Hidden standards make USB-C a pain to deal with | Hacker News Reader

Hidden standards make USB-C a pain to deal with

onezero.medium.com

34 pointsnotmine133744 comments

I completely disagree. I can put a Type C plug into the port of any of my devices on the first attempt, the quality of the plug is much higher, the plug is more stable in any of the ports of my devices, I can charge all my devices with one charger, I have a docking station that just delivers 4k/UHD@60hz with both of my laptops (MBP, Pixelbook) and all I need is one Type C plug in my laptop.

I have a GaN 68W charger with 2x C and 1x A and it's the best travel companion I've ever had. It's lightweight, small and it just works.

I roll my eyes everytime people complain about Type C. The second Apple delivered Thunderbolt 3 via Type C it was clear that the dream of having one universal port for everything doesn't work, then came various ways of quick charge, some even proprietary. But you can rest assured that anything that's just Type C will work with any Type C and charging isn't always quick but it works.

This is so much better than anything we had before.

Charging is only one of the things USB-C is capable of. But most people aren’t complaining about that;[a] they’re complaining about the “alternate modes” such as DisplayPort, Thunderbolt, etc. And if one wants to use those “alternate modes,” you can’t use any old cable; you need specific ones capable of them. It should be obvious why this is a problem: just because two cables look identical, they may not be.

And not even that! There’s also the problem of bad cables. If the USB Consortium actually cared about consumer safety, they would aggressively enforce their trademark against cables falsely advertising what they’re capable of. But they don’t.

Remember Benson Leung? The guy who was testing USB-C cables from Amazon so you wouldn’t have to? And then one cable ended up being so bad, it fried his Pixel Chromebook.[0] Not good.

TL;DR: USB-C was supposed to be “one connector to rule them all,” and it succeeded at the expense of consumer confusion and possible harm.

[0]: https://www.amazon.com/review/R2XDBFUD9CTN2R/

[a]: Except for maybe the people who don’t understand the basics of USB-PD

DisplayPort/Thunderbolt requirements only affect a few people and occur in contexts where "the correct cable" was already a requirement prior to USB-C. The problem with USB-C cables from Amazon is they are from Amazon...same as for regular USB wall warts, SD cards, and many many other categories where Amazon willingly sells whatever crap it can take a cut of.

>This is so much better than anything we had before.

I haven't really noticed much difference between having a USB-c charger for my latest phone vs an older USB on my previous phone other than a lack of extra charging cords available in my house.

Plugging the cord in upside down, inexplicably changes my phone to slow charging. The plug, after a couple years now, has worn down exactly the same as other USB sockets has. The cord's loose when plugged in, needs to be wiggled to make it charge.

As far as I can tell, besides being able to plug it in both ways, it might as well be no different.

At least in terms of charging my USB-C devices the experience has been good for me. I know I can carry a single charger (ie: my Dell 100W USB-C charger) when travelling and it will charge my cellphone and Nintendo Switch as well.

Need to be careful about using the port on the switch. It isn't a usb C port even though it uses the connector. It has been known to brick or damage the device when used with certain combinations.

That really annoys me (companies using connectors without conforming to the standard). I have no idea what the state of the legislation is but I find that EU proposal to make everyone standardise around proper USB-C tempting.

These are just teething pains for an emerging, universal charging standard. I will take some incompatibility between USB-C even for the next decade, if that means that I no longer need to carry more than one wire and more than one charger to power all my devices.

Just dongles for any inexpensive / older hardware.

Older hardware will naturally get cycled out of use by exponential-like decay; it won't ever go to zero because of legacy requirements and retro device enthusiasts, but it will be basically zero for most people given enough time. Inexpensive hardware will just stay with the cheap subset of USB-C without fast charging, which is perfectly fine.

When is this supposed to happen? Lots of USB-A and micro USB devices are still sold today.

Replace: USB-C with RJ-45 and his argument is just as silly.

It turns out both simplified our lives dramatically, and both require you to do a bit of research before plugging things in.

If you don't want to brick your expensive devices, buy approved chargers. No different than being able to buy cheap knock-off batteries for devices off ebay. Will they work most of the time? Sure. Are you risking bricking your device or burning down your house using that $25 laptop battery off ebay? Absolutely.

