More USB-C speed won't fix users' problems with cables

2»

Comments

  • Reply 21 of 36
    Why not have a way to look up cable specs in settings. It would be an easy way to confirm your cable is licensed, manufacture, power, and speed rating. 
  • Reply 22 of 36
    JP234 said:
    If you can't tell what bandwidth you're getting from your USB cable/device, you probably don't need to know.
    My iPad Pro uses a USB-C cable for charging. My wife’s M1 MacBook Air uses a USB-C cable for charging. The iPad cable doesn’t work for charging the Air (and maybe vice versa). I can’t tell by looking at them which one works with which device but we obviously need to know or at least one device won’t be charging when we think it is. 
    Don’t iPad Pros come with a 1-metre cable and MacBook Airs with a 2 metre?  That’d be one way to tell I would’ve thought.   :smile: :wink: 

    Also, are you sure they’re both Apple cables and one isn’t a knock-off?

    Don’t worry, I’m completely on your side and this author’s on wanting better clarity.  It’s becoming more difficult to know what’s what.
    edited September 2022
  • Reply 23 of 36
    dewmedewme Posts: 5,415member
    The EC has the pull to lock our devices into this deeply flawed physical connector, but hasn't gone far enough to actually make it useful.

    Mandating labelling of cable ends and sockets with protocol, speed, and power capacity would've been more useful - but they're politicians and not technologists and can only pick something that's currently available and mandate its use.

    I've always found USB connectors to be deeply flawed since plug ends always have designated a shaped hollow piece of metal which is used as a channel that goes into a similarly shaped hole on the device with the plug end being female and the device end being male. This results in an unstable connection with tolerances too tight to support blind docking (mating the connectors without being able to see them).

    Common sense dictates that the plug end should be male and the socket end female. The physical connection could then sport a connection with less critical tolerances making blind docking possible. An optional lock like an ethernet cable would also be desirable since that connector may also be used for high speed or high power mission critical connections whose interruption could have dire consequences.

    Take a look at an RJ-type connector (RJ-11 phone or RJ-45 ethernet) connector to see what I mean. Those connections are relatively easy to blind dock and have locking clips so the connections are rarely interrupted.

    Now compare that to USB-C and try to insert a plug into a socket without looking and see how easy mating those connectors are - the tolerance of the sleeve is so high your practically need to have eyes on to do it easily.
    Very good points. Not sure whether I'd categorize the very exacting insertion angle requirements as being "deeply flawed" but I know exactly why you are bringing this issue up. I personally find the previous generation Thunderbolt ports to be very intolerant of insertion angle misalignments as well. When you're reaching around the back of a 27" iMac to plug in a TB or USB-C connector it's easy to miss the "hole" and scrape the connector across the back of the Mac trying to find the hole. In most cases I end up turning or tipping the iMac so I can see the target hole.

    It would be very nice to have a standard connector where the socket side was more tolerant of insertion alignment errors. A lot of mechanical connectors that involve a plug and socket type of arrangement, like screws, bolts, and rods do provide a chamfered transition on the plug side to make it easier to get the plug part into the socket. Once the plug part is started in the socket it can then be aligned to the exacting mating requirements. If you look at an RS-232 DB style connector you'll see the metal edges of the plug rim are slighly rounded/chamfered for this exact reason. 

    Apple's MacBook MagSafe power connector uses magnets that help with plug+socket alignment. It's a cool and clever idea, but a lot of the other connectors could provide for much easier insertion simply by adopting some of the tried & true techniques that have been around for many decades. The fact that USB-C is reversible should have made coming up with a solution even easier. I haven't looked at a USB-C plug under a magnifier, but I wouldn't be surprised to see some chamfering on the connector. Maybe they needed to do something on the socket side as well, but yeah, these things are very very picky. 
    tmay
  • Reply 24 of 36
    chasmchasm Posts: 3,335member
    I wish the USB-IF would get a grip and think about the buyers for just one hot second. The USB-C form factor is fantastic: reversible, reliable (in my experience), satisfying snap when its in, and incredibly versatile. The naming convention, on the other hand, is so bewilderingly stupid I suspect the marketers of the Golgafrinchan Ark “B” are in charge of that side of things.

