at least one HDMI mode on the Rpi4 has changed

[mwrich4]

I have a 'known good' monitor I've used for ALL of my Raspi boards from the first through 3B.

Having it not work for my new Raspi 4-4GB, sent me on the will not boot path for a few hours. I've seen others mention it in the forums but not as a clear topic.

I tried with the 3.2 noobs (no boot screen) and the full install of the latest Raspi Buster image. Neither even showed the rainbow screen. I ~finally~ gave my main monitor a trial and voila, working board after all.

My Raspi 3 gives the following responses:

volumio@volumiom:~$ sudo /opt/vc/bin/tvservice -s
state 0x12000a [HDMI DMT (4) RGB full 4:3], 640x480 @ 60.00Hz, progressive

volumio@volumiom:~$ sudo /opt/vc/bin/tvservice -d edid.dat
Written 256 bytes to edid.dat
volumio@volumiom:~$ sudo /opt/vc/bin/edidparser edid.dat
Enabling fuzzy format match...
Parsing edid.dat...
HDMI:EDID version 1.3, 1 extensions, screen size 26x14 cm
HDMI:EDID features - videodef 0x80 !standby !suspend !active off; colour encodin g:RGB444|YCbCr422; sRGB is not default colourspace; preferred format is native; does not support GTF
HDMI:EDID found monitor S/N descriptor tag 0xff
HDMI:EDID found monitor range descriptor tag 0xfd
HDMI:EDID monitor range offsets: V min=0, V max=0, H min=0, H max=0
HDMI:EDID monitor range: vertical is 50-75 Hz, horizontal is 30-85 kHz, max pixe l clock is 150 MHz
HDMI:EDID monitor range does not support GTF
HDMI:EDID found monitor name descriptor tag 0xfc
HDMI:EDID monitor name is MotoAttach
HDMI:EDID found preferred DMT detail timing format: 1366x768p @ 60 Hz (86)
HDMI:EDID established timing I/II bytes are 00 00 00
HDMI:EDID standard timings block x 8: 0x0101 0101 0101 0101 0101 0101 0101 0101
HDMI:EDID parsing v3 CEA extension 0
HDMI:EDID monitor support - underscan IT formats:no, basic audio:yes, yuv444:yes , yuv422:yes, #native DTD:1
HDMI:EDID found DMT detail timing format: 1366x768p @ 60 Hz (86)
HDMI:EDID found DMT detail timing format: 1366x768p @ 60 Hz (86)
HDMI:EDID failed to find a matching detail format for 0x1080i hfp:88 hs:44 hbp:- 116 vfp:2 vs:5 vbp:16 pixel clock:0 MHz
HDMI:EDID calculated refresh rate is 0 Hz
HDMI:EDID guessing the format to be 0x1080i @24 Hz
HDMI:EDID found unknown detail timing format: 0x1080i hfp:88 hs:44 hbp:-116 vfp: 2 vs:5 vbp:16 pixel clock:0 MHz
HDMI:EDID found CEA format: code 1, 640x480p @ 60Hz
HDMI:EDID found CEA format: code 3, 720x480p @ 60Hz
HDMI:EDID found CEA format: code 18, 720x576p @ 50Hz
HDMI:EDID found audio format 2 channels PCM, sample rate: 32|44|48|96|192 kHz, s ample size: 16|20|24 bits
HDMI:EDID found HDMI VSDB length 5
HDMI:EDID HDMI VSDB has physical address 1.0.0.0
HDMI:EDID HDMI VSDB has no extension fields
HDMI:EDID adding mandatory support for DMT (4) 640x480p @ 60Hz
HDMI:EDID adding mandatory support for CEA (2) 720x480p @ 60Hz
HDMI:EDID adding mandatory support for CEA (17) 720x576p @ 50Hz
HDMI:EDID filtering formats with pixel clock > 162 MHz or h. blanking > 1023
HDMI:EDID best score mode initialised to DMT (4) 640x480p @ 60 Hz with pixel clo ck 25 MHz (score 0)
HDMI:EDID best score mode is now CEA (1) 640x480p @ 60 Hz with pixel clock 25 MH z (score 43432)
HDMI:EDID best score mode is now CEA (2) 720x480p @ 60 Hz with pixel clock 27 MH z (score 45736)
HDMI:EDID CEA mode (3) 720x480p @ 60 Hz with pixel clock 27 MHz has a score of 4 5736
HDMI:EDID DMT mode (4) 640x480p @ 60 Hz with pixel clock 25 MHz has a score of 3 6864
HDMI:EDID CEA mode (17) 720x576p @ 50 Hz with pixel clock 27 MHz has a score of 45736
HDMI:EDID CEA mode (18) 720x576p @ 50 Hz with pixel clock 27 MHz has a score of 45736
HDMI:EDID best score mode is now DMT (86) 1366x768p @ 60 Hz with pixel clock 72 MHz (score 5188835)
HDMI:EDID preferred mode remained as DMT (86) 1366x768p @ 60 Hz with pixel clock 72 MHz
HDMI:EDID has HDMI support and audio support
/opt/vc/bin/edidparser exited with code 0

