Intel Proposes to Use USB Type-C Digital Audio Technology | Hacker News Reader

Intel Proposes to Use USB Type-C Digital Audio Technology

anandtech.com

105 pointsnjaremko183 comments

It's recently occurred to me that the prolificness of the 3.5mm audio connector is something to revere. Wikipedia tells that it's been around since 1964, with its fame really coming with the Walkman in 1979. So, a connector introduced over fifty years ago is still in wide use today. How is that even possible? RJ11 phone jacks I think were introduced around the same time, but they seem archaic and old-fashioned in a way the 3.5mm audio jack doesn't. It just astonishes me that the audio connector has been a solved problem for this long, and with all the other advances in tech we've seen.... audio connectors didn't need any improvements.

It makes me very skeptical of any replacements. If a fifty year old connector has been essentially flawless all this time, it's gonna be a tough sell to convince me that something else is really needed now.

An interesting piece about the 3.5mm audio connector from the BBC:

The 19th Century plug that's still being used

"The 3.5mm headphone jack is essentially a 19th Century bit of kit - it is a miniaturised version of the classic quarter-inch jack (6.35mm), which is said to go back as far as 1878."

http://www.bbc.co.uk/news/magazine-35253398

I still remember watching WW2 films and seeing scenes of phone exchanges where people were switching manually ussing the quarter inch jack. It was surprisingly thrilling. I was working during the day at a place where I sold the same type of jack that was in use for similar purposes all the way back in world war 2.

I had no idea their history went back so much further than that.

I have a bunch of old electronics parts inherited from my father, who was a Ham and an EE. There are a few old quarter-inch phone plugs, from the 30s or 40s, in the collection, with classic Bakelite casings. They work just fine in modern equipment.

The old Cannon/Amphenol connectors are still my favorite, and they supposedly go back to the 30s:

https://en.wikipedia.org/wiki/MIL-DTL-5015

Have you ever had to assemble one of those things? Ugh. T_T

Yes! I've assembled probably 100 17 pin ones.

How is that even possible?

Only the computer industry has rapid obsolescence. Other standards aren't nearly so vulnerable to churn. Look at your power sockets. Or your pipe fittings, nuts, bolts, and so on.

Tell that to my dad, who worked as a plumber back in the days of copper pipes etc. These days it is all plastic, so he had to go out of his way to get tools and parts the last time they did anything to their home.

Said home also still have a bunch of ungrounded sockets, because back then regulations didn't require it outside of the kitchen or some such.

> These days it is all plastic, so he had to go out of his way to get tools and parts the last time they did anything to their home.

The main difference between plumbing and other de-facto standards like the mini jack is that plastic components offer some important advantages over the metallic ones, such as price.

It's only natural that obsolete standards are replaced by newer standards when the newer versions offer important improvements.

Isn't that the same with computers? They introduce new connectors to overcome limitations in the old ones. Moore's law is the reason for this I assume.

The article talks about the possible improvements: "USB Type-C Digital Audio".

Moving the DAC and amplifier out of the noise electric environment of a device and closer to the headphone can improve the sound quality. Of course this makes the headphone more expensive but its an improvement in quality for most users. Two examples:

https://www.indiegogo.com/projects/z-ero-the-world-s-1st-dig...

http://www.sony.com/electronics/headband-headphones/mdr-1ada...

Plug sockets are about the only thing in my flat which aren't based on obsolete standards - general consensus is that the blocks were lashed up by cowboys. Which makes it a right bugger when you want to replace a bathroom tile (can't - they stopped making that size about 25 years ago); replace the oven (no-one makes them that narrow any more); replace the toilet (cistern is in another room); etc. etc.

Someone should start if there's that much demand. Good way to make some money.

About the only two things i could see that might need changes for it, are longevity of the connector (make it last more insertions) and maybe making it thinner for devices that don't seem to want battery life in favor of being thinner/lighter.

Even then it'll still be a really hard sell because they've been improving the longevity for a half-century and it'll be hard for something else to come up to that kind of ease and reliability.

I would love a standard that doesn't produce a loud pop when you plug it in to amplified speakers.

I don't know if that can be done. There's going to be two causes to that, capacitance in the line you're plugging in or there being a DC offset from the device. The amplifier should filter out both of those in the long term pretty easily with some capacitive coupling. However that will lead to a sharp peak when you plug it in, that makes the speakers pop.

You might be able to require a sense line however, and then the amplifier can keep a timed relay on the plug before it actually starts amplifying the signal.

My understanding is that 1/8" has this problem because it mates signal connections before ground, which is why its less common in RCA or XLR connections that mate ground at the same time or before signal.

Also, a digital interconnect will not inherently have this problem.

I think it needs to die and be replaced by a standard wireless audio technology of some sort, more efficient than bluetooth. I honestly don't know how people can go about their daily commutes with cables dangling from their ears. To me, that is very uncomfortable. Bluetooth has served me well over the years but I think we need something better and way more energy efficient.

I respect your opinion, but I don't ever want to recharge my headphones or change batteries.

