pg27aqdm screen flicker when shutting down computer&switch source from pc, at 1440@240hz@10bpc

[dfr1238]

Update: ASUS customer service has sent me the MCM104S firmware specially compiled for this problem. After the update, this problem no longer occurs.

phpBB [video]

Kinda look like this, is normal? and does it effect panel life-span?

Screen
Model: PG27AQDM
Firmware version: MCM104, update firmware via PC.
Connection: DisplayPort(1.4)

Sec-screen
Model: Optix MAG274QRF-QD
Connection: DisplayPort(1.2a)

Computer Equipments

OS : Windows 11 x64 Home Edition, build 22621.2361
GPU : MSI GeForce RTX™ 4070 Ti SUPRIM X 12G
GPU DRIVER : 537.42

NVIDIA control panel settings:

PG27AQDM

• V-Sync : G-Sync compatible mode

• Resolution Setting : 2560x1440@240Hz, RGB, 10 bpc, Full Range

Optix MAG274QRF-QD

• V-Sync: Traditional Fixed

• Resolution setting: 2560x1440@120Hz, RGB, 10 bpc, full range

Tried and true troubleshooting:

• Reinstalling the graphics driver to the latest version did not work.

• Refreshing the firmware through the computer did not work.

• Replacing the DisplayPort cable did not work.

• Waiting a few minutes after power shutdown and reconnecting, and waiting for the auto-triggered pixel cleanup in standby to complete but it didn't work.

• Turn off the secondary monitor (But DP cable is still connected, power still plug.) but it doesn't work.

• Before using the PG27AQDM, I had not experienced this problem when using the Optix MAG274QRF-QD for almost a year (one month if only counting the monitor with the current GPU). However, the Optix MAG274QRF-QD is 8 Bit+FRC and is DP 1.2a

It seems to be a successful but not perfect temporary solution

• PG27AQDM resolution change to 2560x1440@240Hz, RGB, 8 bpc, full range, work fine with 8 Bit and 8 Bit + FRC when HDR On.

Tried another way of replicating the problem without encountering problems:

• PG27AQDM HDMI connected to Google Chromecast (Google TV 4K) test, entering sleep mode with 4K@60hz HDR 10Bit(Not sRGB) on did not trigger the issue.

• The issue is not triggered when switching between signal sources on the monitor side and when the monitor is turned off from monitor power button.

[dfr1238]

Update:

After a series of tests

NVIDIA GPU (same 4070Ti)

• The replication rate of the problem is greatly reduced when using 1440@240@10 bpc on a single screen.

• The problem is still reproduced with the same probability when the primary monitor (PG27AQDM) is connected by DP and the secondary monitor is connected by HDMI.

AMD Internal Graphics (AMD Radeon(TM) Graphics)

• The problem cannot be replicated when using a single screen.

• This issue cannot be replicated with dual monitors (secondary monitor connected to NVIDIA GPU and AMD internal Graphics ).

[jorimt]

1. Displays in DSC mode can introduce multi-monitor issues due to bandwidth limitations, etc.
2. Random black screen flicker is currently a known Nvidia driver issue:
https://us.download.nvidia.com/Windows/ ... -notes.pdf

nvidia-driver-notes.jpg (44.14 KiB) Viewed 3599 times

Whether the above is contributing to your particular issues on your specific model is the question.

Have you tried setting the PG27AQDM to 240Hz 2560x1440 8-bit to rule out DSC? Because that should prevent DSC from kicking in. You can also check to see if your monitor OSD has an option to disable DSC.

[Akso]

1. Displays in DSC mode can introduce multi-monitor issues due to bandwidth limitations, etc.
2. Random black screen flicker is currently a known Nvidia driver issue:
https://us.download.nvidia.com/Windows/ ... -notes.pdf
nvidia-driver-notes.jpg

Whether the above is contributing to your particular issues on your specific model is the question.

Have you tried setting the PG27AQDM to 240Hz 2560x1440 8-bit to rule out DSC? Because that should prevent DSC from kicking in. You can also check to see if your monitor OSD has an option to disable DSC.

I was thinking but does 10 bits takes more bandwidth than 8 bits ? Never got an answer on the internet for it.

[jorimt]

does 10 bits takes more bandwidth than 8 bits ?

Yes, it requires DSC on your model.