I hear you, but RJ45 never pretended to be anything other than a physical standard. USB-C, on the other hand, is the latest successor to the USB standard, which from the beginning was both a physical specification and a protocol specification. The article correctly points out that there are now so many variations on the protocol implementation that it's confusing to the consumer. And he's clearly right. The whole point is that you shouldn't have to "do a bit of research" before you plug in the only cable connector your device has.

Citation?

The USB-C specification is the physical specification. All the other stuff is a USB specification but it has nothing to do with USB-C.

https://www.usb.org/sites/default/files/usb_type-c_language_...

https://webstore.iec.ch/publication/25676

The layman confusion the USB-PD and USB-3.2 specifications is no different than someone thinking they can get 10gbe over a 1000m cat-5 cable because the ends plug into their 10Gbe adapter.

You're absolutely right, I stand corrected.

But the author's complaint is still legitimate: in the context of billions of consumer devices, the diversity of behaviors is a problem. With RJ45, as long as the cable wasn't defective, it generally wouldn't fail completely. And relatively few consumers ever wired their own networks, whereas virtually everyone has to deal with USB-C-equipped devices.

>With RJ45, as long as the cable wasn't defective, it generally wouldn't fail completely.

That's just not true, and I don't mean this to be demeaning but you clearly haven't had to install new hardware into an old datacenter. I can't tell you how many times I've run into situations where a 1Gbe port would auto-negotiate to 10Mbps half-duplex. Did it technically not fail? I guess... but the line was so lossy you couldn't get a ping to complete much less an actual session of any sort.

I can tell you that, as a consumer who has wired a lot of small office and home Ethernet, I've never seen a failure like that. On the other hand, I've seen lots of unexpected behavior with USB. I don't doubt your sincerity, but the reality is that very few consumers have to rebuild data centers. Lots of them need to power personal devices with shared USB cables and power supplies.

OK, but for example of a problem I recently ran into, I left the (excellent) magnetic surface charger at my office and needed to charge my surface pro 7. Not a problem, I figured my phone charger (pixel) should be fine, if a bit slower. Nope, no dice. I had to Google around to remind myself that "power delivery" is what I need. OK, exceedingly stupid name, but it jogged my memory that the portable battery I have supports that, and I have used it for that before. Cue more frustration when I realize that the only 2 cables I could find at the house don't support this charging standard either.

At this point I feel like an absolute idiot, since now I've wasted 1/2 hour plus 2-4 Google searches just to charge a laptop. None of the equipment is a sketchy aliexpress/Amazon no name part, except possibly 1 of the cables. If it's this hard for someone with electronics experience who is into tech, I can't even imagine explaining it to a family member in a way that would make sense.

Some devices have greater power demands than others. So there are two design choices that can be made: (1) require all chargers and cables to support charging all devices; or (2) allow "lesser" chargers and cables.

Choosing (1) would have meant that everything would work as you expect, but then even the cheapest devices would have to come with an unnecessarily beefy (and therefore expensive) chargers.

Choosing (2) results in potential for the situation that you described. Lesser chargers and cables cannot charge more demanding devices.

I'm pretty sure (1) would simply have not been practical. It would have been wasteful and the expense would kill adoption. So they went with (2).

Seems reasonable to me.

The key missing choice is (3) Make it possible to tell the difference between the cables/devices.

Thats what everyone is complaining about, that there is no way to know without trial and error what combination of capabilities across (at a minimum) three different elements you're dealing with.

Make it so that we can tell the difference! As per teh articel, they tried to do this with the thunderbolt standard on top of USB-C with a lighting bolt symbol, but even that doesn't mean you know what you're getting in practice.

USB3 for instance mandated blue plastic for connectors, so you could tell the difference from previous generations.

A better solution is to just have a different physical connector so it's obvious at a glance if you have the correct cables. But that: involves carrying another cable, and kind of defeats the purpose of a unifying connector.

I've bought stuff that supports everything I own, and so I don't have to carry multiple cables or chargers. Different physical connectors would be a step backwards for me.

Why would you need to buy "approved" chargers? If the vendor gets to dictate which chargers I buy, I'm worse off than before the days of USB charging.