Given that AM radios can be powered by the radio waves, I'll bet small headphones can be powered by small solar cells.

I believe you are talking about crystal radios, where the max volume is a whisper and the sound quality is like a tin can. Above that, it requires a wire stretched several meters to operate, which doesn't scream portable.

They'll be in your ears, so volume isn't the problem. And the sun will likely provide more energy than AM radio unless you're next to the transmitter, even for small solar cells.

If they're in my ears, how do they have sufficient cell surface area and incidence angle? Do I have antennae sticking out the sides of my head like a 1950s movie alien?

20 mW average maximum available power supply for headphones. Add some for the circuitry and radio, I bet you can manage at 500mW.

1368 W/m^2 solar energy on earth on average, or 0.137 W/mm^2 = 137 mW/mm^2. Let's assume 15% solar cell efficiency, so you'll get 20 mW/mm^2 out.

500mW / 20 mW/mm^2 = 25 mm^2. That's how large cells it be exposed to the sun.

Cover both the headphones and cables, and you'll be able to charge any batteries enough to survive the night.

You started out talking about wireless headphones. What cables?

Between your earbuds.

That sounds like solar roadways.

Bluetooth sucks at pairing. A lot of the time I have to try multiple times and it takes me or minute or so to pair. Whereas with headphones connecting takes like two seconds. I want to like Bluetooth but it's not making it easy.

The Plantronics Bluetooth stereo headset I have never fails to connect to my devices on startup. It will connect to two or three at first power up, announcing them as it does so. It has the reverse problem, I then want to signal which device should take priority over the limited bandwidth/processing on the device for audio quality. But I mention it because not all Bluetooth headphones pair to only one device at a time --though all the rest of mine share that limitation. And if USB-C is also an option, pairing fixes could be as easy as the iPad Pro's Pencil pairing: just plug it in and unplug when you want...

I use BT most of the time with a Philips bluetooth audio connector for my stereo. That works most of the time without problem. As long as you use one connection, it works. When I want to use another device to connect to the stereo, that's a problem. Then I suddenly have to reset the connector. And again when I want to go back.

And I honestly can't figure out how people can tolerate connection flakiness, quality issues, and battery drain for a set of headphones. I have tried many BT audio solutions and this is not wanna-be-audiophile pickiness, but the experience is universally substandard as far as I can tell.

Watching a video on my mac, and listening over bluetooth, the audio lags the video by at least 0.5 second, it's unwatchable if there's any sort of dialogue.

That hasn't been my experience at all.

with bluetooth you have a 'choice' between shitty lossy codec and proprietary patent encumbered one.

> How is that even possible? RJ11 phone jacks I think were introduced around the same time, but they seem archaic and old-fashioned in a way the 3.5mm audio jack doesn't.

There hasn't been a major change in the amount of bandwidth or function for headphones.

RJ11 was, and still is, a perfectly fine standard if all you need is two or four wires. But that wasn't enough to meet the requirements for hundreds or thousands of megabits of data.

The bandwidth of the human ear hasn't improved, and over short distances the analog transmission of headphones isn't usually a problem.

The only thing we've added to headphones in the recent years are a limited number of playback control buttons and a microphone, which have evolved the port only a little bit.

Let's lift a glass to Edison base light bulbs, eh?

Edison bulbs are awful. The BNC style ones much of the rest of the world uses are vastly superior.

RJ11 has never seemed archaic to me - I've always treated it as just a smaller RJ45, and it still works fine for a lot of uses (phone landlines, ADSL broadband etc.)

The humble XLR is a much, much better connector ( it locks, it's possibly gas-tight and it has lots of surface area ) , but it wasn't exactly going to fit on a Walk, was it?

Plus, you can actually solder an XLR.

You can definitely solder headphone jack connectors...

How I know this: I've done it at least three times.

But on a decent quality XLR, you can re-solder 20 times, drive a truck over, yank on it until the cable snaps, ... and it will still work beautifully ;-).

Judging from some of cables I've seen, that's also the use that most XLR connectors are exposed to.

> But on a decent quality XLR, you can re-solder 20 times, drive a truck over, yank on it until the cable snaps, ... and it will still work beautifully ;-).

Same for a humble headphone jack, if we're dealing with one which has been designed with that purpose in mind. Just unscrew the metallic cap, unsolder the cables, solder them again, you're good to go.

The XLR's robustness may actually be a problem. When an accident occurs, it's better to just break the 10€ cable than the 1k€ equipment it's connected to.

Indeed. You haven't really worked stage until you've accidentally yanked a cable that's still connected to something on the other end.

There is mini XLR, which should easily fit on a Walk...

That's true. I've only seen pictures of those.

I have one on my headphones

AKGs by any chance?

Indeed :)

I'm sad the K812s moved to a proprietary connector. I have a stash of mini-xlr's

That's not a standard, though. AKG and other companies have different "mini XLR" connectors and they vary by pin number (3 or 4) and pin mapping.

The basic design may still be around, but if you compare the original with one that is used today on modern devices, you will find it divided into may more bands.

This because originally you had mono. Meaning you only needed two wires.

