Hidden Lightning connector found in Apple TV 4K ethernet port

AppleInsider

A closer inspection of Apple TV 4K's internals reveals Apple's streaming box includes what appears to be Lightning connector terminals hidden in its ethernet port, potentially providing a direct path to critical system hardware.

In a review of iFixit's 2017 teardown of Apple TV 4K, Apple TV hardware and tvOS specialist Kevin Bradley on Thursday spotted solder joints on the device's logic board that correspond to Apple's Lightning connector protocol.

Bradley, better known by his Twitter handle nitoTV, suggests the Lightning port might be used to gain access to Apple TV 4K firmware, a route that could lead to a jailbreak.

"[N]one of us looked THAT closely to the hardware of the AppleTV 4K [sic] and the magic locked in the ethernet port until fairly recently. its [sic] going to take time to figure out what is possible and how," Bradley said in a tweet.

A subsequent tweet from self-described "maker" Steven Barker reveals Apple slyly disguised external access to the Lightning contacts with sliding slat at the back of the ethernet port. Shifting the door up exposes the set of pins, but it appears that a specialized connector is required to interface with the terminal.

Developer Steven Troughton-Smith believes the Lightning pins are used by Apple for hardware debugging. The company has a history of incorporating similar "hidden" connectivity elements in its hardware for internal purposes.

For example, Apple secreted away an Apple Watch data port in a groove designed to accept watch straps. The company has since utilized the port for diagnostics and connection to an interactive pedestal previously used to display the device in Apple stores.

What, if anything, can be accomplished by accessing the Apple TV 4K terminal is at this point unknown, but the presence of Lightning is an interesting find considering the device was released more than two years ago.

coolfactor

Creative! A great example of Apple hiding ports that are not meant to be used by consumers.

Now only if they would put *some* ports on the FRONT or side of their computers for easier access. 🤣

rhythmagic

Lol you just had to open my bag ; )

apple ][

That's weird. The Apple TV 4k was released a long time ago, and I'm sure that various sites did teardowns after it was released, and this went undiscovered until now?

I guess those teardowns weren't as in depth as they could have been.

edited October 2019

nllarsen

The Lightning connector has 8 pins, but the image in the tweet only has 7 solder points. How does that work?

netrox

Ok, how the heck did they not see the hidden port for the past two years?!?!

d_2

nllarsen said:

The Lightning connector has 8 pins, but the image in the tweet only has 7 solder points. How does that work?

It’s a custom adaptation that likely doesn’t require the use of pin 5 / power.

cornchip

ok. that. is prettyfukkin cool.

dewme

It would be very cool if Apple built a version of the Apple TV that served in the originally intended role of the Raspberry Pi, which is to provide a low cost platform to learn programming and computer controlled automation. The programming part is pretty straightforward and would of course be focused on Swift, but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support, and potentially direct support for one of the popular Industrial Ethernet protocols like PROFINET, EtherCAT, or EtherNet/IP. The Apple TV also has HDMI so connecting a screen is no big deal. Expose the hidden Lightning and enable the USB so connecting a keyboard and mouse is no big deal. It seems like 95% of the hardware work is already done.

This is very doable and would be a huge benefit for high school and above technical/vocational education and a boon for hobbyists. Call it a Mac Micro or HomeKit Homebrew Kit, or whatever, but I'd be first in line to give it a try.

swineone

dewme said:

[...] but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support [...]

I'm very familiar with embedded systems development, and I can tell you that, short of playings with LEDs and push-buttons and such, a remote GPIO gateway is useless. If you need to talk to an I2C or SPI device, you'll be toggling pins every few microseconds, hundreds of nanoseconds, or even tens of nanoseconds. The latency introduced by an Ethernet connection would make this completely unfeasible.

dewme

swineone said:

dewme said:

[...] but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support [...]

I'm very familiar with embedded systems development, and I can tell you that, short of playings with LEDs and push-buttons and such, a remote GPIO gateway is useless. If you need to talk to an I2C or SPI device, you'll be toggling pins every few microseconds, hundreds of nanoseconds, or even tens of nanoseconds. The latency introduced by an Ethernet connection would make this completely unfeasible.

I'm talking a $100.00 learning tool for educational, vocational, and hobbyist use, not a high speed controller for running a paper machine, high speed sortation, or chemical reactor. To a budding programmer, seeing some LEDs blink or a motor spin in response to a program that you wrote is a huge learning experience.

macpro

Searching my 2013 Mac Pro for a hidden TB3 port ... /kidding

macgui

apple ][ said:

That's weird. The Apple TV 4k was released a long time ago, and I'm sure that various sites did teardowns after it was released, and this went undiscovered until now?

I guess those teardowns weren't as in depth as they could have been.

Maybe, just maybe Apple was a little more clever than you want to give them credit for.

netrox said:

Ok, how the heck did they not see the hidden port for the past two years?!?!

Again a dumbass question. How 'bout Apple's aforementioned cleverness? Like you would have spotted it. As fucking if.

Monday morning quarterback hacks.

cornchip said:

ok. that. is prettyfukkin cool.

This ^^. It absolutely is.

dysamoria

Way to go with the hostile commentary 12 comments in.

edited October 2019

lilstevie

macgui said:

netrox said:

Ok, how the heck did they not see the hidden port for the past two years?!?!

Again a dumbass question. How 'bout Apple's aforementioned cleverness? Like you would have spotted it. As fucking if.