The best score listed above is indeed the native resolution for my Motorola Lapdock.

edid.zip

edid.dat and edid.parsed

(1.36 KiB) Downloaded 19 times

[andrum99]

A dump of what the Pi 4 produces would have been more helpful - can you provide that? Note that this is not to do with HDMI modes changing, but rather that the hardware itself is significantly different.

You can force a specific display mode using parameters in the config.txt file. So for your setup it would be:

Code: Select all

hdmi_force_hotplug=1
hdmi_group=2
hdmi_mode=86
hdmi_drive=2

Although I note that the mode table at https://www.raspberrypi.org/documentati ... t/video.md says that DMT mode 86 has reduced blanking, so there may be subtle differences in the Pi 4 hardware that cause it to no longer work properly with your setup.

[andrum99]

Actually, per https://github.com/raspberrypi/firmware/issues/1202 can you try this instead:

Code: Select all

hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

[mwrich4]

the changes suggested allow me to glimpse the rainbow and 4 berries screens

I will try more later.

edid.dat.gz

rpi4 output

(172 Bytes) Downloaded 31 times

edid.parse.txt.gz

rpi4 output text form

(1.03 KiB) Downloaded 38 times

[aBUGSworstnightmare]

use 'edid-decode' for parsing your EDID data!

[FTrevorGowen]

use 'edid-decode' for parsing your EDID data!

+1 but, unless things have changed recently, it's not (currently) "part of Buster" and has to installed from here:
https://git.linuxtv.org/edid-decode.git/about/
(FWIW, some examples of its output cf. that of parse-edid are here: http://www.cpmspectrepi.uk/raspberry_pi ... uster.html )
Trev.

[aBUGSworstnightmare]

use 'edid-decode' for parsing your EDID data!

+1 but, unless things have changed recently, it's not (currently) "part of Buster" and has to installed from here:
https://git.linuxtv.org/edid-decode.git/about/
...

i've installed it with 'sudo apt install edid-decode'

[mwrich4]

I installed edid-decode as recommended above.

edid-decode.txt.gz

edid-decode output from Rpi4

(841 Bytes) Downloaded 16 times

With the recommend change to /boot/config.txt, I didn't see the post screen just now. Plus, the Xconsole was not started without me first booting with my main screen connected instead of the Lapdock. The pi desktop is set to 1280x720 because 1366x768 was not in the list of choices.

After booting with my Samsung 2770, the xconsole started and I was also able to finally use VNC. I also reconnected the LapDock and see the console now. It appears that I can't start the Rpi4 headless (with these settings) or with the LapDock connected. I can't seem to rsh to this Rpi unless my samsung is connected to the HDMI port. I'm not in a rush. The work-around only requires that my samsung be connected. I currently use the DVI input for my PC so I can live with this while playing with the new Rpi4.


Aside: I know more changes are needed for the distro to mature. I was hoping these little details might help find loose ends. There was a post in troubleshooting where the fellow had a similar trouble and settled for the workaround for using his smaller HDMI device as well. I made a comment to that thread but failed to notice that it was marked [solved]. I believe that things that once worked and don't work now bear a little investigation to help smooth the path for others later.

I'm a bit out of my depth here with HDMI match-making, but I'm willing to gather whatever you need to 'further the cause'

Have a good day,
mark

[FTrevorGowen]

use 'edid-decode' for parsing your EDID data!