Then came stereo, but you could get away with using three wires.

So with the plug end being designated ground, they divided the upper band in two.

then comes phones and the idea of putting a remote on the headphones wire. "Simple", divide the bands again giving a total of 4.

Thus when you plug it in, you have ground down at the bottom, and then some arbitrary divide of left, right, and remote.

It's because of Moore's law. Power consumption decreases, size decreases, and of course throughput increases as well. When that happens, connectors must change.

How is that even possible?

Take a look at the MIDI connector. It has been around for 30+ years, also, and is still .. ubiquitous!

The answer is, in my opinion, simple: the hardware is still there. Its still quite useful. Connections are forever!

> If a fifty year old connector has been essentially flawless all this time

3.5mm jacks are honestly not that great. Not enough surface area and all the important contacts on the plug are exposed and subject to corrosion and abrasion. It has lasted because it's good enough, not because it's flawless.

Not to mention that it shorts itself out on every plug/unplug. There's a reason this style of connector never took off for any other application.

From everyone who has invested heavily in high-quality audio equipment over the years and understands the dangerous path to DRMed digital, I say "Fuck you Intel!". We don't want your USB-C. We'll happily stick to 3.5mm stereo analog outputs.

The proposal includes stereo analog output.

For now. Think DVI-A.

Great example. VGA and DVI-I are still supported by lots of new video cards and laptops even though analog video has been dead for quite a long time.

Yes, it's a great example, because VGA is broken by the newer standards because of HDCP. As others in this thread have mentioned, HDCP is included in this new USB spec.

There has been a lot of effort over the years to close the so-called "analog hole" at the end of the DRM chain of trust. The stereo output exists only to distract people during the transition period. The analog out hardware may exist in the future, but the software will refuse to send it any data. This is not theoretical; we already see this with Blu-ray software, where you get an error message if you use a VGA monitor without HDCP.

Bonus: with Win 10 forcing updates, once the hardware is common, you won';t get a choice about the update that disables the "analog hole".

As others have pointed out, this may be a means to foist DRM on the audio output of phones, tablets, and possibly desktops/laptops. But the assumption following that is that DRM is being implemented in an attempt to prevent piracy. That is, in my opinion, not the case. DRM's primary purpose with respect to the video/audio industry is market control. Let's start at the top of the market, the content producers (Warner, Fox, etc). They make a movie/TV show/etc, and publish it to disc. But that disc has encryption on it. So a company like Sony wants to make a device that plays the disc. To do so legally they have to sign a bunch of agreements with the owners of the encryption license. Part of those agreements requires the use of HDCP. Okay, so now the disc is playing, and outputting an encrypted video signal. So a company like LG makes a TV, but the video signal is encrypted. So, they have to sign an agreement with the holders of the HDCP license. But the holders of that license, and the license of disc encryption, are all held, ultimately, by the same media industry oligarchy that holds the rights to the content that started this chain of DRM in the first place.

The end result is that the content owners get to use DRM as a means to force all the companies along the food chain to sign agreements with them, and thus they can exercise power over the entire market. Not a single legit Blu-ray player gets manufactured without signing agreements with these companies. Not a single TV, cable box, repeater, receiver, projector, etc. DRM is not a padlock, it's a parasite. The icing on the cake is that it also nets them a tidy profit. HDCP requires both a yearly licensing fee, and a per-device royalty. It ain't cheap. And there are more aggressive requirements if you plan to implement HDCP yourself, rather than using a pre-made device. So you can either use a pre-made device, which is conveniently manufactured by the same oligarchy and is rather pricey, or make your own and suffer further agreements and expenses.

Intel is the guy that the media industry hired to create HDCP, and who currently manages it. It wasn't long ago that Intel was gung-ho about pushing video DRM on PC's along with Microsoft. Luckily that mostly died, but here we are again, same story, different day.

Wouldn't it still be possible to record the non-DRM analog signal with a USB-C to 3.5 mm adapter? The change would put the amp and DAC inside the headphones. Take it apart and connect directly to the "last leg" going to the analog speakers. If anything, it should be a better quality signal because of the shorter analog transmission distance.

An adapter like that will have the exact same problems as DVI->VGA adapters: HDCP. The software will refuse to send the data to the port without an HDCP negotiation. And just like the VGA adapters, I'm sure it will be possible to make a "striper" that fakes the HDCP handshake, with all the associated legal problems.

This is why the media industry needs to get deprecated and fade away.

Their desire for controlling the market and thus a large fraction of the culture is directly harmful. Copyright maximalism has nothing to do with "rights" or "harm" or "piracy", it is instead all about this, to make sure that the largest market entities are dependent on you and follow your commands. They make themselves gatekeepers.

DRM has always been about control of the client, not piracy.

Is USB-C basically the trojan horse for DRM audio? Likely yes, right? HDCP on HDMI out, equivalent on USB-C out?

"Usage of digital audio means that headsets should gain their own amplifiers, DACs and various other logic, which is currently located inside smartphones. Intel proposes to install special multi-function processing units (MPUs) ... The MPUs will also support HDCP technology, hence, it will not be possible to make digital copies of records using USB-C digital headset outputs."

