Announcement

Collapse
No announcement yet.

Intel Sends Out Initial USB 4.0 Support For The Linux Kernel

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    Originally posted by wizard69 View Post

    Your thought don’t seem to be valid! USB has evolved magnificently over the years. We now have one connector that solves many problems.
    I suppose that's true if we simply omit the 23 iterations between then and now, most of which are still in use, that have no guarantee of interoperability even though they're supposed to.

    Just take look at the micro 2.0 ones. It's a PITA to have to look at the individual copper pins on the inside of a cable to try to determine if it's a charging only cable or a data cable...and then you have oddities like Motorola factory debugging USB cables that have all the pins present but have some pins swapped around to activate special modes in their phones or charging-only cables with all the pins present and no data wires on the inside. Yeah, that was just great over the years

    Comment


    • #12
      Originally posted by wizard69 View Post

      Your thought don’t seem to be valid! USB has evolved magnificently over the years. We now have one connector that solves many problems.
      An alternative take is that we now have one connector that is incompatible with itself.

      IT may or May not support:
      • USB
      • Thunderbolt
      • USB4
      • DisplayPort over Thunderbolt
      • DisplayPort over USB4
      • Power
      • PCie Passthrough
      (Am I missing any?)

      If cables were universal it would solve a lot of things but right now they cherry pick which of the above they support.

      Atleast we are past the point where the cable could cause the device to explode and/or catch fire

      Comment


      • #13
        Originally posted by You- View Post

        An alternative take is that we now have one connector that is incompatible with itself.

        IT may or May not support:
        • USB
        • Thunderbolt
        • USB4
        • DisplayPort over Thunderbolt
        • DisplayPort over USB4
        • Power
        • PCie Passthrough
        (Am I missing any?)

        If cables were universal it would solve a lot of things but right now they cherry pick which of the above they support.

        Atleast we are past the point where the cable could cause the device to explode and/or catch fire
        Until Samsung says "hold my beer" and releases a new phone.

        But yeah, even color coding the cables would be a start. Don't get me wrong, it'll get really old and frustrating when we start having to name off colors like "red, orange, green, purple, blue, and yellow" like we're in a damn Fruity Pebbles commercial while cross-referencing the colors with a chart, but that's better than the current method of "plug it in with fingers crossed" because there are 6 identical looking cables in the drawer that support different features. I could have used something like that with my generic and unlabeled HDMI 1.0 and 2.0 cables.

        Comment


        • #14
          Originally posted by You- View Post
          Atleast we are past the point where the cable could cause the device to explode and/or catch fire
          Now the cable itself can catch fire and/or short hard because it's used to carry power levels it was never designed for
          That's a sidegrade at best.

          Comment


          • #15
            Originally posted by You- View Post

            An alternative take is that we now have one connector that is incompatible with itself.

            IT may or May not support:
            • USB
            • Thunderbolt
            • USB4
            • DisplayPort over Thunderbolt
            • DisplayPort over USB4
            • Power
            • PCie Passthrough
            (Am I missing any?)

            If cables were universal it would solve a lot of things but right now they cherry pick which of the above they support.

            Atleast we are past the point where the cable could cause the device to explode and/or catch fire
            And it gracefully degrades to the highest common version given the host, device and cable you're using, which is really the best we can hope for.
            What's the alternative here? Any phone that wants to be able to read a USB stick has to also support displayport and be capable of providing 100W to another device? Wired mice and headphones have to have thick inflexible cables <1m long that support 40Gbps operation? The USB IF employs a private mercenary army to seek and destroy factories in east asia churning out cheap faulty cables?
            At the end of the day there's a whole load of conflicting requirements for different devices and mandating that everything has to support the highest speed and power is a great way to make sure nobody uses it. Everything using a common physical, electrical and protocol standard so devices can generally interoperate where it logically makes sense for them to be capable of doing so is a massive improvement over having dozens of different cables and adapters.

            USB4 and thunderbolt 3 are so similar it's likely that USB4 hosts/devices will be backward compatible with TB3, and "DisplayPort over Thunderbolt" and "DisplayPort over USB4" are basically the same thing. Also, there's only really 2 types of USB-C cable, the thin/cheap 4-wire type that only supports USB2.0 and the full featured type that supports the rest. You can't make one that supports USB 3.0 but not displayport for example.

            Comment


            • #16
              The problem isnt the device. It should not be limited by the cable.
              • Connect a PC to a Displayport monitor. Oops, wrong type of USB Type C cable!
              • Connect a USB Type C cable for power to a laptop, oops, its Thunderbolt only!

              Ideally all cables should be comparable and features decided by what they are connected to.

              I understand that they cant control quality of most cables due to the nature of the free to impelent status of USB, but maybe have a quality assurance program that if passed gives the cables a different colour? That way you will know those cables are capable.

              Comment


              • #17
                Originally posted by You- View Post
                The problem isnt the device. It should not be limited by the cable.
                • Connect a PC to a Displayport monitor. Oops, wrong type of USB Type C cable!
                • Connect a USB Type C cable for power to a laptop, oops, its Thunderbolt only!

                Ideally all cables should be comparable and features decided by what they are connected to.

                I understand that they cant control quality of most cables due to the nature of the free to impelent status of USB, but maybe have a quality assurance program that if passed gives the cables a different colour? That way you will know those cables are capable.
                Actually this problem has existed for years since USB 2.0. I have two mini-USB cables to charge my kindle but neither of them supports data transmission, which is good imo because I do not want accidentally mount my kindle to my PC.

                Comment


                • #18
                  Originally posted by starshipeleven View Post
                  USB-C is a connector specification. It does not define the cable.
                  This is entirely incorrect. The USB TYpe C specification is specifically titled "USB Type-C Cable and Connector Specification", and most definitely DOES define the cables. It contains a huge amount of requirements for the electrical and physical requirements of the cables, what signals have to be wired for what types of USB-C cables, etc.

                  Release 1.3 of the specification, dated July 14, 2017 and included as part of the USB 3.2 specification, includes requirements for cables in:

                  section 3.3 "Cable Construction and Wire Assignments"
                  section 3.4 "Standard USB Type-C Cable Assemblies"
                  section 3.5 "Legacy Cable Assemblies" (which is for Type C to legacy cables)
                  section 3.6 "Legacy Adapter Assemblies"
                  section 3.7 "Electrical Characteristics"
                  section 4.9 "Electronically Marked Cables"

                  and additional information in other portions of the standard. Over 46 pages out of 241 total are devoted to cable specifications.




                  Comment


                  • #19
                    Originally posted by starshipeleven View Post
                    Now the cable itself can catch fire and/or short hard because it's used to carry power levels it was never designed for
                    That's a sidegrade at best.
                    If the USB cables and equipment meet the USB specifications (including USB Type C, and optionally USB PD), then the cable will not catch fire or reach dangerous temperatures. The USB specifications give strict requirements for cables including voltage and current requirements. All USB-C cables are required to safely support 3A of Vbus current. The host and device are PROHIBITED by the standard from providing more than 3A of current on Vbus unless the cable is electronically marked as being capable of supporting that.

                    There may be many substandard cables in the world which do not actually meet the specifications, but that's nothing unique to USB or USB-C. Any cables carrying the USB trademark that do not actually meed the USB specifications are actually unlawful to sell in the United States and probably most other countries, as that is a trademark violation.

                    Comment


                    • #20
                      40 Gbps you say?

                      And how are we doing with ethernet these days?

                      Comment

                      Working...
                      X