+1 but, unless things have changed recently, it's not (currently) "part of Buster" and has to installed from here:
https://git.linuxtv.org/edid-decode.git/about/
...

i've installed it with 'sudo apt install edid-decode'

O.K. That's good news. I'm guessing it's been added sometime in the last couple of weeks or so - been busy with non-Pi things (external DIY) whilst the weather's been better.
Trev.

[andrum99]

the changes suggested allow me to glimpse the rainbow and 4 berries screens

Which changes worked? Forcing DMT mode 86, or the custom mode using hdmi_cvt? Note that on the Pi 4 the display is handled differently when the Pi initially boots up, and once the graphics X desktop is running. It seems you've get the first one working, but not the second.

[mwrich4]

Actually, per https://github.com/raspberrypi/firmware/issues/1202 can you try this instead:

Code: Select all

hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

adding/changing these values in /boot/config.txt may have helped a bit.

I'm not sure whether ultimately the above worked or if was simply connecting the Samsung to the Rpi4 for the initial boot.

The short version is even with the settings above I do not see the rainbow screen or the 4 raspberries when using the LapDock after a coldboot.

The main issue here is none of this was needed with the Rpi0 through Rpi3. They just worked, so what has changed. I know, 'the video hardware and new software'. I had hoped that providing the edid.dat would lead to an aha moment.

Additionally, why doesn't the system load properly to the xconsole with no monitor connected. I can't rsh or use VNC until after booting with my samsung attached first.

[andrum99]

Actually, per https://github.com/raspberrypi/firmware/issues/1202 can you try this instead:

Code: Select all

hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

adding/changing these values in /boot/config.txt may have helped a bit.

I'm not sure whether ultimately the above worked or if was simply connecting the Samsung to the Rpi4 for the initial boot.

The short version is even with the settings above I do not see the rainbow screen or the 4 raspberries when using the LapDock after a coldboot.

The main issue here is none of this was needed with the Rpi0 through Rpi3. They just worked, so what has changed. I know, 'the video hardware and new software'. I had hoped that providing the edid.dat would lead to an aha moment.

In https://github.com/raspberrypi/firmware/issues/1202, 6by9 (a Raspberry Pi engineer) said:

Please remember that the Pi4 has completely replaced the HDMI, HVS, and Pixel valve blocks. Where issues occur we will investigate, but have to set priorities.

These blocks being the part of the chip that produces the display - HVS being Hardware Video Scaler.

A fix for the bug will be available in the next few days. Once it is, run the following to get it:

Code: Select all

sudo rpi-update

Note that this installs pre-release software on your Pi and may break something else on your system. (Although unlikely, it is a possibility).

Additionally, why doesn't the system load properly to the xconsole with no monitor connected. I can't rsh or use VNC until after booting with my samsung attached first.

The Pi should still boot with no monitor connected - you should be able to SSH in, assuming you have enabled the SSH server on the Pi - it is disabled by default. The rsh command on the current Raspbian redirects to SSH - I would not expect the rsh protocol to be enabled at all, ever, since it is not encrypted.

VNC not working with no monitor attached is the expected behaviour. By default with the RealVNC server, unless the X desktop can be brought up, the VNC server will have nothing to connect to, and therefore nothing for VNC to "mirror" to remote clients. The workaround in this case is either to force a particular video mode - tell the Pi to ignore the fact that there is no screen attached and bring up HDMI 0 in a single mode using settings in /boot/config.txt. Alternatively you can run the VNC server attached to a "virtual" desktop that exists only in memory, not on the (not powered on) local screen. A remote client will then connect to the otherwise invisible "virtual" desktop via VNC. The official RealVNC docs are at https://www.realvnc.com/en/connect/docs ... y-pi-setup. The "virtual" mode for VNC also allows you to run 2 separate desktops - one on the local screen, the other for just remote clients via VNC.

[FTrevorGowen]

....
.... Alternatively you can run the VNC server attached to a "virtual" desktop that exists only in memory, not on the (not powered on) local screen. A remote client will then connect to the otherwise invisible "virtual" desktop via VNC. The official RealVNC docs are at https://www.realvnc.com/en/connect/docs ... y-pi-setup. The "virtual" mode for VNC also allows you to run 2 separate desktops - one on the local screen, the other for just remote clients via VNC.

FWIW, there are a couple of examples of using this method within my webpages here:
http://www.cpmspectrepi.uk/raspberry_pi ... uster.html
Trev.

