Hacker Read top | best | new | newcomments | leaders | about | bookmarklet login
Whoever invented USB-C deserves a Nobel Prize (thenextweb.com) similar stories update story
1 points by Brajeshwar | karma 59882 | avg karma 5.56 2021-12-30 08:38:30 | hide | past | favorite | 17 comments



view as:

More like an Ig Nobel.

Someone call the waaaaa-mbulance! Patient down.

Anything in specific got your goat?


I see many devices where the USB-C connector doesn’t seem securely connected to the circuit board and where there are electrical problems that cause devices to detach.

Then there is the ‘twisty little connectors that all look alike’ problem.


hardly ignoble-worthy. Sorry to say.

Personally I have a couple of issues with USB-C (and the whole USB Ecosystem as a whole);

#1 - the female connector of the port has a circuit board present which may be more readily damaged than a port without it (ala Lightning - which has it's own problems).

#2 - Because the USB Standards Committee did not put a mandatory minimum capability on cables using the connector it is impossible to tell at a glance what the capabilities of a cable are, there are no identifying markers - No detail about Power Delivery, data transfer speeds, or which generation of USB it belongs (a Type C - USB 2.0, 3.0, 3.1, 3.2, 4 cable, all look the same at a glance)

#3 - the only way to buy the most capable USB-C cable is to give up altogether with looking for a USB-C Cable and to just buy a Thunderbolt 3 or 4 Cable, which isn't even USB!

#4 - the USB Standards committee insistence on renaming existing versions of the protocol with the newer standards leads to untold confusion among consumers - Any non-technical person will look at at USB 3.2 Gen 1, USB 3.1 Gen 1, USB 3.0 and believe there is a difference, when they are the same.

#5 - The USB Standards Committee insists on naming it's protocol revisions in the least user friendly way possible - I have never once seen a device say it requires a host device with a "SuperSpeed USB 10Gbps" or "SuperSpeed USB 5Gbps" port, coupled with how badly named transfer modes are USB 3.2 Gen 2×2, USB 3.2 Gen 2×1, USB 3.2 Gen 2, USB 3.2 Gen 1 means it is far too easy to trip up on.


> #3 - the only way to buy the most capable USB-C cable is to give up altogether with looking for a USB-C Cable and to just buy a Thunderbolt 3 or 4 Cable, which isn't even USB!

The most capable USB-C cable isn't just a TB3 cable.

TB3 cables doesn't guarantee USB 3.2 Gen 2x2 support. Nor does it imply USB PD support. Also, does not guarantee Displayport Alt Mode support.


> Nor does it imply USB PD support.

All USB-C cables do PD. They all do 20V 3A. ALL OF THEM. This is wrong. What FUD.

> TB3 cables doesn't guarantee USB 3.2 Gen 2x2 support. ... Also, does not guarantee Displayport Alt Mode support.

There's no mode other than 40Gbps that is official, from what research I've done. But I do see some cables that list as 20Gbps. To me, this seems like a spec violation, but I could be wrong.

In practice, nearly all TB3 cables are 40Gbps capable. They will almost completely certainly run USB 3.2 Gen 2x2 more than fine and will run DisplayPort Alt Mode more than fine. They'll do great.


> All USB-C cables do PD. They all do 20V 3A. ALL OF THEM. This is wrong. What FUD.

No you are the one who is wrong. Nothwithstanding optical TB3 cables which can't even carry power, there are a large number of USB-C cables that only do 5V 3A on Amazon such as this one:

https://www.amazon.com/AmazonBasics-USB-C-Cable-USB-IF-Certi...

> There's no mode other than 40Gbps that is official, from what research I've done. But I do see some cables that list as 20Gbps. To me, this seems like a spec violation, but I could be wrong.

https://en.wikipedia.org/wiki/Thunderbolt_(interface)#Thunde...

Intel shipped at least 2 controllers that uses PCIe 3.0 x2, which only allows 20 Gbps.

Here is a CERTIFIED TB3 cable only capable of 20 Gbps:

https://www.amazon.com/Certified-Cable-Matters-Thunderbolt-U...

> In practice, nearly all TB3 cables are 40Gbps capable. They will almost completely certainly run USB 3.2 Gen 2x2 more than fine and will run DisplayPort Alt Mode more than fine. They'll do great.

No they won't. Some don't even support USB 3.0 (5 Gbps), especially if they are longer than 0.5m.

https://tidbits.com/2017/08/23/how-some-thunderbolt-3-cables...

"Instead of carrying up to 5 Gbps of USB 3.1 Gen 1 data, these active Thunderbolt cables throttle down to USB 2.0 speeds, offering about one-tenth as much throughput."


> No they won't. Some don't even support USB 3.0 (5 Gbps), especially if they are longer than 0.5m.

You got me on the long cables. The active TB3 cables are weirder than I thought. I'm not going to hold my breath USB4 fixed this, alas. Other than active long cables, I expect success (although tbh there's numerous beyond-spec (but likely adequate) ~0.8m cables too). As for 20Gbps, that sucks; I thought half-lanes was for ports only, not cables, and had rarely seen (& promptly ignored) 20Gbps cables when shopping. Thankfully these half-breed cables seem exceedingly rare, but yes, a pitfall. I still think your overall premise, "The most capable USB-C cable isn't just a TB3 cable," distracts from just how good an option TB3 cables are for most folks. >95% of TB3 cables on Amazon (old-ish & affordable now) are the most capable USB-C cable, but you've honed in on picky little nits & exceptions & made mountains out of mole hills. I protest; this is immoral.