    What the hell is wrong with calling this major advance in the USB spec … wait for it … USB5? And mandating that any cable labeled USB5 must deliver the full charging and full speed capability of the spec? If you want to label it USB5 PD, then it must deliver the full electrical spec (and can be used at the “passive” speed for data/video). Done. That’s it. Two labels for cables. Lesser cables don’t get to use the USB5 label.

    To quote from Zoolander, “I feel like I’m taking crazy pills!”
    edited September 2022 crowleyMplsPsphericFileMakerFeller
  • Reply 25 of 36
    The total separation of physical connector from its purpose must have been a fantasy of the various USB committees. 'One connector - a thousand standards'.

    This absolute mess needs rationalising into levels where each level guarantees a minimum set of service standards in terms of data rate, charging power, Thunderbolt standard, video capabilities AND these must be indicated on both sides of both connectors of all USB-C cables. It won't stop the fakes but if you buy from a reputable manufacturer from an authorised seller you at least know what your are getting. 
    FileMakerFeller
  • Reply 26 of 36
    DAalsethDAalseth Posts: 2,800member
    I disagree with the USB-C critics here. At the moment I have a mix of USB-C and Lightning and I find USB-C to be easier to plug in, either directly, or when I’m reaching around to do it blind, and more solid. I really prefer it.
    But the problem is the cable. First I’ve come across several cables that don’t charge, just do data, or only charge and don’t do data. I’ve assumed these were defective and threw them away. Apparently those are some of the “standard” forms? Well F-that, a cable needs to do everything or it’s garbage to me. That was a very stupid thing for them to allow. Then there are cables that look identical but will transfer data at different rates? Oh but they are marked on the package that they came in. The package that I immediately pull them out of losing the marking.  Another very stupid move.
    tl:dr USB-C is a very good plug handicapped by very poor implementation.
    FileMakerFeller
  • Reply 27 of 36
    sphericspheric Posts: 2,572member
    DAalseth said:
    I disagree with the USB-C critics here. At the moment I have a mix of USB-C and Lightning and I find USB-C to be easier to plug in, either directly, or when I’m reaching around to do it blind, and more solid. I really prefer it.
    But the problem is the cable. First I’ve come across several cables that don’t charge, just do data, or only charge and don’t do data. I’ve assumed these were defective and threw them away. Apparently those are some of the “standard” forms? Well F-that, a cable needs to do everything or it’s garbage to me. That was a very stupid thing for them to allow. Then there are cables that look identical but will transfer data at different rates? Oh but they are marked on the package that they came in. The package that I immediately pull them out of losing the marking.  Another very stupid move.
    tl:dr USB-C is a very good plug handicapped by very poor implementation.
    You're not disagreeing with the USB-C critics. That is EXACTLY what we're critical of. 

    Now step back and think for a second: How is a manufacturer in 2016 going to build a cable that will supply all protocols and voltages that will be part of the standard in 2023? How the hell is the manufacturer supposed to predict that the cable will need 240W power delivery and active electronics for 80 Gb/sec data transfer, when none of that has even been proposed yet? 

    There really was no sensible way to approach it, except to REQUIRE labelling of cables and ports. And that would *still* be confusing, and would require completely different naming structures, as there are, at current, potentially SIXTY different kinds of USB-C cables, if you allow for every possible combination of specs the USB-C connection protocols technically offer. 

    Or, you'd have to drastically reduce the number of options, but that would mean that, for example, USB-C power supplies to come with cables that do way more than just supply power. Increasing costs. Etc. This would still be my preferred option: Make cables USB-C v1, v2, v3, v4 up to v8 or whatever, and make labelling on ports and cables MANDATORY as part of licensing.

    The idiotic naming conventions are just the tip of a huge shitberg. 
    edited September 2022 FileMakerFeller
  • Reply 28 of 36
    dewmedewme Posts: 5,415member
    spheric said:
    DAalseth said:
    I disagree with the USB-C critics here. At the moment I have a mix of USB-C and Lightning and I find USB-C to be easier to plug in, either directly, or when I’m reaching around to do it blind, and more solid. I really prefer it.
    But the problem is the cable. First I’ve come across several cables that don’t charge, just do data, or only charge and don’t do data. I’ve assumed these were defective and threw them away. Apparently those are some of the “standard” forms? Well F-that, a cable needs to do everything or it’s garbage to me. That was a very stupid thing for them to allow. Then there are cables that look identical but will transfer data at different rates? Oh but they are marked on the package that they came in. The package that I immediately pull them out of losing the marking.  Another very stupid move.
    tl:dr USB-C is a very good plug handicapped by very poor implementation.
    You're not disagreeing with the USB-C critics. That is EXACTLY what we're critical of. 

