7 ms·
I'm curious -- what's the reason for something seemingly as simple as a Lightning cable to have chips inside of it? For most of computing history, I was genera
by crazygringo 6y ago
I'm curious -- what's the reason for something seemingly as simple as a Lightning cable to have chips inside of it?
For most of computing history, I was generally under the impression that cables were "dumb" -- each pin connects to a wire which connects to a pin on the other end. There's fancy bundling and twisting and whatever involved, but it's still ultimately just conductive wires.
When did cables start getting chips, and why? Did Apple start it or somebody else? Is it solely to try to prevent third-party manufacturers? Is it for the cable to announce to the port that it supports certain specifications of power or data? (But why would that require a chip instead of just some kind of "dumb" extra pin that has some basic electrical property that can be read?) Is it something else?
I mean I understand why certain dongles have chips, because they're connecting between sets of pins that aren't 1-1 or even in the same data format at all. Or why the same might be true for USB-C to Lightning.
But for cables to go from "dumb" to "smart" seems like it kind of breaks all expectations of what a cable even does, and therefore how a consumer will even know what to purchase -- which, of course, has famously been a HUGE issue with USB-C.
Would it be better for us to go back to dumb cables without chips? Or are there good reasons why this is the future, where at some point we'll expect all computer cables to have them?
- userbinator 6y agoIs it solely to try to prevent third-party manufacturers? Is it for the cable to announce to the port that it supports certain specifications of power or data? (But why would that require a chip instead of just some kind of "dumb" extra pin that has some basic electrical property that can be read?) Is it something else? My theory is that it started as the former and then they realised it would seem less suspicious and more friendly to also use it for the latter.
- nikanj 6y agoIn a nutshell, the answer is that the Lightning cables just seem simple, but they're not.
- mmastrac 6y agoLightning added chips that weren't necessary IMO. You don't need smart control over the low voltages/power that go over a lightning cable. Lightning chips were likely more about control. USB-C _does_ require e-marked cables in certain cases where passing large amounts of power over a thin cable could be dangerous. I don't believe these chips are proprietary nor expensive - just marking silicon that replaced the old resistor system that was somewhat brittle and often wrong.
- jorvi 6y agoThunderbolt 3 (and probably USB4) requires chipped ('active') cables at lengths over 0.5m, but considering you're pumping 40Gbit of data over it, it kind of makes sense. For Lightning, yeah, there's basically no reason except for the 'Made For iPhone' program, and it isn't even effective at that since Chinese cable makers just clone the keys of the official stuff.
- nitrogen 6y agoCouldn't whatever EQ/delay compensation is performed by the chip just be moved to the PHY?
- p_l 6y agoThe chips in MFI program were pretty much DRM.
- userbinator 6y agoIndeed, I believe 40Gbit Ethernet doesn't need active cabling either.
- p_l 6y agoNot exactly - at very short ranges, you can use DAC cabling, "Direct Attachment Cable". Those directly connect SFP signals, but afaik they still have circuitry inside its just easier because the cable is known length and characteristics - in exchange the cable is rather thick and unwieldy. Generally it works well inside rack for connecting servers to ToR switches and that's it. 40Gbit DAC is even more annoying because unless the cable is a bit more active, it's essentially 4 10gbit DAC in one sheath.
- leecb 6y agoThe chip's responses are how the iPhone knows what device / cable type is connected to the port- at least that's what I understand from this.
- cortesoft 6y agoWhy wouldn't the device ask the port directly? A chip in the cable means that the device asks the cable, which then asks the device.... why not cut out the middleman?
- WrtCdEvrydy 6y agoIf the device is dead, then it can't ask the port. This is also tied to fast charging (and whether the device will support it)
- cortesoft 6y agoCouldn't it provide the low power until the device is powered enough to answer the fast charging question?
- cortesoft 6y agoAlso, wait... if the device is dead, how does the CABLE know that fast charging is supported by the device?
- warricksothr 6y agoI think the link describes this tangentially. The minimal 10mw - 15mw current that high-five allows on a cable that hasn’t completed a handshake is enough to charge the capacitor for secure boot and trident which then can handshake with 0x74 to receive 5v 1a source. When the kernel boots the rest of the handshake process can be completed to increase the current. That’s at least my understanding from the document.