Yet another form of irritating DRM which won't actually prevent piracy.

Take apart any device which supports the copy protection, connect to output of DAC, bypassed.

/facepalm

And just like Apple TV[1], enjoy getting HDCP errors requiring restarting playback several times.

1: Might have been the Toshiba screen it was connected to.

Dunno why you're being downvoted, that's exactly the type of frustrating experience that exemplifies HDCP.

It's not really about stopping piracy though. it's about making sure they can sell licenses to manufacturers.

Yeah. If they stuck to old-fashioned 3.5mm jack, they couldn't charge a licensing fee on every headphone manufactured.

Who makes copies using their headphones jack anyhow?

I doubt anyone is spending serious money on audio DRM. The ship sailed, the standard for audio is non-DRM. Sure, there's streaming, but the vast majority of the music you can stream you can buy DRM free from pretty much anywhere. I don't think you can put that cat back in the bag.

It would be more likely to be used with Hollywood movies, where the world is still flat and of course users should be glad that they have been granted the privilege of audio at all.

Hey, we lived just fine 50 years without sound, we can live again without it if there is not a DRM solution with it - Random Hollywood Exec

Judging by all of the comments complaining about DRM and having to buy new audio equipment, it seems like few people are actually reading the whole article. USB-C can support analog audio output via audio adapter accessory mode[1].

In the near future you'll be able to buy passive USB-C→3.5mm cables for cheaper than normal USB-C→USB-C cables. You will be able to use them like any other aux cable.

Personally, I still prefer the 3.5mm jack so that I don't need a passive adapter for using my earbuds. Requiring a passive adapter that gives you a perpendicular 3.5mm jack, no matter how small, would be ugly and obtrusive.

Fortunately, at least for over-the-ear headphones with a 3.5mm jack, this will at least be much less of a hassle. Passive USB-C→3.5mm cables can just be your new aux cable.

[1] https://i.imgur.com/y6xCS9u.png

While it's true the USB-C spec provides for this possibility as part of the extensions to the USB Audio standard, Intel is also discussing an MPU specifically to implement HDCP for audio outputs from the USB-C connector in this announcement. You say we didn't read it, but we read it loud and clear. This is an enabler for DRM that does nothing to help consumers, many of whom have already significantly invested time and money into audio gear that works perfectly fine.

You can't DRM standard analog output (and still be compatible with normal analog headphones), which is why I mentioned the DRM concern.

Most of your current digital audio output ports already support DRM. S/PDIF, HDMI, DisplayPort, etc. all support DRM for digital audio.

You're making the big assumption these devices will always support analog audio output over USB-C. There's nothing preventing that from being disabled, and the digital output can definitely be forced to DRM only. We're already seeing this with things like the Lightning connectors on iPhones vs the old Dock connector that had analog and digital signals unimpeded, now with Lightning it requires an MFi certified device to decode the digital signal and no analog signal is provided at all.

The signal has to make its way to a speaker in analog form, eventually. In a way, analog output will always be supported.

But the conversion can be part of the headphone circuitry. I'm betting that's what Intel is thinking here.

In that case, it'd be simple enough to snip the wires to the transducers and take the analog signal from there. Even if there is wonky EQ for the specific transducers in use, it'd be simple enough to do a frequency sweep to figure out the filtering needed to undo it.

I respect and appreciate the hacker spirit that lets us understand that we can always work around things like DRM, but I'd prefer to fight it up front rather than deal with it later. There's MILLIONS of devices that currently work with 3.5mm stereo jacks around the world. Just because we can bypass the DRM in some way, it is unfair to the population at large to make this change when it incurs no real benefit just to satisfy the DRM desires of content producers.

In order to do that, you first have to buy a pair of the (likely expensive) DRM-enabled headphones, then destroy them and solder to the annoyingly hard-to-solder wires inside. Ultimately, this isn't going to stop piracy so much as it is people buying headphones that Intel haven't approved and charged a per-unit licensing fee on - it's about making sure they get their cut. It's probably a win-win for Intel and the headphone manufacturers too; Intel get to take a cut and the headphone manufacturers get to charge a premium to consumers, safe from competition from $3 earbuds from China (which are actually pretty good these days).

Until that is made illegal and you go to jail because you modified your device with the intention to circumvent content protections you agreed with when you bought the device.

The real long term expectation on your phone should be to use bluetooth headsets anyway. No wires is better than one less port and one more adapter OR two ports and a direct analog wire.

The key word is can, not must.

IMO the 3.5 mm jack is better because it can rotate 360 degrees. The USB connector is rigidly fixed, which puts stresses on the cord and connector. I've rarely seen a 3.5 mm fail to make proper connections or wear out the port with use, whereas both are really common with USB.

Generally I agree with you, but my macbook's headphone connector won't hold the cable in anymore.

I dunno man, I typically use stuff beyond failure, and all my headsets have failed at some point due to broken wires near/inside the connector.