See:
https://pcmonitors.info/reviews/asus-ro ... ifications

*10-bit can be selected in the graphics driver at any refresh rate, up to the native resolution using DP 1.4 (with DSC).

Also see:
https://www.reddit.com/r/OLED_Gaming/co ... &context=3

I tested the Asus and the LG model, and I kept the LG because HDMI 2.1. I couldn't use DLDSR 10bit at highest refresh rate with the Asus model, probably because DLDSR isn't supported when DSC is in use.

[dfr1238]

1. Displays in DSC mode can introduce multi-monitor issues due to bandwidth limitations, etc.
2. Random black screen flicker is currently a known Nvidia driver issue:
https://us.download.nvidia.com/Windows/ ... -notes.pdf
nvidia-driver-notes.jpg

Whether the above is contributing to your particular issues on your specific model is the question.

Have you tried setting the PG27AQDM to 240Hz 2560x1440 8-bit to rule out DSC? Because that should prevent DSC from kicking in. You can also check to see if your monitor OSD has an option to disable DSC.

8 bpc mode was normal without any issues. So, I was thinking about that driver glitch too.

The OSD has options to choose the Displayport version, 1.2 and 1.4

[dfr1238]

1. Displays in DSC mode can introduce multi-monitor issues due to bandwidth limitations, etc.
2. Random black screen flicker is currently a known Nvidia driver issue:
https://us.download.nvidia.com/Windows/ ... -notes.pdf
nvidia-driver-notes.jpg

Whether the above is contributing to your particular issues on your specific model is the question.

Have you tried setting the PG27AQDM to 240Hz 2560x1440 8-bit to rule out DSC? Because that should prevent DSC from kicking in. You can also check to see if your monitor OSD has an option to disable DSC.

> Displays in DSC mode can introduce multi-monitor issues due to bandwidth limitations, etc.

Is this a model-specific issue or is it a general issue when DSC is enabled?

> Random black screen flicker is currently a known Nvidia driver issue

Yes, I'm speculating on this part. Because there is no problem in 8bpc mode.

It’s just that it specifically states a black screen instead of an error like the one in the video, so I’m not sure.

[jorimt]

The OSD has options to choose the Displayport version, 1.2 and 1.4

DisplayPort 1.2 mode will only support 2560 x 1440 at a maximum of 144Hz, so there's no reason to switch to it. Keeping it at 1.4 and switching to 8-bit will disable DSC mode on your monitor while maintaining 1440p at 240Hz.

Is this a model-specific issue or is it a general issue when DSC is enabled?

General issue.

\
Yes, I'm speculating on this part. Because there is no problem in 8bpc mode.

It’s just that it specifically states a black screen instead of an error like the one in the video, so I’m not sure.

Likely just a general description of DSC issues on the latest Nvidia drivers.

If you want to rule the driver out, you can temporarily download and install a version from before the issue existed (I don't know that version off-hand).

[dfr1238]

The OSD has options to choose the Displayport version, 1.2 and 1.4

DisplayPort 1.2 mode will only support 2560 x 1440 at a maximum of 144Hz, so there's no reason to switch to it. Keeping it at 1.4 and switching to 8-bit will disable DSC mode on your monitor while maintaining 1440p at 240Hz.

Is this a model-specific issue or is it a general issue when DSC is enabled?

General issue.

\
Yes, I'm speculating on this part. Because there is no problem in 8bpc mode.

It’s just that it specifically states a black screen instead of an error like the one in the video, so I’m not sure.

Likely just a general description of DSC issues on the latest Nvidia drivers.

If you want to rule the driver out, you can temporarily download and install a version from before the issue existed (I don't know that version off-hand).

Then I guess I'll just wait for the driver to fix this.

There is no big impact on the current use, mainly because the performance of 8 Bit + FRC in HDR seems to be close to that of 10 Bit, at least it is difficult for me to tell.

Another question I would like to ask is, will the performance of using 8 Bit + FRC on a native 10 Bit panel be better than that of a native 8 Bit panel?

[jorimt]

will the performance of using 8 Bit + FRC on a native 10 Bit panel be better than that of a native 8 Bit panel?

8-bit + FRC can be virtually indistinguishable from native 10-bit outside of some test patterns even on an 8-bit panel, so it certainly shouldn't be worse on a 10-bit panel, but probably not notably better either, since the 10-bit panel will probably be running in 8-bit mode with FRC anyway.