    Now step back and think for a second: How is a manufacturer in 2016 going to build a cable that will supply all protocols and voltages that will be part of the standard in 2023? How the hell is the manufacturer supposed to predict that the cable will need 240W power delivery and active electronics for 80 Gb/sec data transfer, when none of that has even been proposed yet? 

    There really was no sensible way to approach it, except to REQUIRE labelling of cables and ports. And that would *still* be confusing, and would require completely different naming structures, as there are, at current, potentially SIXTY different kinds of USB-C cables, if you allow for every possible combination of specs the USB-C connection protocols technically offer. 

    Or, you'd have to drastically reduce the number of options, but that would mean that, for example, USB-C power supplies to come with cables that do way more than just supply power. Increasing costs. Etc. This would still be my preferred option: Make cables USB-C v1, v2, v3, v4 up to v8 or whatever, and make labelling on ports and cables MANDATORY as part of licensing.

    The idiotic naming conventions are just the tip of a huge shitberg. 

    Lamenting the plethora of issues around USB is one thing. Doing something about it is extraordinarily more difficult. I'm certain those responsible for USB standards and governance would love to hear about pragmatic solutions that not only address forward looking concerns but backward compatibility ones as well. We've reached a point where it's nearly impossible to unwind all of the decisions and consequences of non-decisions that have already taken place, but perhaps we can get the industry to agree on standards and conventions going forward.

    Instilling labeling requirements may help. Some sort of coding is needed. Text labels? Text would have to be standardized for languages, locales, and units of measure. Perhaps a color coding standard could also be applied, where cable jackets/sheaths are specifically color-coded according to application compatibility. Color coding conventions have been used for over 100 years for electrical components like resistors. Electrical connection terminal connectors have a color coding convention based on wire gauge compatibility. Not everyone follows it. There was a loose convention, which fell short of being a de facto standard, of color coding USB3 A-sockets with blue guides but not all vendors, Apple included, went along with it. I think I've seen other colors on USB A-sockets, e.g., yellow. The obvious issue with color coding is the need for a "decoder ring" to match up the color codes to their meanings. Camera equipped smartphones could make this job much easier.

    While the situation is kind of a mess, I suppose it could be worse. There's always been a strategy in hardware design, at least in some industries like military and industrial, to put the side of the removable connection that's more likely to fail, get damaged, or more difficult to repair on the removable component, i.e., the cable. In other words, if compromises were required, like male pins that are more likely to get bent, put them on the cable side rather than the equipment side. I believe that this may contribute somewhat to there being a culture of tolerance for shoving the vast majority of the crappiness and ugliness of a connection standard into the cables. After all, if you're an equipment vendor, all those cable issues are someone else's problem, not your problem.
    FileMakerFeller
  • Reply 29 of 36
    sphericspheric Posts: 2,572member
    It needed to happen at least six years ago. 

    There's no real avoiding or remedying the clusterfuck we have now, but at least they could try to avoid jerking on the cookie by adding ever more idiotically confusing descriptions with multiple different version numbers within a single name. 
  • Reply 30 of 36
    nrg2nrg2 Posts: 18member
    Despite claims here that there's nothing wrong with the USB-C connector, I find that it really sucks.