They cannot be serious. Replacing what works great with all kinds of cheap or expensive and easy to repair equipment with what? Another device on the Universal Serial Bus. I see the point of attaching storage devices, sound cards and anything else that is not simple and needs a controller and logic to USB, but audio output? I do sometimes use a USB headset because it has its own soundcard built-in, but I actually prefer a simple headset with a 3.5mm mic-in/head-out connection. Why? Well, it works always, it doesn't load another driver, and it doesn't rely on the USB bus. USB is nice but occasionally hiccups and resets itself, which is not something you'll see happen with your mic and headphones because they are not overcomplicated. To be totally honest, I'm also one of those who prefers keyboards attached via PS/2 because I've had USB keyboards reset when I attached another USB device to the shared bus. That PS/2 has real interrupts is an added bonus for something as crucial as keyboard input. With all that being said, as long as this happens in the phone and tablet space, I guess I can live with it, but having to carry around an adapter from USB to 3.5mm audio will be a PITA.

This is just another, let's change it to make money with adapters and sell new implementations due to bugs in the old controllers and drivers, scheme.

Next year, USB power cords, and you'll have to rewire your house.

> but I actually prefer a simple headset with a 3.5mm mic-in/head-out connection. Why? Well, it works always

I can say with certainty that it does not always work. If I plug a pair of Apple earbuds into my Android device, the audio up/down do not work. This functionality isn't consistent even within the Android device world. Having a consistent headphone jack that provides consistency here would be a pretty big win. (Even bigger if Apple gets on board, which it won't.)

> I guess I can live with it, but having to carry around an adapter from USB to 3.5mm audio will be a PITA

Pretty sure the endgame here is that your headphones have a USB C connection on them instead of a 3.5mm audio jack.

> Next year, USB power cords, and you'll have to rewire your house.

You can already buy outlets with USB ports. They're increasingly common, but thankfully the transformer is in the outlet so you're not running low-voltage wire all over.

I think this is a false comparison for two reasons

1) adding the microphone and volume function to the headjack is a different cable than a normal TRS 3.5mm aux cable, which is the cable we all love. I would rather get rid of the 4 wired microphone enabled 3.5mm than lose the 3.52mm altogether.

2) all of the problems you mentioned that revolve around software - Apple cables not working on android, etc - are only going to get word by introducing USB into this. The idea is to remove proprietary things from analog audio, that's the whole reason the standard survived.

It also goes without saying that there is an inherit loss of utility caused by this switch, since so many of the products I already own use 3.5mm.

I'm not attached sentimentally to any cables. I'd love for audio, volume, skip, and microphone functionality to work with all my devices and all my headphones. Would it be annoying to lose compatibility (or require a converter) for all my existing speakers and headphones? Of course. I'd still take that penalty if it meant the overall experience was better and eventually more consistent.

Frankly I'm not sure 3.5mm is going to survive in the mainstream anyway. Bluetooth might eventually replace it for mainstream use cases.

>Bluetooth might eventually replace it for mainstream use cases.

As somewho who uses Bluetooth audio I strongly disagree. I think Bluetooth audio is going to remain niche until batteries get better. Who wants to have to charge their headphones or Bluetooth (often battery powered) speakers every few days?

I may be spoiled but whether I use bluetooth or old-school RF headsets, it's always less reliable and lower quality than wired.

I don't think I'm that "non-mainstream", but nothing annoys me more than even split-second connectivity drops with Bluetooth. When I'm listening to music, reliability and consistency in hearing an entire chunk of audio trumps all else -- and 3.5mm has that.

That's a problem with all things apple. Not with the 3.5 mm jack. Samsung, LG, Microsoft, Sony etc all follow the same standard for buttons. I bet if Apple implemented a USB port in iPhone it would only support accessories who have paid license fees to Apple.

No they don't. That's why most Android earbuds only have play/pause and not volume. The volume is implemented inconsistently on different phones. I did research into earbuds for my Windows phone a couple years ago and learned that manufacturers were inconsistent. You could get earbuds with volume control for the Samsung Galaxy and they wouldn't work on LG devices. The same mess applied to Windows phones.

And then when that $1 transformer (probably off by a factor of 10) fails, you have to replace your whole outlet? No thanks.

Outlets are cheap and easy to wire. Running low-voltage cable throughout your house to supply USB ports in outlet panels isn't - especially since, at least in the US, you can't run HV and LV cable in the same conduits and still pass inspection. (Nor should you! Suddenly having 120VAC@10A on a USB port's power pins is a Bad Thing.)

Yeah. I'd rather do the conversion once or twice - say, 2x high quality 5x USB ports from Anker or similar than to have it somewhere I need an electrician to modify.

> You can already buy outlets with USB ports

The few I've seen examined in YT teardowns seem to rate as "horrific". Lowest quality, minimum component count stuff. More than enough to put me off buying.

> Pretty sure the endgame here is that your headphones have a USB C connection on them instead of a 3.5mm audio jack.

Like most people, I want to keep using my audio equipment that does not use USB, so I'll need an adapter.

Industry signaling a strong desire to move from analog to digital

Which industry would that be, I wonder? The audio hardware industry, or the content creation industry?

