Slashdot Mirror
Intel Wants To Eliminate The Headphone Jack And Replace It With USB-C (9to5mac.com)
An anonymous reader writes: With rumors circulating about how Apple may do away with the 3.5 mm headphone jack on its upcoming iPhone 7, Intel has shared a similar desire, citing "industry singling a strong desire to move from analog to digital." Intel believes USB-C is the future audio jack. They believe USB-C has more potential than the 3.5mm audio jack as it allows users to add additional smart features to headphones in the future. For instance, a future pair of headphones could monitor one's pulse or inner-ear temperature for fitness tracking, something that could only be possible if the headphones were connected to a smartphone via a USB-C cable. What's also worth mentioning [quoted from 9to5Mac]: USB-C already supports analog audio transfer through sideband pins simplifying the engineering steps necessary to swap 3.5mm with USB-C in device designs. In the second quarter, Intel should have a finalized USB-C standard for digital audio transfer. Intel does note that the transition from analog to digital will be expensive as the headphones have to include amplifiers and DACs, but scale will offset the early costs over time.
There are reasons to add USB-C connectors to PCs(if nothing else, a baseline USB-C connector is just a USB3 connector that works with USB-C cables; and depending on the situation it can also have additional advantages, as a replacement for various proprietary laptop charge ports, as an alternate-mode video-out, etc.); but what this proposal will do is make it even less predictable(and it's already fairly unpredictable) what a given USB-C port will or will not be capable of.
USB-C supports analog audio through the sideband pins, so a given port might support ordinary passive headphones with nothing more than a mechanical adapter or change in connector. However, on a device with more than one USB-C port, or with USB-C ports that predate this plan, it isn't likely that all the USB-C ports have analog audio, so those passive headphones will only work on certain ports, perhaps none; but 'active' headphones with a USB audio chipset will work on any of them(including USB1.1 or better ports with a mechanical adapter). For extra fun, if USB-C headphones become ubiquitous, even devices without any USB functions will probably want to implement sideband-only USB-C ports, so people can plug headphones into them; but those will only work with passive headphones since they won't actually have a USB host controller.
As with a number of USB-C design decisions, this seems like a pretty good idea if all you care about is bleeding edge cellphones; a troublesome-but-probably-worth-it one if all you care about is cellphones and ultra-skinny laptops; but a morass of confusion and suffering the more broadly you try to make it work. The USB-C port already suffers from the 'might be capable of anything, only actually promises to be capable of almost nothing' and this will only expand that unpleasant aspect.
In principle, the state of USB power delivery is such that this should be doable(with an external dongle of some kind, if the phone has just one USB-C port and the headphones don't have a power plug you obviously need some additional hardware just to have somewhere to plug in the power); the ugly detail is that nobody actually seems to obey those specs yet(as the Google guy on a Quixotic crusade against dodgy USB-C peripherals has discovered you can't even trust a cable to not kill your device on occasion); and when it comes to something more complex like "connect to a phone's USB-C port, accept a DC input and pass through USB-C audio" your mileage will vary, probably enough to make shopping a giant PITA. Until that settles down, odds are that we'll see a lot of enthusiastic cashing in from phone OEMs on the fact that(while nominally 'standardized'/'standards-based') the market is unpredictable and untrustworthy enough that anyone without a moderately techie understanding of USB-C and a masochistic desire to shop by trial and error will basically have to purchase the accessory from whoever they bought their phone from in order to have a reasonable expectation of it actually working.
In the noble world of theory, USB-C can actually be used to do some really cool stuff(something like Microsoft's Lumia dock, while not known to actually be supported on anything except select models of Windows Phones, apparently doesn't require doing anything freaky and nonstandard over the USB-C connector); but the quality varies so widely, and the number of possible combinations is unpredictable enough, that it's hard to make use of the potential without getting burned by crap or sticking exclusively to first-party accessories.
I know this is a sarcasm thread, but 24 bits is actually a lot for an ADC. You're talking 0.2uV/LSB with a 3.3V reference. Even getting close to that requires careful attention to noise sources and PCB layout. 16 bits is pretty hardcore in its own right. 8-, 10-, and 12-bit ADCs are far more common.
Visit the