Your TV set switches to the ‘active source’, which is normally intended. How it reacts on simultaneous "active sources" depends on your TV. As a rule, the active source that is reported last wins. This could lead to the situation that you have to wait until the satellite receiver reports itself as an active source after startup. You could also start this satellite receiver only and patiently wait until this activates the TV.
With some TVs, you can set a specific external input as the standard for watching television.
Maybe a mitigation is possible if you disable "Switch source to this device on startup" and/or "Wake devices when deactivating the screensaver" in the "CEC Adapter" settings of LibreELEC: Settings -> System -> Input -> Peripherals -> CEC Adapter
However, this could also lead to some problems, e.g. that the remote control via CEC not working properly or partially only because another CEC device is being controlled or similar.
Posts by HarryH
-
-
Stuntel ,
You should make sure that you use the HDMI port that is closest to the USB-C port. Then CEC will work again. 12.0.1 temporarily supported CEC on both ports, but there were too many unresolved issues, so support was limited to the first HDMI port. -
It's not currently on the ToDo list as the V5 has nothing in common with the previous ONE enclosures other than the name and the manufacturer. In addition, I don't have an OLED device to test the function, and I would need capable testers with the ONE V5 case to get useful feedback in a timely manner during the development/debugging phase.
As far as I can see in Argon40's implementation with their scripts, the OLED requires additional Python modules like Luma.OLED and some binaries provided by Argon40. The latter without licence information. This hinders an integration, not only the additional abundant effort.
What expectations would you have regarding OLED support? Only the few cyclically repeated system information as with the argon1v5.sh script?
-
Please keep in mind that LE is reduced to just the required components for KODI and hosted in a readonly root filesystem. All additional things (more than a simple script), which are not already available in the base image or as add-on, you want to install should be a package. This can be an add-on package or directly included in a self build image. Many things are documented in the Wiki, at GitHub and also in the code itself.
As for your example with the Python background process, you can set it up manually as you have already done, or create an KODI add-on (script or service) that does the same. Adding new additional Python libraries can be significantly more difficult and depends heavily on how experienced you already are with compiling and packaging Linux packages.
Maybe I'm just misunderstanding you. But your wishes/goals sound very ambitious and require a good understanding of hardware/specialised knowledge in some areas. There is no such thing as a simple all-in-one guide. You should also not ignore the fact that not everything can be realised or is necessary on every hardware platform.Some starting points:
-
You are welcome. It was just a guess on my part that you also moved the fan_control.sh shell script to the subfolder called ‘scripts’. You are right that it makes no difference where it is located as long as the path in the autostart.sh script matches.
If you set FAN_OFF = 0, the fan should be stopped until the CPU temp value cross the threshold of MIN_TEMP. So you could lower the value MIN_TEMP near to the idle temperature for testing purposes, to trigger ON/OFF earlier and check the functionality of the script. -
Because it relies on gpiozero you could be affected by this (RPi gpiozero): https://libreelec.tv/2025/01/22/libreelec-nexus-12-0-2/
To find out where the problem comes from, what is the output of python3 /storage/scripts/fan_control.py ? Are there any exceptions/error messages? Or is this what you mean by?QuoteI was able to control my fan just fine from the interactive python so I know I have everything hooked up correctly.
If this works, the next step should be to test the wrapper script:
sh /storage/scripts/fan_control.sh start and sh /storage/scripts/fan_control.sh stop -
Version 1.1.12 released:
- new GUI languages: Korean, Chinese (Simplified Han script)
- missing Dutch addon description and summary added
-
Sounds to me like the issue which was already discussed here: RE: Nightly Builds
-
Sounds promising. I think they mean some kind of speed ramp (the longer you hold a button down, the shorter the distance to the next repetition) and I'm keeping my fingers crossed.
PS: Regarding the mouse mode, where I was wrong: As I didn't know the model you have, I chose it at random: 65RTS4K
-
Yes, it's currently not a solution for your situation, but it looks much better. The timeout looks realistic and the different can clearly identified. So it should be technical possible to fix the issue via firmware. I have seen at the Cello webpages, that the remote control has a mouse mode. Does it make a different if you switch on/off this feature after you switched to the HDMI input?
Just to be sure, since you didn't start with a fresh install. Can you remember if you created an edid.bin file via getedid at some point in the past that might have remained in the directory structure (ls -la /storage/.config/firmware/edid/edid.bin)? If so, remove that via getedid delete and reboot.
Have you already tried a different HDMI port on your TV? -
Has the output of the cec-client changed in regards to the 2 things I mentioned? I assume you only get 1 event when you hold down the button, not multiple, right?
Do you have already checked if a newer firmware version for your Cello TV is available?
-
Theoretically yes, if you put the image file in the /storage/.update directory and then reboot. Please make a backup before.
The main problem could be that you have to reinstall additional add-ons because the architecture for the RPi4/5 has been changed from arm to aarch64. This is the reason why it is not recommended to go this way.
-
Can you try it with a recent LE 12.0.2 from a spare SD card whether it behave different?
-
This part looks strange to me, or better: is different to the behaviour I know from my SONY TVs.
(rel:500,rep:0,prs:500,rel:0)
The prs: counter increases continuously by 1, starting at 1, as long as I hold down a button. In your case, it seems to be fixed to 500. Also timeout:30222488ms looks wrong -> 8h 24min.
Which RPi do you have and which LE version do you currently use?
-
You can also use cec-client to get more CEC details than with the KODI debug logging.
- Stop KODI process: systemctl stop kodi
- Start CEC client: cec-client
This CLI client will announced as "CECTester" on the TV. Depending on the behaviour of the TV it could be necessary to switch to the selected HDMI port again, to get CEC to work.
In general it's sometimes a good idea to disconnect the HDMI devices and TV physical and make all the components powerless simultaneous for some minutes. Afterwards the HDMI handshake will be initiated again and sometimes the CEC issues are like "magical" blown away.
-
Appears like a firmware issue of your Cello TV. To diagnostic what happens in detail, you could enable the KODI debug logging and watch with tail -F /storage/.kodi/temp/kodi.log which key events will be received from your TV via CEC during you held the up/down arrow keys and if there is a different to left/right arrow keys.
Some TVs interprets a long press signal from the remote control as a different function or the remote control sends a different key instead of repeating. If there is send a different key via CEC, then you could map this to Channel Up / Down to have page wise scrolling to mitigate the issue. -
Try and play a little bit with these settings for the CEC adapter:
- Remote button press repeat rate (ms): 100
- Remote button press delay before repeating (ms): 200
- Remote button press release time (ms): 0
Settings -> System -> Input -> Peripherals -> CEC Adapter
-
My experiences: some VC1 Bluray releases for example "The Bucket List" or "The Red Baron" are not playable fluent anymore, since the VC1 hardware decoder has been dropped with RPi4. Overclocking the RPi4 above 2.1 GHz also seems to make it fast enough to decode these titles via software, but it's like riding the razor's edge as to whether these settings are stable enough to enjoy the movie.