image into Preview for further review on the MBA/M2Max.

👍