Stories
Slash Boxes
Comments

SoylentNews is people

SoylentNews is powered by your submissions, so send in your scoop. Only 16 submissions in the queue.
posted by martyb on Wednesday September 04 2019, @05:55AM   Printer-friendly
from the Waiting-for-Godot,-again dept.

As reported at C|net, USB4 is ready to go.

USB4 is done, the group developing the next version of the immensely successful USB connector technology said Tuesday. USB4 doubles speeds compared to today's fastest USB 3.2 by incorporating Intel's speedy Thunderbolt technology that you already see on high-end laptops and peripherals. The USB Implementers Forum announced the completion of the technical specification Tuesday, a move that frees hardware and software engineers to get cracking building the actual products to support it.

Today's USB 3.2, which enables data transfer speeds up to 20 gigabits per second, is still something of a rarity; most of us have earlier versions of the technology that works at 5Gbps or 10Gbps. USB4 promises a speed boost to 40Gbps, helpful for things like using multiple external displays or fetching files from external hard drives.

What is the Serial Bus equivalent of, "Looks like I'm going to have to buy the White Album again."?


Original Submission

Related Stories

DisplayPort Alt Mode Updated for USB 4, Allows Uncompressed 8K @ 60 Hz HDR Video Over a USB-C Cable 11 comments

DisplayPort Alt Mode 2.0 Spec Released: Defining Alt Mode for USB4

As the tech industry gears up for the launch of the new USB4 standard, a few more parts first need to fall into place. Along with the core specification itself, there is the matter of alternate modes, which add further functionality to USB Type-C host ports by allowing the data pins to be used to carry other types of signals. Keeping pace with the updates to USB4, some of the alt modes are being updated as well, and this process is starting with the granddaddy of them all: DisplayPort Alt Mode.

The very first USB-C alt mode, DisplayPort Alt Mode was introduced in 2014. By remapping the USB-C high speed data pins from USB data to DisplayPort data, it became possible to use a USB-C port as a DisplayPort video output, and in some cases even mix the two to get both USB 3.x signaling and DisplayPort signaling over the same cable. As a result of DisplayPort Alt Mode's release, the number of devices with video output has exploded, and in laptops especially, this has become the preferred mode for driving video outputs when a laptop doesn't include a dedicated HDMI port.

If you're willing to accept Display Stream Compression... New DisplayPort spec enables 16K video over USB-C

VESA press release.

Previously: Forget USB 3.2: Thunderbolt 3 Will Become the Basis of USB 4
DisplayPort 2.0 Announced, Triples Bandwidth to ~77.4 Gbps for 8K Displays
Speed-Doubling USB4 is Ready -- Now We Just Have to Wait for Devices


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 1, Interesting) by Anonymous Coward on Wednesday September 04 2019, @06:05AM (9 children)

    by Anonymous Coward on Wednesday September 04 2019, @06:05AM (#889389)

    I'm old enough to remember dialing in to local BBS systems and being able to type on my keyboard faster than the modem could transmit the characters over the line.

    Now the text of every book I ever read in my life could probably be transferred in under a minute (and I've been a very prolific reader most of my life).

    I wonder how fast serial transmissions will be in another 20 years, when they're turning me into Soylent Green.

    • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @07:50AM (1 child)

      by Anonymous Coward on Wednesday September 04 2019, @07:50AM (#889412)

      Technically, the later revisions of USB are parallel, not serial. That is one of the ways they managed to increase the speeds so much and bring other sorts of improvements.

      • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @09:51AM

        by Anonymous Coward on Wednesday September 04 2019, @09:51AM (#889463)

        So, the Universal Serial Bus is parallel.

        This world is a lie. Probably the cake is a lie too.

    • (Score: 2) by Immerman on Wednesday September 04 2019, @02:17PM (4 children)

      by Immerman (3985) on Wednesday September 04 2019, @02:17PM (#889535)

      Did you also walk uphill both ways to school?

      It appears the first commercial modem was the Bell103, at 300bps, designed to exceed the 150bps of teletype machines. Assuming 8-bit characters (though 7- wouldn't be unusual at the time) that's 37.5 characters per second, or ~450 words per minute. I have a hard time believing you could type faster than that.