> No you are the one who is wrong. .. there are a large number of USB-C cables that only do 5V 3A on Amazon such as this one:

I don't get what Amazon's up to, but you'd have to try real hard to make a cable that can do 5V 3A but can't handle 20V 3A. Cables are just not that sensitive to such low voltages, and connectors are too standardized to notice. And indeed, one of the (if not the top) most trusted, thorough, technical cable reviewers on the planet claims this cable is quite good, low resistance, and showed it doing 20V 3A just fine. Perhaps Amazon just lacked test equipment to confirm 20V, but whatever the case, to fail 20V would be difficult to pull off. https://www.amazon.com/gp/customer-reviews/R2DXKTH7UAJIZ4/re...

Re-asserting: no, you do not need special cables for USB-PD. If a USB-C cable can't do 20V 3A, it's broken & out of spec & not USB. You're wrong & USB-C & the Type-C connector in general does not deserve this baseless & incorrect slander.


> Re-asserting: no, you do not need special cables for USB-PD. If a USB-C cable can't do 20V 3A, it's broken & out of spec & not USB. You're wrong & USB-C & the Type-C connector in general does not deserve this baseless & incorrect slander.

USB-C spec doesn't require cables to support PD. You're wrong.

https://www.reclaimerlabs.com/blog/2017/1/12/usb-c-for-engin...

"Even without supporting USB Power Delivery, it is possible to get as much as 15W through the USB connector. VBUS is still limited to 5 Volts, but the current can be as high as 3 Amps."


The quote you're citing is chiefly discussing what happens if you use a charger or device doesn't have USB-PD support. Perhaps there are some cables somewhere if you search long hard. Perhaps. But again you're dealing in extremely rare & exceptional cases & sowing doubt, where I really don't think any such misgivings are deserved. I dislike seeing such unfair maligning & sowing of doubt. I do not think this sense of fear & afraidness is reasonable. A USB-C USB3 cable without CC lines is out of spec & cannot be called such.

1. I've had usb-micro ports die on lots of phones /devices. Never had a usb-c port fail, anywhere. I can accept maybe it's still not good enough. Personally though the durability has been exceptional. I've been so so happy at USB-C as a mechanical interface. Overwhelmed with joy. I'm surprised there are complaints. You and another poster both made it your top complaint, and I'm shocked; I had no idea there were complaints from anyone, anywhere. For it to be a top complaint is astounding to me.

2. It definitely sucks getting usb2 cables. It would be good if cable were better marked but a usb3 cable will give 3A, 20V, 5Gbps. If you want more, you gotta go higher. Look for a 5A (also 20V) capable cable for more power, or look for something 20Gbps capable. Don't bother looking inbetween. This goes to point 3:

3. I just look for USB4, which standardizes things better.

4. Agreed the various USB3.x specs are frustrating. Not a USB-C problem.

5. This feels like #4b. Agreed that the USB spec proliferation has mostly confused not informed. But I'm not super sympathetic to these complaints, because I don't think it affects 99.99999% of uses. Most of us still use USB3.2 Gen 1 devices & ports: 5Gbps. There's a ratcheting up of speeds that is now well defined: some devices/hosts/cables support 10Gbps signalling (and often cables not rated for 10Gbps still run it fine in my experience): USD 3.2 Gen 2 or Gen 2x1; 10Gbps. Either your host or your device might not support this signal speed, ok, sorry, no go. If you need more there's USB Gen 2x2, 10Gbps x 2. There was a time when 5Gbps x 2 was a thing, but 1x2 is kind of obsolete, kind of crap. Most cables do 10Gbps signalling, whether they advertise it or not: there's enough slot, enough to work. Needing a whole extra set of lanes is a much bigger ask. It's easy to get overwhelmed, but there's a middle-ground that's simply not worth evaluating, that doesn't actually really exist. USB 3.2 defined away & deprecated a lot of that questionable middle ground. Thankfully.

I don't see a lot of these issues as USB-C's fault. They're things USB in general did. What's missing from these mew'ings is recognition that USB-C also enabled it's killer feature: multiple channels/lanes. That we can plug in a USB-C connector & get USB3 AND DisplayPort, or USB3 AND HDMI is absolutely killer, just amazing leaps and bounds. It's such a huge win for connectivity. Giving us 3A/20V is another epic win. USB-C is just tops.

USB4 pivots us into a future where we no longer need dedicated lanes for different protocols. It is an epic leap forward. But I am so happy USB3 lead us here, opened the doors, became the one connector that could do it all. That it also happened to have some short-lived semi-confusing mixed modes... it's not 100% clear, but fuck it, it was an obvious & available & great engineering win at the time. Short-lived but it was part of the race for better & it cost us nearly nothing to add, was using the cables & transceivers we already had, so I get it. I am 100% sympathetic. This grumbling & grousing... it's pedantic & boring & irrelevant. Everything still worked, even if you weren't getting crazy speed. Most people won't notice nor care. Those who do care can figure it out in 15 minutes (after making the mistake once). USB-C rocks (particularly now that USB4 raises the bar, re-sets expectations, & does away with the USB3.2 multi-tiering).


My beef:

Before you could looks at a connector and if it fit it worked. Now? Who the he'll knows. It's getting better slowly but it's a mess.


There was a period before USB ('universal serial bus' pretty much sums up the USP of USB), and I'd say the general adoption of USB to power and connect random things was overall more impactful than USB C specfically. A million tiny wins, e.g. buying christmas lights that don't come with their own powerbrick, but plug into USB-A. But yes, generally heading in the right direction and good to recognize the benefits of open standards.

At least with USB-C, I no longer have to try 3 times before I get the plug the right way round to fit in the device, as happened with Mini- and Micro-USB. Always puzzled me that, given there were only 2 possible orientations to choose from.

Legal | privacy