[mwrich4]

In https://github.com/raspberrypi/firmware/issues/1202, 6by9 (a Raspberry Pi engineer) said:
These blocks being the part of the chip that produces the display - HVS being Hardware Video Scaler.
A fix for the bug will be available in the next few days. Once it is, run the following to get it:

Code: Select all

sudo rpi-update

Note that this installs pre-release software on your Pi and may break something else on your system. (Although unlikely, it is a possibility).

This is very expected, I realized it's early and things like this happen.

I have tried the VNC server setup from allowing multiple separate Xconsoles when I was testing my 320x480 screen while also auditioning the Pi3 as music server with clementine (I settled on volumio for now). IIRC that Rpi3 always booted even without an HDMI device. There may have been more changes added when the LCD setup was applied than I recognize.

I appreciate the detailed response in the thread. I wanted to make sure my minor finding was reviewed by the people looking at 'surprise' issues. Especially since another had a similar issue.

I have written and tested a bare metal bootloader and system on another platform. It's tedious and easy to overlook 'assumed' good areas even when I was the only author. I imagine with such a large and complex system it's not a favorite thing for anyone.

Thanks,
Mark

EDIT:
After reading the referenced article above, if someone would ping this thread when I can get the change via rpi-update, I'll give it a try.
When I do that, should I first remove the suggested /boot/config.txt changes.

[mwrich4]

Ran 'rpi-update' today on my latest card image. After a cold boot, the Rpi4 with my Atrix seems to be working as it did for previous RasPis.

I'll update if I find this is not true. Thanks again for the response JamesH on this thread "https://github.com/raspberrypi/firmware/issues/1202".

I expect that eventually as the dust settles these fixes will be part of the regular distribution.

cheers

[mwrich4]

I was so happy to have something work, I forgot this was in my config.txt

# uncomment to force a specific HDMI mode (this will force VGA)
hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

When I commented out the 3 lines like a clean system, the HDMI lapdock screen was not detected. Unplugging the HDMI cable and replugging the cable did not give me a display either. Not even the backlight was triggered.

So, the issue still exists to some extent in the firmware boot

[jamesh]

I was so happy to have something work, I forgot this was in my config.txt

# uncomment to force a specific HDMI mode (this will force VGA)
hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

When I commented out the 3 lines like a clean system, the HDMI lapdock screen was not detected. Unplugging the HDMI cable and replugging the cable did not give me a display either. Not even the backlight was triggered.

So, the issue still exists to some extent in the firmware boot

Expected - rpi-update doesn't get the modified firmware - you would need to install it from the zip file in the linked thread. That firmware adds reading of the 'detailed' timings section of the EDID, which may, or may no, be how you display advertises its capabilities.

[mwrich4]

OK, here a link to a zip of the test firmware.

https://drive.google.com/open?id=1xBfTK ... Lh5Sr8apaE

This is very much beta. I'm interested to see whether it correctly detects odd monitors with non-standard timings. I don't have anything like that so am relying on EDID dumped from other devices.

Thanks.

I think I found your reference this time. I'll try it later today.

EDIT: I see in the zip
fixup.dat
fixup_x.dat
start.elf
start_x.elf

I assume I place them in the /boot directory overwriting the existing files. (I'll save a copy first)

[mwrich4]

I tried the images linked above on the latest 'rpi-update'(ed) code for my Pi4 connected to the Lapdock screen and still no rainbow and no 4 raspberries. I also don't see the desktop (set at 1366x768).

I have hdmi_force_hotplug=1 so the system will allow VNC with no monitor connected.

Let me know if you have any further items to try.

Thanks

[FTrevorGowen]

...
I have hdmi_force_hotplug=1 so the system will allow VNC with no monitor connected.
...

Alternatively you might find using the VNC server's virtual mode useful if you need to check the hotplug signal's present see my examples here FWIW:
http://www.cpmspectrepi.uk/raspberry_pi ... uster.html
Trev.

[benicehavefun]

Actually, per https://github.com/raspberrypi/firmware/issues/1202 can you try this instead:

Code: Select all

hdmi_group=2
hdmi_cvt=1366 768 60
hdmi_mode=87

For Motorola Atrix Lapdock users -- these changes are still what's needed for the 2020-02-13 release of Raspbian Buster. One of the easier ways to the add these lines to your config.txt file is to put the SD card in a Windows system (or probably any OS) and edit the file directly.