- derefr 6y agoOne use-case comes up in the article: the chip in the cable is able to probe information from the device even when the device is unpowered, essentially acting as a microcontroller grabbing some of the device's less-guarded peripherals for itself.
- liminalsunset 6y agoLightning cables have exposed contacts - While at 5V, this does not pose a significant shock risk, if full current were available, brushing the connector against metal could result in sparks and damage to the connector or power supply. Additionally, if the connector gets wet, the DC voltage will result in electrolytic corrosion. This still does happen even with the limited current provided before the handshake, just slower. (Don't!!!) try charging a phone with official cables and a power bank in the rain or letting a plugged in lighting cord's lightning end sit in some soup Cables with electronics in them are useful for protections like this, and to facilitate safe interoperability of multiple voltage and current sources and sinks like on USB Type-C. The question becomes, is the protocol implemented in a simple, open consumer friendly way or is it implemented with other nasty antifeatures.
- jcrawfordor 6y agoCouldn't all of the protections on current delivery be implemented in the interface on the other end of the cable, though? This is the approach taken by basically every other standard for negotiated power delivery. Even things like 802.3af which manage to deliver 48v at nearly an amp and with the complicating case of many valid devices not really tolerating any current delivered.
- sjwright 6y agoThe other end of the cable is generic USB and can often be a naked, logic-free 5V supply.
- jojobas 6y agoHow would you maintain such a margin on the cables if they could be "dumb"?
- jxi 6y agoHopefully this is sarcasm.
- neotek 6y agoThe last time I looked, the money Apple earned from the entirety of the MFi program — not just cables, but every single third-party Apple accessory on Earth — made up less than one half of one percent of their annual revenue. It's such a vanishingly small part of their business. It's far more likely that the purpose of MFi is to make sure that any officially licensed accessories aren't pieces of shit that Apple will cop the blame for. Most users aren't going to blame the cable manufacturer when their phone won't charge properly, since the vast majority of people have no idea about the complexities of modern charging protocols, so they're going to mistakenly think the phone is at fault and blame Apple.
- williamdclt 6y ago> every single third-party Apple accessory on Earth made up less than one half of one percent of their annual revenue That makes me even more angry about the ridiculous price of these accessories
- rconti 6y agoConsidering they're already being ripped off anyway, and most consumers don't know they're "smart", the margins would be higher selling a "dumb" cable that costs less to make, at the same price as they sell them at today.
- PinguTS 6y agoIn a nutshell: EMI (Electromagnetic Interference) is the main reason why cables got smart in the first place. Many people in IT think, that data is just 1s and 0s. But in reality we are living in a physical world and as such some form of power is used to transmit those 1s and 0s. This power may is electrical power in terms of voltages with defined generated current or in terms of current with defined voltages (20mA communication heavily seen in industrial equipment). If you have a slow communication link, EMI and other environmental thinks like temperature is not much of an issue. But as faster you like to communicate you need to cope with those things. The electrical contacts in the connectors doesn't help. So people started to put the transceivers into the cables. Those transceivers with additional electronics like ringing suppression circuitry and other "magic" are then best fitted for the desired cables for the desired environments. Then you started to makes those cables smart like to detect the capabilities of the cables like is it really possible to talk at 10Gbps or is it more a cheaper type where 100Mbps is the max. As more smarter those cables get people get creative to put other things in it like "hey, the cable could become a converter as well".
- alkonaut 6y ago> In a nutshell: EMI (Electromagnetic Interference) is the main reason why cables got smart in the first place. Considering the main use case (almost exclusive use case) for lightning is for charging and nothing else, I wonder why they have to be so complicated? Shouldn't charging cables work with Vcc/Gnd only wired and a dumb power 5V/2A source? Why is my phone doing some complex detection/negotiation of cables/chargers when I plug it in? Why is a charger considered an "acceessory"?
- dijit 6y agoPrimary use-case for a smart phone is the web-browser and networking array. That doesn’t mean you can remove the microphone. Optimising the common case in this instance will remove all other possibilities.
- alkonaut 6y agoYou don't have to remove negotiation and complexity from the system when it's used for data. My question isn't why those systems are there so much as why they are needed for the charging use case. E.g.: why do chargers even need more pins wired than +5V and ground? Is there really such a large benefit from making chargers "smart"?