    If USB plugs had a simple spade connector like lightning I think the cables would be reliable - but with that stupid female connector within a sleeve design, you just don't get a durable and reliable connection.
    Exactly my experience. USB-C connectors and ports are pure garbage. Working in IT support, we are regularly replacing both motherboards from failed USB-C/Thunderbolt ports and the Thunderbolt docks because the connectors no longer work. (The docks regularly lose some sort of functionality either display, data or power delivery.) I do think part of this is our company primarily uses brand “D” computers and docks and the fault may be heavily on ”D’s” poor engineering skills. We have a number of Mac users too, but those never have had such issues. At one point the onsite service tech sent by “D” to replace a motherboard said they are replacing boards all the time from such issues. He even said Apple had the USB-C ports first and they are the only one to have gotten it right. 
  • Reply 31 of 36
    mpantonempantone Posts: 2,065member
    dewme said:
    spheric said:
    DAalseth said:
    I disagree with the USB-C critics here. At the moment I have a mix of USB-C and Lightning and I find USB-C to be easier to plug in, either directly, or when I’m reaching around to do it blind, and more solid. I really prefer it.
    But the problem is the cable. First I’ve come across several cables that don’t charge, just do data, or only charge and don’t do data. I’ve assumed these were defective and threw them away. Apparently those are some of the “standard” forms? Well F-that, a cable needs to do everything or it’s garbage to me. That was a very stupid thing for them to allow. Then there are cables that look identical but will transfer data at different rates? Oh but they are marked on the package that they came in. The package that I immediately pull them out of losing the marking.  Another very stupid move.
    tl:dr USB-C is a very good plug handicapped by very poor implementation.
    You're not disagreeing with the USB-C critics. That is EXACTLY what we're critical of. 

    Now step back and think for a second: How is a manufacturer in 2016 going to build a cable that will supply all protocols and voltages that will be part of the standard in 2023? How the hell is the manufacturer supposed to predict that the cable will need 240W power delivery and active electronics for 80 Gb/sec data transfer, when none of that has even been proposed yet? 

    There really was no sensible way to approach it, except to REQUIRE labelling of cables and ports. And that would *still* be confusing, and would require completely different naming structures, as there are, at current, potentially SIXTY different kinds of USB-C cables, if you allow for every possible combination of specs the USB-C connection protocols technically offer. 

    Or, you'd have to drastically reduce the number of options, but that would mean that, for example, USB-C power supplies to come with cables that do way more than just supply power. Increasing costs. Etc. This would still be my preferred option: Make cables USB-C v1, v2, v3, v4 up to v8 or whatever, and make labelling on ports and cables MANDATORY as part of licensing.

    The idiotic naming conventions are just the tip of a huge shitberg. 

    There's always been a strategy in hardware design, at least in some industries like military and industrial, to put the side of the removable connection that's more likely to fail, get damaged, or more difficult to repair on the removable component, i.e., the cable. In other words, if compromises were required, like male pins that are more likely to get bent, put them on the cable side rather than the equipment side.
    This is why the Lightning connector is superior in basic design to the USB-C connector. The male side is on the cheap-to-replace cable. Understandably the EU picked the popular USB-C connector as the common charging standard because there's no freely available superior alternative.
    edited September 2022 spheric
  • Reply 32 of 36
    macguimacgui Posts: 2,369member
    netrox said:
    But why do we want just PD cable? It's actually promoting waste. It cannot be more economical to have just PD cable. 
     


    The value of a PD-only cable is being able to charge from any USB-C port in any country and not fear data being stolen or app installed on your device from a malevolent source. It's not something most users will experience but it's a worthwhile precaution to have and use. Security shouldn't be about economy. There's far too much of that mindset already.
    FileMakerFeller
  • Reply 33 of 36
    chasm said:
    I wish the USB-IF would get a grip and think about the buyers for just one hot second. The USB-C form factor is fantastic: reversible, reliable (in my experience), satisfying snap when its in, and incredibly versatile. The naming convention, on the other hand, is so bewilderingly stupid I suspect the marketers of the Golgafrinchan Ark “B” are in charge of that side of things.
    Nah. That bunch of mindless jerks were the first up against the wall when the revolution came. :wink:
  • Reply 34 of 36
    LaddLadd Posts: 2member
    Over the years, the spec has expanded so much, I've purchased a cable tester and identifier to be sure that what I've bought is what I got, and to guarantee my own labels are accurate.

    Still no reply as to what brand/model of cable tester the author has purchase!
  • Reply 35 of 36
    Mike WuertheleMike Wuerthele Posts: 6,868administrator
    Ladd said:
    Over the years, the spec has expanded so much, I've purchased a cable tester and identifier to be sure that what I've bought is what I got, and to guarantee my own labels are accurate.

    Still no reply as to what brand/model of cable tester the author has purchase!
    Read the entire comment thread.
    edited October 2022
Sign In or Register to comment.