Apple's Lightning Connector Uses Adaptive Technology to Dynamically Assign Pin Functions

[jclardy]

I mean, wouldn't it be practical to have the same connection across all devices, that can do data and power, like was touted? Or is it not thin enough?

Thunderbolt is not thin enough, plus it would create really expensive cables as essentially the TB cable takes care of transmission, which is why it can eventually be used with optical cables rather than copper. They just have to have the transcoder built into the cable.

For an iPhone you don't want a super expensive cable, which it would be if you wanted a thunderbolt -> USB cable. And TB->TB would exclude way too many people (Everyone without a 2010 or newer mac.)

Has anyone done a teardown of the lighting cable?

 

[UnSainted]

Intel

I mean, wouldn't it be practical to have the same connection across all devices, that can do data and power, like was touted? Or is it not thin enough?

Thunderbolt is a joint development with Intel, probably would not work with an ARM based SOC.

 

[D.T.]

Think about how you combine USB OTG and HDMI output signals over a single 8-pin + shield Lightning or standard 10-pin + shield USB 3.0 Micro AB connector...

There's not enough pins to map them 1 to 1. Even if you dynamically switch signals, there's still not enough to fully support HDMI. The easiest way to do this is to multiplex the signals, dump them through a high-speed SerDes, and then sort them out on the other side.

The two competing technologies for doing exactly this for smartphones are MHL and MyDP. MyDP is based on DisplayPort which Apple is fond of, is royalty free, and does not specify a physical connector... Gee, I wonder what Lightning is?

Good stuff, nice contribution.

I have this head unit.... http://www.amazon.com/gp/product/B0039YXNVC/

and it works great with it. Just as it did with the 30pin cable

Oh yeah? I thought the "iPod mode" that HUs used for controlling the device wasn't available via the adapter[?] I have an Alpine X100 HU that controls an iPod/iPhone and provides display of the music/playlist/art/etc., through the HU ... figured that wouldn't work with an adapter.

 

[everything-i]

Now we know why the adapters are expensive, they aren't just a couple of simple connecters wired back to back. I have to give Apple credit, this is a pretty innovative concept, this new adapter can essentially evolve into anything while using the same physical connector. The change is a pain but its a good change that shouldn't need to be done again for a very long time.

 

[vikingdave]

The phone is probably doing the adaption

Interesting article, but I'm betting that the circuitry in the phone is actually adapting and detecting/changing pin assignments, not the cable itself.

 

[Navdakilla]

I don't get it.

me either.

seems cool tho

 

[Geckotek]

Interesting article, but I'm betting that the circuitry in the phone is actually adapting and detecting/changing pin assignments, not the cable itself.

Likely based on a signal from the cable.

 

[citi]

I mean, wouldn't it be practical to have the same connection across all devices, that can do data and power, like was touted? Or is it not thin enough?

Thunderbolt is not on PCs so it can't be the standard. It actually makes more since to make it USB 3/USB 2 backwards Compatible. There new Macs just came out with them. That's bizarre to me.

 

[croooow]

I guess we now have the answer to this question:

"I heard the connector's all digital! What does that even mean!?!?!"

That sums up Samsung: Make fun of Apple for something that is actually beneficial because most consumers would not really understand it.

If the 30 pin connector had "Adaptive Technology" Apple could have avoided that whole "Firewire/USB" charging switch that left a lot of iPod accessories useless a few years back.

 

[jrswizzle]

New lightning cable which dynamically adapts to different signals sent by different peripheral devices. Adaptable to a wide variety of peripherals - even ones developed down the road in the future.

vs.

"flip to mute"


Hmmm.....

 

[dguisinger]

On a quick note, future variations could make use of both sides of the connector, and the pins could become decoupled. It would require more complex circuitry in the plug and a updated connector on the phones, but that would allow a singular design today to be updated in the future with 2x the pins.

For people interested in the more technical implications behind this move:

People fail to grasp the reason USB 3.0 has so many pins. Its because the signals between the analog USB 2 serial interface and the SerDes interface aren't compatible, you'd blow out the SerDes drivers if you plugged in something that used USB 2.0... hence they doubled the pins and ran the signals side-by-side.

Interestingly enough, the USB interface with the iPhone no longer needs to fully reside on the phone. A chip could determine in the cable what you are plugged into at the host side and act as the physical interface layer, eliminating the need for the USB 2 pins going to the phone. Since we already know the phone supports HDMI over the connector, we already know the pins use a multi-gigabit SerDes.

What standards use multi-gigabit serdes interfaces (which are very standard)?

PCIe
USB 3.0
Ethernet (MAC to PHY)
SATA
HDMI
DisplayPort
and.....
ThunderBolt... as Thunderbolt is an multiplexed version of PCIe and DisplayPort.

Why does Apple call it a Lightening connector?
Maybe it supports a single lane of Thunderbolt in future hardware (or even current hardware). The DP/TB connector is dual-lane... a single lane would still provide a HUGE increase in throughput; you would likely never need more than that for interfacing a mobile device.

This connector, along with the ability to put physical layer chips in the connector, opens up the world to doing almost anything with the phone.

The real question I have is what is the SerDes hardware like on the phone side? I'm assuming most of these interfaces aren't supported due to lack of physical hardware. Without an FPGA or physical hardware, things like SATA, USB3, Ethernet, etc wouldn't be practical from software implementation... but the possibility for more and more protocols with future phones/devices is intriguing.

 

[scootermafia]

Read the article. The pins are continuous from one side to the other.

No they are not. Look at the picture I drew (I'm Peter from DHC). Some of the pins are continuous top to bottom, others (like USB V-) are mapped so that when you flip the connector, they are still going into the same contact on the jack. Top pin 1 goes to bottom pin 8, etc.