I can say with absolute certainty that I would have spotted it, and I ask the exact same question. I went in knowing that Apple must have some way, looked at the board and saw there was too many signals leading into the ethernet port and just had to know how it worked.

esquirecats

Since Apple appear to use some clever non-standard connectors for debugging/repair/in store demos. It does raise a few questions about user repairability:

• Should Apple be required to provide the custom port cord/dongle to home/unqualified repairers?
• Should the proposed "right to repair" control the design of the object (e.g. forcing apple to use certain parts or standard ports for diagnostics), and similarly what truly defines user-repairability, since Apple's designs rely on skilled repairers.

I think this provides a clearer view that Apple devices have a lot going on in them, for even the relatively simpler ones.

macman1984

Wouldn't have been nice if Apple had included an non-hidden lightning port so we could all charge our remotes directly from the Apple TV instead of having to connect it to some other device/wall socket to charge it?  Seriously, if it's there anyway, couldn't they have let us use it?

crowley

macman1984 said:

Wouldn't have been nice if Apple had included an non-hidden lightning port so we could all charge our remotes directly from the Apple TV instead of having to connect it to some other device/wall socket to charge it?  Seriously, if it's there anyway, couldn't they have let us use it?

There is no lighting to lightning cable sold by Apple though?  USB-C would be better, older Apple TVs had a USB port.

wizard69

dewme said:

It would be very cool if Apple built a version of the Apple TV that served in the originally intended role of the Raspberry Pi, which is to provide a low cost platform to learn programming and computer controlled automation. The programming part is pretty straightforward and would of course be focused on Swift, but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support, and potentially direct support for one of the popular Industrial Ethernet protocols like PROFINET, EtherCAT, or EtherNet/IP. The Apple TV also has HDMI so connecting a screen is no big deal. Expose the hidden Lightning and enable the USB so connecting a keyboard and mouse is no big deal. It seems like 95% of the hardware work is already done.

This is very doable and would be a huge benefit for high school and above technical/vocational education and a boon for hobbyists. Call it a Mac Micro or HomeKit Homebrew Kit, or whatever, but I'd be first in line to give it a try.

This is a very good idea but I think you mis the one big advantage Raspberry PI has, that is the hardware I/O header. The options the current Apple TV have just are not good enough, especially for those interested in lower level programming.  That could be easily corrected by Apple in a number of ways.  One is the addition of an I/O header, the other would be running TB out a port to an expansion header.  In any event your general idea is very sound.  

I’m not sure why Apple abandoned education, maybe they became too focused on over priced hardware.  Whatever the cause it would be great to see low cost hardware available to schools in the USA.  Hardware and of course software to teach programming and the basics of embedded hardware.  As you note Apple TV is almost there right now.  

wizard69

dewme said:

swineone said:

dewme said:

[...] but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support [...]

I'm very familiar with embedded systems development, and I can tell you that, short of playings with LEDs and push-buttons and such, a remote GPIO gateway is useless. If you need to talk to an I2C or SPI device, you'll be toggling pins every few microseconds, hundreds of nanoseconds, or even tens of nanoseconds. The latency introduced by an Ethernet connection would make this completely unfeasible.

I'm talking a $100.00 learning tool for educational, vocational, and hobbyist use, not a high speed controller for running a paper machine, high speed sortation, or chemical reactor. To a budding programmer, seeing some LEDs blink or a motor spin in response to a program that you wrote is a huge learning experience.

While I understand your point, from an educational standpoint flashing an LED is only good for one class or learning experience.  If you really want students to learn about low level hardware you need to expose them to serial and parallel I/O solutions.  The big problem is Apples.chips and operating system it being able to handle this low level I/O well.  Driving a hobby servo for example requires that pulse width of the control signal be tightly controlled.  Often on a micro controller this is done with the aid of hardware, I’m not even sure Apples chips can do this.  

What this highlights is the huge advantage more open systems have for education.  Personally I’d like to see Apple offer something to compete against Raspberry PI and frankly Apple TV isn’t far from having the required capabilities.  Getting Apple to address this market though is a huge problem.  Todo so they would need to build a device as open as the Apple 2, that includes full documentation with respect to the SoC and the external schematic.    They have shown zero interest in more than a decade now (actually a lot more than one decade).  So how do we convince Apple to show some interest in education.  

randominternetperson

wizard69 said:

dewme said:

swineone said:

dewme said:

[...] but the Apple TV, unlike the Pi, has no exposed access to GPIO. This is not really an issue because Apple TV already has support for HomeKit based automation, it has Ethernet, and it would be simple enough to build a separate HomeKit gateway/IO controller that provides HomeKit -> IO integration, with both digital and analog IO support [...]

I'm very familiar with embedded systems development, and I can tell you that, short of playings with LEDs and push-buttons and such, a remote GPIO gateway is useless. If you need to talk to an I2C or SPI device, you'll be toggling pins every few microseconds, hundreds of nanoseconds, or even tens of nanoseconds. The latency introduced by an Ethernet connection would make this completely unfeasible.

I'm talking a $100.00 learning tool for educational, vocational, and hobbyist use, not a high speed controller for running a paper machine, high speed sortation, or chemical reactor. To a budding programmer, seeing some LEDs blink or a motor spin in response to a program that you wrote is a huge learning experience.

While I understand your point, from an educational standpoint flashing an LED is only good for one class or learning experience.  If you really want students to learn about low level hardware you need to expose them to serial and parallel I/O solutions.  ...

Who is to say that learning about "low level hardware" is the only valid "education."  Apple has invested (some) in teaching kids how to program, not how to be hardware hackers.  They are doing with Swift Playground (I think that what's it called), but it would slick if they extended that a bit further with an inexpensive hardware device.