This is ironic. It used to be the analog medium that had a sort natural copy protection built in.

>Industry signaling a strong desire..

Is this the same industry that gave us the HDMI "blessing" ?

"The headphones, audio cables and the jack adapters are too cheap.. We can solve that !"

HDMI does have nice features, like ARC and CEC.

(I'd rather use DisplayPort though)

Neither of which require DRM, I'm sure.

HDMI causes a lot of problems (and price increases) because of HDCP.

> HDMI causes a lot of problems (and price increases) because of HDCP.

And the way around it (in the specific case of HDCP) is often to buy cheap, probably non-standards compliant equipment, which isn't a good situation for anyone. I recall reading that the VitaTV had HDCP, yet I hadn't encountered that issue. Turns out the HDMI switcher I was using (the cheapest one I could find on Amazon with two outputs) was just stripping it off.

Then you got lucky. If that practice was pervasive, 1) honest people wouldn't face as many issues with HDMI; and 2) the Hollywood DRM-lobby would go crazy on manufacturers

CEC is really fantastic. For Raspberry Pi media centers connected to a TV, your TV remote just works without any additional setup.

Previously, it was a massive pain in the ass to set up key mappings with a bluetooth keyboard, fiddling around with OpenElec config files over SSH until it supported the keyboard. It also means the media centre is non-geek usable.

CEC has given me nothing but pain. Namely inputs switching randomly again and again. I had to disable it to make my TV usable.

I imagine as a mobile device manufacturer you would be pretty excited to get to cut the total number of ports on your device in half and sell high margin cables/adapters/hubs as accessories.

Let the dongle wars begin...

So now instead of buying a single DAC in my phone, I have to buy a DAC for every set of speakers and headphones I ever want to plug in to it?

Coming from the reverse perspective: now my expensive wireless bluetooth headphones will make use of their own good DAC all the time, instead of relying on my phone's crappy DAC whenever I plug them in directly.

> expensive wireless bluetooth headphones will make use of their own good DAC all the time.

Thank god. Otherwise you'd miss out on the subtle nuance and coloration from the compression artifacts that you can't enjoy with 19th century headphone technology.

You can already use your own DAC with smartphones.

How does that work?

You plug it into the USB port.

On Android phones that support USB-OTG (most of the better/more modern ones) you can literally just plug a standard USB DAC into the onboard USB port. Honestly, it shouldn't be that hard to just include a decent onboard DAC and headphone amp these days though - Sandisk's MP3 players and Allwinner's range of ARM SoCs have even managed to integrate both on the same die as the main CPU no problem.

Apple devices have generally excellent DACs as well - to the point where it's basically a waste of money to buy an external DAC. I wouldn't be surprised to learn that there are Android devices with good onboard DACs, but I also wouldn't be surprised to learn that such devices are at the high end of their manufacturers' ranges and come at a premium price.

> A good thing about USB Type-C headsets with MPUs is that they are going to be software upgradeable and could gain functionality over their lifespan.

yeah, unless it never happens, like it typically doesn't :(

Maybe we can rely on the community for patches for headphone 0 days?

> > software upgradeable and could gain functionality over their lifespan.

That usually means "now we get to ship it broken".

I just can't wait to have bugs in my HEADPHONES of all things.

Or for them to be connected to the IoT.

Why the hell does a digital serial bus interface have an analog audio "mode"?

I mean, USB-C is already supposed to do daisy chaining, some 100w power bidirectional power transfer, and support video channels over it according to displayport spec. Oh, and its also thunderbolt.

Seriously, is this port supposed to cost more than its weight in gold to manufacture, and be such an extreme nightmare to program for we should expect exploits every other Tuesday? I am totally on board with a high bandwidth even parallel standard port for digital data exchange with good power delivery metrics, but all this specificity over how it performs with what data while supporting analog modes is... feature creep, by definition, in my book.

You just hook up your existing DAC to the USB-C port instead of a 3.5mm jack, and the rest is just software. It's not like they need to use special circuitry that isn't already in your phone. All phones in existence already have a DAC (or you couldn't make phone calls using the built-in speaker).

This reminds me of the Game Boy Advance SP and how it removed the 3.5mm jack for seemingly no reason other than to have me buy an adapter.

I won't buy a phone without a 3.5mm jack. It just works. I don't need digital audio. My (medium-price-range) headphones sound utterly glorious, there is no utility here.

To be honest, I'm not sure whether I'll ever upgrade my 2013 Moto G. It'll probably break eventually.

When did I become a luddite? It's like, at some point, things stopped getting substantially better, and just became sidegrades with annoying tweaks for the sake of it.

I want my toaster to take... bread. Not tomatoes. Bread is what I eat for breakfast, not toasted tomatoes. :P

Initial steps toward DRM on the audio path?

From the article:

> The MPUs will also support HDCP technology, hence, it will not be possible to make digital copies of records using USB-C digital headset outputs.

DRM on the audio path has been attempted since the earliest days of PC soundcards. It took the industry a little too long to work out that anyone could route a CD signal to the Record Input inside most typical PC soundcard mixer apps. Once they did work it out, that option disappeared.

