brightness changes randomly failing (LG 24MB56) #67
Labels
Comments
|
What you're experiencing is an unfortunate behavior I see for some
features on some monitors. Sometimes, verification succeeds, but there
is no change in the monitor state. Other times, verification fails, but
the monitor is visibly changed. Sometimes the problem only occurs for
"large" changes, e.g. "setvcp 10 5 fails", but "set vcp 10 55 succeeds".
For example, on my Dell U3011 "setvcp 16 5" to set the Red gain to a low
value changes the screen, but reports failure. "setvcp 16 100" both
changes the screen and does not report failure.
Note that setvcp verification is just a getvcp after the setvcp.
Bottom line: If I had a solution this would be a priority. This
behavior is especially problematic in the GUI version of ddcutil that's
under construction. Unfortunately, it appears to be a problem in
particular monitors' DDC/CI implementation.
But thank you for reporting the problem. I'll add your monitor to the
rogues gallery of monitors at www.ddcutil.com/monitor_notes, and perhaps
add a discussion of this behavior to the FAQ.
Sanford
…On 11/30/18 10:53 AM, qlum wrote:
Sometimes when I try to change the brightness on my monitors It returns:
|Verification failed for feature 10|
After that the brightness on the particular monitor remains unchanged.
This is the command I use:
|sudo ddcutil -d 1 setvcp 10 50|
While this is not a huge issue and doesn't happen often enough when
changing the brightness on 5 identical monitors.
I hope this helps / I know this is probably not a priority.
probe.txt <https://github.com/rockowitz/ddcutil/files/2633983/probe.txt>
capabilities.txt
<https://github.com/rockowitz/ddcutil/files/2633984/capabilities.txt>
interrogate.txt
<https://github.com/rockowitz/ddcutil/files/2633985/interrogate.txt>
output.log <https://github.com/rockowitz/ddcutil/files/2633986/output.log>
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#67>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/ANhsbiJxfdUM5vRX7kTZIqFXq9qxAvQmks5u0VRggaJpZM4Y76aL>.
|
|
One further comment. Some monitor features can be changed only if the
monitor is in certain input modes. For example, typically Hue and
Saturation controls will be work only if the input color format is YPrPb
or such, as opposed to RGB. So verification will be telling the truth
in those cases. But I've never seen a monitor for which brightness
control (feature x10) is input mode dependent.
Sanford
…On 11/30/18 10:53 AM, qlum wrote:
Sometimes when I try to change the brightness on my monitors It returns:
|Verification failed for feature 10|
After that the brightness on the particular monitor remains unchanged.
This is the command I use:
|sudo ddcutil -d 1 setvcp 10 50|
While this is not a huge issue and doesn't happen often enough when
changing the brightness on 5 identical monitors.
I hope this helps / I know this is probably not a priority.
probe.txt <https://github.com/rockowitz/ddcutil/files/2633983/probe.txt>
capabilities.txt
<https://github.com/rockowitz/ddcutil/files/2633984/capabilities.txt>
interrogate.txt
<https://github.com/rockowitz/ddcutil/files/2633985/interrogate.txt>
output.log <https://github.com/rockowitz/ddcutil/files/2633986/output.log>
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#67>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/ANhsbiJxfdUM5vRX7kTZIqFXq9qxAvQmks5u0VRggaJpZM4Y76aL>.
|
|
Just to specify when setting the brightness form for example 20 to 30 and back to 20 again it will randomly fail with no visible changes it works 9/10 times but about 1/10 it fails. So in my particular case it doesn't seem to be related to smaller or larger steps. The verification failed does seem to correspond with the brightness actually changing or not. |
|
I've taken a closer look at the interrogate output. It looks like
you're using both the on-chip Intel video (i915 driver) for some
displays and a Radeon card with the radeon driver for the remainder.
The amdgpu driver is loaded but not used. The system logs are showing
errors for DisplayPort attached display(s) on the Radeon card. The
ddcutil's status show a fair amount of retries.
So my questions are these:
- Are you seeing the problem with only the Radeon attached displays or
the ones using on-chip Intel GPU?
- Are the problems limited to DisplayPort attached displays?
- Are you seeing intermittent problems with commands other than setvcp?