 

[dguisinger]

Interesting article, but I'm betting that the circuitry in the phone is actually adapting and detecting/changing pin assignments, not the cable itself.

Actually, its been shown there are chips within the connector of the cable. The cables are intelligent.

 

[verniesgarden]

I mean, wouldn't it be practical to have the same connection across all devices, that can do data and power, like was touted? Or is it not thin enough?

not enough macs have thunderbolt. same with usb3

 

[repoman27]

Oh yeah? I thought the "iPod mode" that HUs used for controlling the device wasn't available via the adapter[?] I have an Alpine X100 HU that controls an iPod/iPhone and provides display of the music/playlist/art/etc., through the HU ... figured that wouldn't work with an adapter.

The Lightning to 30-pin dock connector adapter supports USB charge/sync and analog audio. It does not support either analog or digital video output. "iPod display mode" was a way of outputting what was on the iDevices screen as digital video so it could be reproduced and used by the accessory. Basically, as long as your 30-pin gear only requires USB or analog audio, it should work with the adapter.

 

[D.T.]

The cables are intelligent.

That would explain why I found my cable on the sofa next to an open book on string theory...

 

[DotCom2]

To tell the truth, I was a little scared when I first saw it.
It seemed so tiny and fragile. Hope this is not the case.
I wonder how it's gonna do with the big honkin' iPad once it gets it which I'm sure will be on the next ipad interation.

 

[saturn88]

Intelligent cable: the computer chip is there so you don't have to flip it! What a great innovation. Can I get AppleCare with it?

Sounds like a typical American solution. Reminds me a story when Americans spent $10 million to develop a pen for astronauts. Russians just used pencil.

So get your credit card ready to pay $30 for each cable instead of $0.99 for USB.

 

[repoman27]

What standards use multi-gigabit serdes interfaces (which are very standard)?

PCIe
USB 3.0
Ethernet (MAC to PHY)
SATA
HDMI
DisplayPort
and.....
ThunderBolt... as Thunderbolt is an multiplexed version of PCIe and DisplayPort.

You missed the two most likely candidates: MHL and MyDP.

 

[D.T.]

The Lightning to 30-pin dock connector adapter supports USB charge/sync and analog audio. It does not support either analog or digital video output. "iPod display mode" was a way of outputting what was on the iDevices screen as digital video so it could be reproduced and used by the accessory. Basically, as long as your 30-pin gear only requires USB or analog audio, it should work with the adapter.

OK, that makes sense ... I believe the display is generated from the catalog on the phone via USB (I think it caches it during the initial handshake with the device), then it's my understanding the IDA series HUs use the digital out (vs. analog, i.e., bypassing the devices internal DAC).

Thanks. Some great content you're posting too!

 

[JAT]

I'm hoping they can make the iPad charge faster once it's implemented on the device.

If you charge the Li-Ion batteries too fast they'll heat up too much and heat is bad.... I'd be surprised if they can go much faster without harming the life of the batteries.

It's the charging system in the devices that would need an upgrade, partly for cooling. I haven't done enough research, but it does seem these are relatively inefficient. But that's a problem with the size of tablets, none of them are big enough to dissipate that heat. Possibly if they had a more efficient pattern to which cells charge, when, but they may have already addressed that. Don't know.

This is really cool, but it also seems like it could easily malfunction/get confused if there is some wear and tear involved. I guess we'll have to wait and see about the long time reliability of these cables.

Same as any other cable.

 

[Kaibelf]

I think the problem with the iPhone 5 is that it doesn't feel like a stellar leap. The biggest changes are on the inside and this isn't really that tangible to your average consumer.

At least 5 million average consumers disagree with you.

 

[Yr Blues]

Intelligent cable: the computer chip is there so you don't have to flip it! What a great innovation. Can I get AppleCare with it?

Sounds like a typical American solution. Reminds me a story when Americans spent $10 million to develop a pen for astronauts. Russians just used pencil.

So get your credit card ready to pay $30 for each cable instead of $0.99 for USB.

actually, when fumbling in the dark ....... connecting it correctly isn't 50/50 anymore

it's a time saver and doesn't risk damaging the ports

people will want this feature on every cable for now on

 

[Kaibelf]

Intelligent cable: the computer chip is there so you don't have to flip it! What a great innovation. Can I get AppleCare with it?

Sounds like a typical American solution. Reminds me a story when Americans spent $10 million to develop a pen for astronauts. Russians just used pencil.

So get your credit card ready to pay $30 for each cable instead of $0.99 for USB.

I recall Russians standing in line for 17 hours for a loaf of bread too. What's your point?

 

[dguisinger]

You missed the two most likely candidates: MHL and MyDP.

I would say its neither of them; because its a group of raw SerDes signals, nothing more. MHL and MyDP are function specific and have specific connectors.

You missed my point.

You could implement MHL and MyDP through chips in the cable I'm sure, because they are built upon standard protocols.

I'm referring to technologies that use SerDes links as their physical layer of communications. Most high speed protocols these days use it, therefore, depending on the guts of the phone or future phones/ipads, the cables should have access to many of those interfaces I listed.

For example, when you come from the FPGA world you can see the advantage. With FPGAs, you have the ability to connect to generic SerDes ports on the chip with custom logic, you can build your own SATA controller, PCIe bridge, ThunderBolt controller, etc and wire them up on the fly to a SerDes interface. Once you realize how powerful it is to have SerDes transceivers, the future opens up to a lot of possibilities - even if the current hardware doesn't let you use them all at this time.