Hopefully Software Freedom Conservancy's GPL compliance lawsuit against Vizio will result in more of the necessary source code becoming publicly available, merged into existing operating systems like OpenWRT, Alpine or Debian and then projects like Kodi or KDE Plasma Bigscreen ported to run on top. If Conservancy win under their current strategy, then that would enable any recipient of a Linux based TV to sue the vendor for GPL compliance.
Seems acceptable from a warranty pov. Obviously if they refuse to replace your cracked screen because you've changed the firmware then it's unacceptable.
Yeah, I'm surprised that we have aftermarket firmware for phones and some embedded devices, but not for TVs or printers. Maybe the space is just more fragmented?
I expect that aftermarket firmware would be an issue for content providers. For example, is Netflix going to allow third party firmware developers to create an app to access their content? They will already have contracts with content distributors like Lionsgate that mandate DRM in their own apps to prevent piracy.
If your aftermarket firmware doesn't support Netflix, Prime, HBO, etc then it might have limited appeal. Browser based apps might get around that in a limited way.
As for printers, I think of them as loss leaders to generate ink sales.
Don't know either space particularly well so might be wrong.
Yeah, I wondered about the Android situation; Android itself is semi-standard, in that once you get the hardware support and a way to run your image it's not too hard to build AOSP for nearly any phone. On the other hand, that bit about getting a way to run your own ROM is, on average, awful and device-specific. Still, I suspect that it's better than TVs that run completely different OSs to start with. Edit: Oh, and AFAIK a lot of those OSs aren't FOSS in the first place, which makes things harder; at least with Android the vendor is legally obligated to give you the kernel source and then you can stick that in AOSP and probably get a working image. Other than WebOS, I don't think TVs have an equivalent.
Agreed that content support and DRM are a difficulty.
I know that the ink price gouging is part of why printers are weird, but that results in the default firmware being user-hostile in a way that I would have expected to make people want aftermarket firmware all the more. I suspect that the difficulty there is also something about fragmentation and perhaps a lack of devices that even try to be open (Google publishes all the documentation and source to build AOSP for a Pixel phone; I'm aware of zero printer vendors doing so).
If the aftermarket firmware doesn't support Netflix, then users will have to choose between not having Netflix and having the features the aftermarket firmware enables.
If that aftermarket firmware enables better playback of pirated content, and the Netflix catalog deteriorates more and more with time, it's more and more likely that of the people who'd consider alternative firmware in the first place, more and more will decide against Netflix in that scenario.
Not that that may be enough people for Netflix to care, but it could easily be enough that "no netflix" is no longer a blocker for acceptance of the alternative firmware.
I'm out of my league here but could it be that the tech behind TVs is changing too rapidly? LCD, LED, OLDED, QLED, MINILED, and all sub-differences in each generation. Plus all the brands: SAMSUNG, LG, BOE, ect. Plus all the different screen dimentions and levels of HD, dolby vision ect. Whereas with smartphones, everyone is basically using a samsung display, snapdragon or samsung processor and android. Could this be a factor in why there is so little third party implementations? This could be something Khronos could work to standardize; that is, if display makers want it, which I doubt)
I always figured that the TV design is still "panel with one of a narrow range of common interfaces, and the smarts being on the controller."
Otherwise it would be difficult for small manufacturers-- who are typically not custom-ordering panels, and likely switching between vendors based on best-offer-of-the-day-- to integrate in a reasonable manner.
Most likely, especially for non android devices. Most TVs have barely few years of manufacturer support.
Printers at the very least have excuse of a lot of hardware specific code to drive the head but TVs are pretty much tuner + screen + internet connection.
Also people wanting OSS on their TV tend to just use TV as dumb screen and connect the whatever small device with HDMI they have on their hands and not add "runs this board" to the analysis paralysis
Our TCL Roku TV is slow as molasses to turn on and do stuff. Since we solely use it with an Apple TV device, I'd LOVE firmware that simply booted fast and allowed me to switch between inputs.
Yeah, that's the problem, you're essentially getting less hardware for more money. And if you have to source your own SBC might as well just pick the cheapest TV and connect it via HDMI...
Very nice hack. Knowing all the random stuff that basically has an SBC inside makes me wish it were easier to get at least that piece out and use once the device is toast. Especially since many Raspberry Pis wind up in end products anyway!
I see a lot of broken TVs on the curb, and haven't been thinking about reusable SBCs in them. I might have to start.
(I used to carry a pair of screwdrivers when curb-shopping, mainly for parting-out tower PCs that aren't worth carrying the whole thing back, but I got out of that hobby.)
Hmm, sort of feels to me like if I want some parts of a free-box machine it becomes my problem to dispose of the rest... at least, once I've pulled it apart it becomes even less likely that anyone else will take what's left...
I understand what you're saying, and was aware of that (I saw some questionable behavior by others), and operated responsibly. I can't get into a long description of it right now.
This TV was fairly modular, as you can see: 5V/12V power supply, Wifi/BT, IrDA are all detachable. The boards you have no use of could be disassembled for parts, too.
heh, reminds me of when my friend was going to throw one out a few years ago. It was pretty ancient, but I sold the video card for $40 because the crypto-boom ate up the supply of new cards and even a basic card was hard to find.
I had a brief stint as a technician repairing Foxtel IQ set-top boxes (Pay TV). I was just as surprised to find a UART present on the scart connector too! I guess it's relatively common
There's no mystery here, so I didn't mention it in the write-up. The SoC has an LVDS output. The screen has a very long PCB with a specialised controller; it accepts LVDS and drives the pixels and backlight.
I think that the display is not used at all. If I understood correctly, the scart connector is hosting a hidden uart interface. This let's you connect a serial terminal via three cables. (You can e.g. buy a serial to usb adapter for 5 bucks on Amazon and use it via putt.)
You are correct. The SoC supposed to have a way to generate composite signal, according to the scarce data I found, but I didn't manage to force the board to do so. LVDS is (presumably) working, but I don't have any compatible LCD panels lying around.
https://github.com/DavidBuchanan314/webos-vncserver/raw/main...
The picture shows a phone VNC'ed into it - it's still running the original Linux-based TV OS (LG webOS), but rooted.