There are several command options that will dump diagnostics, though the
volume of output is a bit overwhelming for normal use. These include:
report DDC communication errors: --ddc
report retry and error stats: --stats tries --stats errors
To make the stats specific to a single display, the following way of
specifying the display will skip the detection phase:
--nodetect --bus <busno>
where <busno> is the I2C bus number.
Sanford
…On 12/3/18 11:30 AM, qlum wrote:
Just to specify when setting the brightness form for example 20 to 30
and back to 20 again it will randomly fail with no visible changes it
works 9/10 times but about 1/10 it fails. So in my particular case it
doesn't seem to be related to smaller or larger steps.
The verification failed does seem to correspond with the brightness
actually changing or not.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#67 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ANhsbsjQ4Z8HlfOoeekt6EXK39xQ943Cks5u1VGTgaJpZM4Y76aL>.
|
|
The problems are on all my monitors both on the radeon and the intel I will add here that the gpu I use is a sapphire flex card which does not have any displayport connectors but instead allows me to connect 3 monitors without using displayport ( I assume it has an active adapter build into the card itself) I have not had any problems with other commands although I must admit that I use setvcp for the brightness quite often and don't regularly use any of the other commands. I just ran the command quite a few times with an attached script (fixbrightness.txt) On the monitor side everything is plugged into an hdmi port in case of dvi with a passive adapter cable I also added the output of one run of my script, in this run display 3 failed. |
|
Hi @qlum & @rockowitz I was getting the same error for 2016/2017 Lenovo ThinkVision X1 27" 4K monitor connected via USB-C cable (DP). |
|
Thank you for reporting your observation @kowalcj0.
There is, of course, no Monitor Control Command Set feature defined to
control DCR - the latest revision is 2.2a from July 2011. It would be
interesting to see if we can determine if there is an undocumented VCP
feature that does this. Can I ask you to run "sudo ddcutil interrogate"
and send the output as an attachment?
Regards,
Sanford
…On 7/1/20 6:00 PM, Janusz Kowalczyk wrote:
Hi @qlum <https://github.com/qlum> & @rockowitz
<https://github.com/rockowitz>
I was getting the same error for 2016/2017 Lenovo ThinkVision X1 27"
4K monitor connected via USB-C cable (DP).
When I had a look at the OSD I've noticed that when a DCR (Dynamic
Contrast Ratio) function was enabled then I wasn't able to change the
brightness even via OSD. Once I disabled it then |ddcutil| started to
work.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#67 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADMGY3SNBEED2XDYPCORISTRZOWWZANCNFSM4GHPU2FQ>.
|
|
Reviewing the interrogate output, there are 4 VCP features that are
candidates for controlling DCR and Low Blue Light:
Feature x68:
- Not listed in capabilities string
- Not defined in MCCS
- From scanning, appears to be a NC feature taking values in the range 0..5
Feature xa8:
- Not listed in capabilities string
- Not defined in MCCS
- From scanning, appears to be a NC feature taking values in the range 0..3
Feature xe9:
- Capabilities string reports that it takes values 0, 1
- In MCCS manufacturer specific range
- getvcp response reports it as unsupported
Feature xf0:
- Capabilities string reports that it takes values 0, 2
- In MCCS manufacturer specific range
- getvcp response reports it as unsupported
We're looking for a NC feature that takes exactly 2 values, which rules
out x68 and xa8.
That features xe9 and xf0 are defined in the capabilities string but
getvcp reports them as unsupported suggests they are write-only.
If you're feeling adventurous, try setting feature xe9 to 0 or 2, and
feature xf0 to 0 or 1, and see if the DCR or Blue Light settings are
affected.
If the getvcp commands put the monitor into a weird state, "setvcp 04 1"
resets the monitor to factory defaults.
…On 7/2/20 9:46 AM, Janusz Kowalczyk wrote:
Sure @rockowitz <https://github.com/rockowitz>
lenovo_thinkvision_x1_interrogate.log
<https://github.com/rockowitz/ddcutil/files/4864634/lenovo_thinkvision_x1_interrogate.log>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#67 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADMGY3WUFAPMQBB2BXPGFQDRZSFTJANCNFSM4GHPU2FQ>.
|
|
Additional questions:
- When DCR is enabled, are other features disabled or only brightness?
- Does enabling the Low Blue Light feature disable features as well?
Thanks.
