Hacker News Reader

Comment by lovedswain | Hacker News Reader

Do you have a few links to share? Pretty sure you triggered quite a few folk here! Loving the idea of a TV dumbification movement

View on HN

https://www.aliexpress.com/i/4000046044120.html

Reviewed at http://web.archive.org/web/20190511065920/http://redlightgre... (discussed at https://news.ycombinator.com/item?id=24877363 )

https://sites.google.com/site/lcd4hobby/6-lcd-as-pc-vga-hdmi...

https://imall.com/product/LCD-TV-Controller-Driver-Board-MSD...

It's hard to keep track of all the different variants as this seems to be a bunch of Chinese companies all copying each other. You can find these boards under names such as MSD3663/3463 DS3663LUA, TSUMV53/V56/V59, etc.

I went to that google link and this popped up.

Ethereum Phishing Detection

> This domain is currently on the Crypto Wallets domain warning list. This means that based on information available to us, Crypto Wallets believes this domain could currently compromise your security and, as an added safety feature, Crypto Wallets has restricted access to the site. To override this, please read the rest of this warning for instructions on how to continue at your own risk.

"oh noes! someone on the internet is trying to con me!!!"

guess an 3rd-party script- and ad-blocker is the way to go

I'm using Brave with uBlock Origin.

I wonder if there are any of these with VRR and lower latency. I'd like to upgrade my TV but it seems pretty wasteful to throw out the panel when it's mostly the firmware that's the problem.

I've sort of dumbified 2 smart TVs by coupling them with cheap small computers with ethernet or WiFI in and HDMI out. I don't know absolutely that the TV's are not spying on us, but I can get to the internet going around the TVs not through them.

If you're already down that route, why have your TVs connected to your WiFi at all?

How do you know the TVs don't connect to any open WiFi network, or simply get shipped with embedded 4G radios ?

I used to think that would be too expensive but then I saw smartphones sell new for USD 20 on Black Friday and I also see one year of 200 MB a month data plans selling for USD 30.

That’s retail.

I have changed my mind on this: if you were selling a million televisions a month, you could absolutely make it work from a money standpoint. They don’t need to send back tons of data and they don’t need to do it particularly fast.

I expect with a lot of these kinds of use cases you could get absolutely rock bottom volume m2m prices by promising to only use it during designated quiet periods when the network is underutilized.

I think you vastly overestimate how progressive telcos are with this kind of thing ...

Maybe, but I feel like there'd be a pretty easy calculus on it at scale— like "your competitor is cheaper. We're going with them unless you can cut your rate in half. One way you can do that is by giving us SIMs that only unlock for an hour once every 24 hours, and you can decide when that hour is, as long as we see our devices at least once a day."

This should be pretty easy to implement on the telco side and it would let them monetize the dead time in the network, same as TOU pricing for electricity.

If you ship a device with a SIM card you then need to follow regulations to ship it. This means you, the buyer, can find out from the FCC what spectrums it speaks on. I don’t think mobile connectivity is a much needed option on TVs so you won’t find it. Connecting to open WiFi hotspots is a completely different beast and doesn’t leave behind a tell-tale trace

You would think it would be easy to implement ... technically ... but one has to remember that as well as being businesses many Telcos are large hulking great slow moving monsters draped with red tape and politics - anything that changes the existing ways of doing things is often a Herculean act and that strips any dynamism in the market. Don’t misunderstand me there are many discussions in the industry around realising the kind of utopian vision you espouse but these are all bound up in committees with a typical lifespan of > 10 years ...

Well, I'll take your word for it, as it sounds like you have first hand experience and I'm just speculating. But it doesn't feel like this should be the kind of thing that requires a standards committee to be involved— it's literally just a SIM whose ID is mapped on the backend to reject connections most of the time, or be so heavily throttled as to be practically unusable. There's no new protocol needed here, just a profile added to existing traffic shaping/limiting systems, all of which already have time of day awareness in support of commonly-available features like free data after 5pm or whatever.

OTOH, perhaps it's simply a question of an insufficiently large addressable market for non-realtime data like this. Maybe there's room for an M2M-focused MVNO to negotiate a block of off-hours data for dirt cheap and offer this kind of service, maybe paired with some kind managed platform to handle the storage and transmission side.

> it doesn't feel like

You’d think ha. Economists have a term for this “deadweight loss”.

MVNO are probably the best place for this kind of dynamism but they are always limited by the tools that the tier 1 carriers give them.

There is stuff like you describe on the way but progress is glacial. You’ll get bits and pieces dripped out bit by bit as the big guys satisfy themselves they’re not missing out on anything.

Just a note that this was one of the puzzles Steve Jobs managed to crack. We had smart phones before the iPhones but progress was stymied by the industry - thankfully jobs recognised the potential and had the wherewithal to put the Telcos over his knee and make them his bitch.

Steve is gone now, and there aren’t too many players like that around now and you can see how gradually the smartphone market has started to tend back towards the more orthodox again.

A 3g module was $2.50 COGS in 2013 when I was looking into an IoT startup.

And data and devices will be even cheaper in another 12-18 months once LoRa is rolled out nationwide in the US.

https://en.m.wikipedia.org/wiki/LoRa

You don't, but connecting them to a WiFi won't solve that problem because they can (and probably do) switch to radio if they can't reach their servers through Wifi connection.

If there's a cellular radio. it has to be declared to the FCC. You can look up your TV's FCC ID to verify that it doesn't have one. I'm not aware of any TV that has an embedded cellular radio.

Panels tend to have one of a few common interface standards: https://www.digitalview.com/blog/lcd-panel-interfaces-a-quic...

I recently came into a pile of V-by-one panels, and discovered boards for them are significantly harder to come by than their eDP or LVDS brethren. But I've found a few on aliexpress and we're up and running:

https://www.aliexpress.com/item/32860791745.html

The nice thing about VBO is the connector is more-or-less standardized, and it's a very durable JAE series, so plugging and unplugging isn't such an issue. (Prototyping with LVDS or MIPI DSI tends to run into fatigue issues unless you're scrupulously gentle.)

Beyond the panel interface, there's also the matter of the backlight. Sometimes the backlight driver is separate, sometimes it's built into the scaler board. The connectors for these are almost always evil and you have to hope your supplier can include the cables, because you'll never find them yourself.

I would loooooove to find a source for the SDK to build the software that runs in the scaler chip. When I buy the scaler boards, they come with random firmware that the supplier isn't particularly interested in updating. Other builds of the same board support DDC/CI but don't have the backlight driver on 'em, and cost more. There's so much stuff about the firmware I'd like to change...

Reply on news.ycombinator.com