DRM killed DAT as a consumer format back in the early 90s.

In fact there were DRM attempts on vinyl and tape, but it's impossible to make inaudible analog DRM, so they never caught on.

The bigger problem now is the move from files to streaming services. If you sell someone a file - downloadable or on CD, it doesn't matter - they can always torrent it.

If all music is locked inside streaming services, they can't - at least, not easily.

It's bizarre. Why would we push the <5mm^2 that an audio amplifier requires in a smartphone into the headphones, where it then inevitably requires local decoupling, a circuit board, power management, something to decode the USB (or can USB-C do analog?) and all the other hassle?

Makes absolutely zero sense.

The amplifier in the headphones can be designed specifically for those headphones. It may well be that a transducer with a bizarre frequency response and an amp/EQ that compensates is a better solution in the end. Where we are now, headphones need to have reasonably flat response (or be called Beats :-) in order to be acceptable. Or perhaps a wildly different impedance makes a better cost performance argument, can't go there now, but with active headphones you can.

It may be a bit like the powered speaker market for PA gear. You can get light weight powered speakers for not a lot of money that perform quite well. The amplifiers do exactly what they need to for the exact speaker, they don't have to be able to handle whatever mystery load you plug into them. They can build in the crossovers and EQ compensation when the signal is small rather than in the speaker cabinets.

Also, noise canceling ear buds become possible when you have power coming to them.

Of course, the heap of Apple Earbuds I have whose clickers no longer work doesn't really inspire me about my new active headphone future.

I’m betting on next year’s Retina Macbook to nix the audio jack for a second USB-C. Likely held off because they don’t want to detract from a big wireless earbud launch with the iPhone 7 this fall…

Maybe 2016 will be Apple's year of USB-C. A single USB-C on the iPhone 7, two on the MacBook, and USB-C earphones?

It could happen.

They just released the refreshed MacBook, they'll probably wait for 2017 OR they'll do that on the new Pros that will be announced at WWDC.

Oh yeah, I know about the 2016 MacBook, but I'm guessing they could lay the groundwork for the 2017 MacBook with their other products this year.

"Industry Signaling a strong desire to move from analog to digital"... which "industry"?

"New digital audio needs to offer significant value at higher end" ... Intel market segmentation at its finest.

Long rein analog.

Looks like they missed a few groups in the job cuts.

Apparently Apple and a Chinese company is the whole industry.

Welp.

Here is to hoping for AMD's zen.

USB audio is nice but it's also nice to not have to buy a new headset or new DAC for my new box.

Edit: spec says analog audio over usb so, there's that.

USB-C has an analog audio output mode (audio adapter accessory mode which uses the device's internal DAC), so you won't have to do that.

There will eventually be passive USB-C→3.5mm cables for use with analog-only headphones.

Do you have a link for this? I've never heard of it but it sounds cool.

Here's a screenshot of part of the relevant section from the USB-C spec: https://i.imgur.com/y6xCS9u.png

> The headset shall not use a USB Type-C plug to replace the 3,5mm plug.

So, the only sane way to connect an analog headset to a mobile phone is forbidden by the spec... Assuming I had a phone with a Analog-Audio-USB-Type-C capable, I'd like to have a headset that directly plugs in, and for which I don't have to purchase, carry, and loose a separate adapter.

So attach the adapter and then pretend it's part of the wire for the lifetime of the device.

Pardon me, but I don't get the relation between AMD's Zen and USB/non-USB audio...

"Industry signalling a strong desire to move from analogue to digital"

Industry can sod off. This is about DRM not any consumer benefit. The decent quality is all in the hifi market that's quite happy with jack plugs.

Why would I want a digital headphone jack? Headphones are analog devices, so you're just going to have to cram a DAC into the headphones anyway, making them heavier, more expensive, and DRM-encumbered.

I think this is a sign that technology is stalling. They are grasping at straws trying to get us excited about technology that adds little value for the consumer but drives another upgrade cycle and even removes features that they can sell back to us. First we get the locked down no dd-wrt routers and now this.

Seeing how gaming keyboards and mice have gone these days. I wonder if we'll need manufacturer specific software and an account and to access our future audio settings and features.

USB seems a lot more flimsy than 3.5mm.

IIRC, USB-C is supposed to be the most resilient USB to date (in terms of average plug/unplug cycles before failure). Too early to say if that's actually true, or how that compares to 3.5mm in any case, but it's not impossible that it's comparable--especially since USB-C finally eliminates the "try to jam it in the wrong way 'round" problem.

> the most resilient USB

At the cost of the devices it's supposed to connect together[0]. When you have to think about having to add cryptographic signatures and verification to cables[1], I cannot see the connectivity standard you built as safe.

0: https://www.amazon.com/review/R2XDBFUD9CTN2R/ref=cm_cr_rdp_p...

1: http://www.bit-tech.net/news/hardware/2016/04/13/usb-type-c-...

yet another form of DRM, for which consumers will pay and corporate will make money, and yet it fails to solve the piracy problem. high fidelity analogue Audio recorders are easily available.