…On 7/1/20 6:00 PM, Janusz Kowalczyk wrote:
Hi @qlum <https://github.com/qlum> & @rockowitz
<https://github.com/rockowitz>
I was getting the same error for 2016/2017 Lenovo ThinkVision X1 27"
4K monitor connected via USB-C cable (DP).
When I had a look at the OSD I've noticed that when a DCR (Dynamic
Contrast Ratio) function was enabled then I wasn't able to change the
brightness even via OSD. Once I disabled it then |ddcutil| started to
work.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#67 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADMGY3SNBEED2XDYPCORISTRZOWWZANCNFSM4GHPU2FQ>.
|
|
Hi @rockowitz Sorry for a belated response, been out of town for few days. I tried to sudo ddcutil -d 1 setvcp xe9 0
Setting value failed for feature e9. rc=DDCRC_REPORTED_UNSUPPORTED(-3005): DDC reports facility unsupportedsudo ddcutil -d 1 setvcp xe9 2
Setting value failed for feature e9. rc=DDCRC_REPORTED_UNSUPPORTED(-3005): DDC reports facility unsupportedre. Additional questions:
|
|
Thank you for the info. Take a look at the Lenovo ThinkVision X1
section on http://www.ddcutil.com/monitor_notes and let me know if I've
summarized your results incorrectly.
As to the setvcp commands, the DDCRC_REPORTED_UNSUPPORTED error is a red
herring. It is the result of the automatic getvcp that's performed to
verify that the value was actually set. Setting a feature value only
returns errors if the request packet is not successfully delivered.
There's no response packet that can indicate the monitor processed the
request properly. Hence the automatic getvcp to verify the result. If
we know the feature is write-only, verification is not performed. But
for manufacturer-specific features we don't know in advance. I'll
change the code to suppress the error message if the getvcp fails in
this case. For now, you can use the "--noverify" option to skip the
verification step.
Your setvcp commands for feature xE9 mix the values for xE9 and xF0.
Please execute the following commands. For each, are any changes in the
monitor appearance or the OSD?
# sudo ddcutil -d 1 --noverify setvcp e9 1
# sudo ddcutil -d 1 --noverify setvcp f0 1
# sudo ddcutil -d 1 --noverify setvcp f0 2
Sanford
…On 7/6/20 2:55 PM, Janusz Kowalczyk wrote:
Hi @rockowitz <https://github.com/rockowitz>
Sorry for a belated response, been out of town for few days.
I tried to |setvcp xe9 0| & |setvcp xe9 2| but got the same error:
sudo ddcutil -d 1 setvcp xe9 0
Setting value failedfor feature e9. rc=DDCRC_REPORTED_UNSUPPORTED(-3005): DDC reports facility unsupported
sudo ddcutil -d 1 setvcp xe9 2
Setting value failedfor feature e9. rc=DDCRC_REPORTED_UNSUPPORTED(-3005): DDC reports facility unsupported
re. Additional questions:
*
When DCR is enabled, are other features disabled or only brightness?
o in "|Adjust overall screen Brightness/Contrast/DSR|" menu:
brightness & contrast features are disabled
o in "|Adjust image properties|" menu: |Color| and |Scaling|
features are disabled
o in "|Options|" menu: |DP Select| is disabled
*
Does enabling the Low Blue Light feature disable features as well?
o When |Low Blue Light| is enabled, then
o the whole "|Adjust overall screen Brightness/Contrast/DSR|"
menu is disabled
o in "|Adjust image properties|" menu: |Color| and |Scaling|
features are disabled
o in "|Options|" menu: |DP Select| is disabled
*
When |DCR| is disabled then:
o in "|Adjust image properties|" menu: only |Scaling| is disabled
o in "|Options|" menu: |DP Select| is disabled
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#67 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADMGY3VUOMVNJH2BPGBRDULR2IMZTANCNFSM4GHPU2FQ>.
|
|
Hi Sanford, LGTM 👍 Thank you for documenting it. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Sometimes when I try to change the brightness on my monitors It returns:
Verification failed for feature 10After that the brightness on the particular monitor remains unchanged.
This is the command I use:
sudo ddcutil -d 1 setvcp 10 50While this is not a huge issue and doesn't happen often enough when changing the brightness on 5 identical monitors.
I hope this helps / I know this is probably not a priority.
probe.txt
capabilities.txt
interrogate.txt
output.log
The text was updated successfully, but these errors were encountered: