You should be able to use the "box" image with the p212 device-tree, read: https://wiki.libreelec.tv/hardware/amlogic
NB: There are no drivers (that work on upstream kernels) that exist for the WiFi chipset in the box.
You should be able to use the "box" image with the p212 device-tree, read: https://wiki.libreelec.tv/hardware/amlogic
NB: There are no drivers (that work on upstream kernels) that exist for the WiFi chipset in the box.
frakkin64 explanation sounds right. The LePotato dtb has no WiFi SDIO node and the VIM1 forces a Broadcom chipset so if boxes have Realtek/Qualcomm chips things either won't probe or will probe the wrong chip and fail.
Current LE12 nightly images have changes to boot files and even on VIM1/LePotato you'll now see an /amlogic folder that has all the dtb files inside (not just VIM1 or LePotato dtb) so experimenting is a simple as editing the dtb name in extlinux.conf.
RPi5 is stable and reliable (with LE12 images) so this sounds like something specific to twitch.
Read here: https://wiki.libreelec.tv/support/log-files
The image is nothing do with us ![]()
It runs on hardware we don't support ![]()
There is no source code ![]()
Can you upload/share a sample file somewhere?
If you disable hardware decoding, does the file play normally?
Are you running a stock LE release or one with "improvements" or "fixes" included?
Kodi supports updates so as long as the Kodi version keeps moving forwards; e.g. K20 to K21, or K21-beta1 to K21-rc1, then settings are preserved when updating and micro bumps from an older nightly to current nightly are low drama when we're late in the Kodi release cycle (in pre-Alpha things can be different).
Kodi does not support downgrades; although minor roll-backs e.g. moving from a current LE nightly to one from a couple of weeks ago to avoid an issue will normally work without issue. Downgrades where the Kodi major version changes cause issues.
Users with good backup habits or update paranoia can always generate a backup in the LE settings add-on ![]()
Oleg's images are largely based on the upstream kernel which supports RK3588 + RK3568. It's the first time I hear of RK3528 and while I don't actively track RK matters that's normally a good sign there's no support for that chipset.
Google shows this patch being advised/required in a couple of places: https://git.launchpad.net/ubuntu/+source…435523d2f374161
So the probable reason it doesn't work is: LE doesn't include that patch, or the udev rules file it's modifying, or the hid2hci binary that would be used if we did.
Simple binaries that aren't compiled/linked against other libraries can often be copied from e.g. Ubuntu and used on LE. So I would try copying hid2hci to /storage/bin/ and the 70-hid2hci.rules file to /storage/.config/udev.rules.d/ and then mod the rules file (with the changes in the patch) so it runs "/storage/bin/hi2hci" (explicit path) not "hid2hci" (which relies on the binary being in $PATH).
Maybe buy a cheap wifi router and set it to wireless bridge mode
^ this is what I do for test/dev simply because it avoids the effort of configuring a WiFi connection. Not that configuring WiFi is such a major effort, but after a decade of test/dev the novelty wears off. I have a bunch of older Apple Airport Express things which are not so fast these days, but compact and (when I got them) cheap.
I think the Q is best asked in the Tvheadend forum: https://tvheadend.org
2024-04-01 19:31:42.719 T:3379 info <general>: ffmpeg[0x2fce05f0]: Stream #0:0: Video: h264 (High), yuv420p(tv, bt709, progressive), 1920x1080 [SAR 1:1 DAR 16:9], 23.98 fps, 23.98 tbr, 1k tbn (default)
...
2024-04-01 19:31:42.720 T:3379 info <general>: [WHITELIST] Searching the whitelist for: width: 1920, height: 1080, fps: 23.976, 3D: false
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Using the default whitelist because the user whitelist is empty
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Searching for an exact resolution with an exact refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] No match for an exact resolution with an exact refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Searching for an exact resolution with double the refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] No match for an exact resolution with double the refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Searching for an exact resolution with a 3:2 pulldown refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] No match for a resolution with a 3:2 pulldown refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Searching for a desktop resolution with an exact refresh rate
2024-04-01 19:31:42.720 T:3379 debug <general>: [WHITELIST] Matched a desktop resolution with an exact refresh rate 3840x2160 @ 23.976025 Hz (26)
2024-04-01 19:31:42.720 T:3379 info <general>: Display resolution ADJUST : 3840x2160 @ 23.976025 Hz (26) (weight: 0.000)
Display More
^ I have a hunch this is trying to output 8-bit/1080p/H264 at 4K and I think this runs afoul of the RPi4's "H264 1080p max" rule so you get no video output. Have a read of https://wiki.libreelec.tv/configuration/4k-hdr and set the whitelist entries correctly. Working?
^ Network tools not System tools.
NB: For any future requests, the contents of the various tools packages are included in the package descriptions in the repo so you can check what's inside without needing to ask in the forum.
You need to ask in the YouTube add-on support thread in Kodi forums. I've no idea if there's a current solution to the problem but it has nothing to do with LE and everything to do with Google being a grade A1+ pain in the rear and moving the goalposts around on what's needed to evade their API usage rules and desire to ruin their user-experience with an unholy amount of adverts. In short, the old solution is no longer the solution. FWIW, I see the same but haven't worked up the enthusiasm to hunt down the solution just yet.
H264 has lots of variations and the H264 codec in the Amlogic vendor kernel that CE uses handles more of the non-standard ones than the RPi4 codec. Two common examples are 4K H264 (RPi4 handles 1080p max) and 10-bit H264 (not a broadcast standard, RPi4 doesn't support it at all). RPi5 sort of solves the RPi4 issue by having no H264 hardware codec at all so ffmpeg falls back to software decode which allows a wider range of media to be played; at the expense of needing more CPU (and RPi5 generally has enough).
NB: To confirm the theory ^ you'll need to share Kodi debug logs and/or media samples.
If you start to poke around in the Gitlab issue log for the Intel DRM driver you'll see reports and triage that reveal the problem isn't simple like LSPCON vs. no-LSPCON; as there are different LSPCON chips (different chip vendors) with different properties, and even identical LSPCON chips can have rather different implementations (different electrical properties, resulting in different bandwidths available on internal connections) on different boards. In some cases firmware changes can tweak voltages and such that drive the LSPCON chip differently so achieve improvements. In other cases the LSPCON implementation is more fundamentally wrong and no tweaking can get the right result. Most of the issue reports are focused on graphics performance; but the underlying issue is related to bandwidth and thus HBR audio (needing more bandwidth) suffers similarly to higher resolution and refresh rate and colour depth graphics (needing more bandwidth). TL/DR; ![]()