It's a shame that apple to be going with a similar but incompatible strategy: http://www.cultofmac.com/401014/apple-now-sells-lightning-he...

Remember when 2.5mm was a thing for phones that could play MP3s?

That sucked.

I never knew what that port was for, thanks for clarifying, really, I'm not kidding.

Some were just for one earphone (mono) and a mic. Some were effectively the same as the iPhone (stereo+mic). Some had weird adapters for other ports if they just had the earphone+mic version. I think I had one of each at different times, all pre-2007.

I wish the industry had moved to 2.5mm back then. 3.5mm is great, it has survived the test of time (introduced in 1964!) because it is so good, but 2.5mm is only better than 3.5mm (except for the manufacturing and migration costs).

If the industry had moved to 2.5mm back then, Apple and Intel and the DRM-lobby would have had one less excuse for phasing out analog audio ports today.

We already have strong and widely adopted digital audio standard, its called I2S. HDMI audio is four I2S channels in parallel. Provision pins for raw I2S, or simple usb endpoint decoding into four I2S, but not another effing DRM shit.

I thought Bluetooth was supposed to eliminate the need for the 3.5mm jack...

if phones don't get two type-C connectors there is no point to this..

The future will be one port on everything breakout dongles everywhere.

Well it is gonna suck not being able to charge your phone while listening to music

Instead of metal weights[1], headphones will contain batteries so first you'll charge your headphones and then your phone will charge off the headphones.

[1] https://blog.bolt.io/how-it-s-made-series-beats-by-dre-154aa...

/s ?

Personally I think this is a terrible idea (I already have too many things that need charging and charging a battery from another battery is inefficient) but I predict that it will happen.

(BTW, does anyone know what happens when you plug two battery-powered USB-PD devices together? How do they decide which direction power should flow?)

Not to worry, by the time USB-C headphones are ubiquitous all that battery tech we keep hearing about will have made it to phones. :)

No, the future will be at least 2 USB C ports on every device.

I think USB Type C on top and bottom of phone would solve the eternal debate between those who want audio port on top or bottom. And provide more ports which will be important for all the uses of USB Type C. Wouldn't need the splitter dongles for providing power like with OTG or HDMI adapters.

What is the purpose of making the audio cables digital? We don't need more audio resolution, we are already beyond the limits of human hearing. It's like making phones with 2000 ppi resolution for no practical benefit other than bragging rights.

It'd rather prefer we had better wireless audio. Bluetooth is too weak for streaming around the house and has slow connection time, is unstable and generally discourages wireless audio.

I'm torn on this one. I know it's a move to push DRM further out the stack, and that annoys. But, I also want higher quality recording and playback from my small devices. The audio circuitry of one of my tablets and my phone is abysmal; my Nexus 7 (second generation) is nice but I don't have a good recording option. Presumably this audio will be two ways, so if I want to stick an ADC in that port I'll be able to record at very high quality, and if I want to stick a DAC in that port, I'll be able to play back at very high quality.

DRM is stupid, of course, and it's just pushing the copying out one more step in the chain (they can't stop you from converting it to analog at some point, because it's gotta be analog to get into your ear holes). And, of course, DRM is made to be broken.

Anyway, the 3.5mm jacks on my devices are about 50/50 unusably bad (either they aren't grounded/filtered properly and end up with a variety of noise, or they aren't loud enough, or they distort at modest volume, etc.), so on the whole, I won't mourn the passing of the 3.5mm jack.

You can already do digital audio on Android, either over OTG USB and normal USB Audio, or via the Android Accessory protocol. Presumably this would work just as well over a USB C connector. Adding DRM is a pure downgrade from this.

I posted this on a thread about this that should be linked somehow:

A true advance, a Jobs and Ives worthy advance, would be to superseed it. Use the 3 prongs of a jack with mic for ground tx,rx,and keep it backward compatible with analog only devices. reply

not a shred of reality displayed here. the analog audio jack they seek to replace is the final analog output of a digital to analog converter or DAC. one cannot speak of a digital speaker, and a digital amplifier does not usually refer to using the loading of the speaker driver to smooth-out the pulse-chain like some motors on a PWM line... not in any high fidelity audio anyway... where is the DAC or audio codec to be located? are we not just pushing it out to the device and pretending it went away? why the assumption for headphones or consumer audio? what about USBs derived clock and jitter? why on

What do they mean by going "digital"? At the end the signal that reaches the speaker driver will have to be analog. What are they suggesting to happen instead?

How about the mechanical loads headphone plugs are subjected while the phone is in a pocket and we are walking?

I am very sure a USB-C is nowhere near as robust as a mini P2.

No.

You can put _analog_ audio over USB-C, too. From the article:

> In fact, USB-C can be used to transfer analog audio in accordance with the specification of the connector. It all comes down as to how that audio is transmitted.

Somehow, I prefer "plain" USB for headsets: it's one connector instead of two, and feels (looks) more modern (plus, requires less force to (un)plug).

No sympathy for clunky DRM-"enhanced" hw, of course :)

Reply on news.ycombinator.com