Also, a lot of official chargers are cheap, bulky and slow. Third party does not and has never implied "cheap knockoff". In this case, they aren't knock-offs at all.

You don't need an approved charger, just one that's well-made. It's easier to use a well-known brand name as a shorthand during the selection process, because it's generally not feasible for the end user to evaluate a charger on a technical level.

You don't, I'm not aware of any vendor that FORCES you to buy their charger, and it's not part of the spec as far as I know.

Op is complaining about a charger bricking something - the solution is: buy the approved charger from the vendor. They've tested it, they're telling you it won't brick your device, and I'm confident if it does, they'll replace the device. If you DON'T WANT TO, buy third party. But then don't whine or blame USB-C because you DID buy a cheap charger that bricked your device. That has literally nothing to do with USB-C and everything to do with being penny wise and pound foolish.

He's blaming USB-C for something that has nothing to do with USB-C.

Aside from that, some manufacturers, for example a certain one named after a popular fruit, decided it would be prudent to save a few bucks and ship cables that are only able to charge devices, not for any form of data transfer.

Ok, maybe it wasn't to save money and instead force people to use cloud solutions for data sharing. This way we will end up with a lot of disabled cables where people wonder why they don't work.

They are ordinary USB 2 data cables per the spec. In fact there is no way to negotiate power delivery without having that ability.

I might wish that Apple provided at least USB 3 speeds in those cables (in fact I do wish it!) but their cables are completely compliant with the USB spec.

What are you talking about?

The USB-C cables which come with MacBook chargers only work for power delivery.

Really? God, even cheaper android phones I have bought recently have charger cables that support data. Isn't that kinda required for USBC charging, since the charger and device have to negotiate power delivery w/ each other?

I don't own any Apple devices; the only time I've encountered a USBC cable that wouldn't support data was a cheap charger cable from a gas station. I took wire cutters to it when I got back from my trip, I don't want that potential fire hazard around.

Edit: I'm sure the apple cables are safe, they do test them. However, a no-name cable that won't support data transfer? For all I know, it may possibly report itself as high power capable, and melt when I try to charge my laptop.

It is true sadly, although I don't know if it is iphones or macbooks or both. Only affects the newer models that use usb instead of lightning. Sadly we have to support Apple but we now destroy the cables that come with new devices. What a waste...

It is very much not true, and you seem to be confused by the various cables and connectors that are available. The only device which uses USB-C and previously used Lightning is the iPad Pro.

Maybe you are correct and it is iPads then and not iPhone or macbooks. But all the latest usb 3 cables from Apple are charge only.

This is simply untrue, and I have just verified this by plugging the cable that was plugged into my charger into a USB-C device and verifying it appears on the bus.

The funny thing is that would actually be a good thing from a security standpoint. If you don't want data from a dodgy hole instead of messing with "USB condoms" using the cable. It would also be consistent with Apple's "it just works/idiot proof" design philosophy.

Chiming in to say I have done the same thing and observed the same result. Seems to work perfectly fine. Mine is a 2017 model MBP.

I believe it works at USB2 speeds only. It's called the "Apple Charge Cable". I can't verify right now as Apple's online store is down.

That’s the default mode of most PD cables, as it is the required standard. USB 3 (err, 3.1 in USB-C-land) is an optional mode in the USB spec.

Man, so true. I’m still mad at LumeCubes for being USB-C, but only being chargeable from a USB-A to USB-C cable because making C to C charging would’ve required additional circuitry.

It's one additional resistor, I think. The first lot of raspberry pi 4s had that same problem, and I've seen it in other hardware that uses type-c to charge.

Interesting. I bought some cheap wireless earbuds and I can’t charge the case with a C to C cable, only with an A to C cable. Now I know why.

usb-c is too fragile. blew out my macbook pro last week when i plugged in the charger. High power through such fine connectors is just asking for a spark / short. So sad. Poor design.

My OnePlus 5t USB-C port wore out within a year, this never happened to other ports.

Mine didn't wear out quite that fast, but it is indeed finicky, no matter how much I clean it out. Same with the Nexus 6P it replaced.

My new PinePhone seems to be more solid all around, so hopefully the USB-C connector on it ain't an exception. It ain't exactly my daily driver right now, though.

Reply on news.ycombinator.com