      Even if you somehow had access to one of the earlier ~100bps modems, that still allowed 150wpm, or 75wpm if half-duplex with echoing.

      I suspect the real bottleneck was not the modem, but either very noisy lines that required much error correcting, slow software, or a BBS that was either inherently slow, or sagging under the load of people using it.

      • (Score: 1) by nitehawk214 on Wednesday September 04 2019, @02:53PM (1 child)

        by nitehawk214 (1304) on Wednesday September 04 2019, @02:53PM (#889547)

        You forget about error correction, encoding, and line noise on the shitty copper cables. Also the computer at the other end was probably some 8088 POS. Seeing characters float across the screen slower than you could read or type was not common, but it did happen during the late 80's "golden age" of BBSs.

        --
        "Don't you ever miss the days when you used to be nostalgic?" -Loiosh
        • (Score: 2) by Immerman on Wednesday September 04 2019, @03:05PM

          by Immerman (3985) on Wednesday September 04 2019, @03:05PM (#889554)

          Actually, I mentioned all of those.

          I remember the echo lagging behind my typing as well, sometimes even on a 14.4 modem - just never because the modem was slow. I suppose the distinction between a slow modem and a slow connection is perhaps a bit pedantic in most conversations, but when we're discussing theoretical bandwidths and the march of technology it seems quite relevant.

      • (Score: 2) by All Your Lawn Are Belong To Us on Wednesday September 04 2019, @04:29PM

        by All Your Lawn Are Belong To Us (6553) on Wednesday September 04 2019, @04:29PM (#889583) Journal

        I know it would happen host-side sometimes on my system. On a couple of occasions I listening to the "conversation", and I could hear my transmit characters go out on typing, stop typing and get a carrier almost immediately, then response would pick up and the remaining characters would display but also instantly including the host computer's response (yet other times there'd be no buffer and all my text would send, delay, then host's response). TRS-80 M III, 300 Baud Modem I. Sometimes I miss those days.... [adafruit.com]
        I very vaguely remember parallelling a phone after the modem but before the wall to do that listening.... I remember ordinarily I'd dial the number, and as soon the other side picked up I'd flip to "Orig" for Originate and hang up the receiver.

        --
        This sig for rent.
      • (Score: 2) by pipedwho on Thursday September 05 2019, @02:44AM

        by pipedwho (2032) on Thursday September 05 2019, @02:44AM (#889820)

        Also keep in mind that in the early '80s with the Bel202/CCITT-V23 half duplex standards, there was the capability of using a 1200/75 baud modem split rate that many modems supported. This standard could be used for 600/600, but more commonly was used at 1200 baud down and 75 baud up). This was mainly to cater for the asymmetric needs of most terminal systems where the screen needed to update much quicker than data would be sent back (typed) to the server system.

        75 baud is approximately 115 wpm. A fast typist can definitely cross that boundary sustained, but even a mediocre typist can type short bursts that exceed that rate. I remember using terminals 'back in the day' where you could easily type a word or two very quickly and have to wait for the buffer to catch up before you risked overflowing it. Same problem with IBM Selectric (electric golf ball) typewriters - they were fast and could keep up at 140 wpm sustained, but you could still type in bursts that it would have to buffer.

        300baud was pretty much impossible to out-type anything more than a couple of characters in a quick typing burst - like hitting adjacent keys almost simultaneously. Keyboard debounce also maxed out at around 50ms or just over 160baud or 200wpm, so the keyboard couldn't even keep up with the 300 baud modem. But, 75 baud was definitely a noticeable slow down for a quick typist.

        A few years later once 1200 and 2400 baud full duplex became popular, error correction such as V22bis was introduced into the standards. This could potentially slow things down if there was a lot of line noise, but for normal mostly quiet lines, they sustained their advertised rates. There were no modem based error correction protocols back in the days of 300 baud systems. So the sibling commenting on error correction causing extreme slow downs (to below typing speed) is only valid for faster modems and VERY noisy lines.

    • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @02:48PM (1 child)

      by Anonymous Coward on Wednesday September 04 2019, @02:48PM (#889545)

      Are you complaining about this, or marveling about it, old man?

      • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @06:22PM

        by Anonymous Coward on Wednesday September 04 2019, @06:22PM (#889618)

        Are you complaining about this, or marveling about it, old man?

        Neither, I'm rambling nonsensically.

  • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @06:21AM

    by Anonymous Coward on Wednesday September 04 2019, @06:21AM (#889392)

    With such a noise margin, it's unlikely your devices would work consistently. It should hopefully downgrade the performance unless you have some gold plated Monster cables.

  • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @06:23AM (2 children)

    by Anonymous Coward on Wednesday September 04 2019, @06:23AM (#889394)

    Requires active cables for max speed, which leads to more of a cable mess than we have.

    • (Score: 3, Informative) by takyon on Wednesday September 04 2019, @07:40AM

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Wednesday September 04 2019, @07:40AM (#889406) Journal

      This basically consumes Thunderbolt 3 and leaves it for dead.

      https://en.wikipedia.org/wiki/Thunderbolt_(interface) [wikipedia.org]

      Compared to Thunderbolt 2, it doubles the bandwidth to 40 Gbit/s (5 GB/s), allowing up to 4-lane PCIe 3.0, 8-lane DisplayPort 1.2, and USB 3.1 10 Gbit/s.

      [...] On 24 May 2017, Intel announced that Thunderbolt 3 would become a royalty-free standard to OEMs and chip manufacturers in 2018, as part of an effort to boost the adoption of the protocol. The Thunderbolt 3 specification was later released to the USB-IF on 4 March 2019, making it royalty-free, to be used to form USB4. Intel says it will retain control over certification of all Thunderbolt 3 devices, though it will not be mandatory.

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
    • (Score: 1, Informative) by Anonymous Coward on Wednesday September 04 2019, @07:44AM

      by Anonymous Coward on Wednesday September 04 2019, @07:44AM (#889408)

      The Thunderbolt 3 specification allows for passive cables for 1.5 feet or less, and connections that are power only. It also allows for fiber optic cables and hybrid cables. With the speeds further increasing and people loathing yet another connector change, I could see the fiber optic cables becoming more widespread in their usage.

  • (Score: 2) by Arik on Wednesday September 04 2019, @09:07AM (1 child)

    by Arik (4543) on Wednesday September 04 2019, @09:07AM (#889448) Journal
    3.2 gives 20gb/s which is way more than any of your peripherals are likely to be able to utilize.

    And so that's a mostly useless metric.

    What's the maximum latency?

    Will USB finally support n-key rollover, or is it STILL inferior to PS/2?
    --
    If laughter is the best medicine, who are the best doctors?
    • (Score: 2) by Immerman on Wednesday September 04 2019, @02:49PM

      by Immerman (3985) on Wednesday September 04 2019, @02:49PM (#889546)

      USB doesn't prevent n-key rollover - that's entirely down to the keypad matrix used by the keyboard manufacturer, and the USB protocols that they choose to implement.

      If they only implement the baseline HID "boot protocol" they're limited to 6-key-rollover (plus the 8 modifier keys). If they also support HID "report protocol" then they can provide n-key-rollover, assuming the computer supports it as well.

  • (Score: 3, Interesting) by jmichaelhudsondotnet on Wednesday September 04 2019, @12:56PM (2 children)

    by jmichaelhudsondotnet (8122) on Wednesday September 04 2019, @12:56PM (#889502) Journal

    Will this mean usb4.1, usb 4.1.3, usb 4.1.35, each with different characteristics for different manufacturers of ports and displays?

    Or did they get it right the first time?

    Because usb 3/thunderwhatever has been a collosal fuckup that should have shaken everyone in the world's confidence in humanity to produce any standard whatsoever, meaning progress went backwards in a sortof big way.

    50 years from now there are going to be people trying to get usb 3 things to work and looking for the rare cable they need to access a device manufactured in 2014.

    A thousand years from now, they are still going to be using usb 2.0, because that shit just works. If I were building a spaceship I would make all of the interconnections usb 2.0 just because I know when I'm in space I will sleep well. If I knew there was a single usb 3.0 device in my spaceship, I wouldn't be able to sleep until I had ripped it out.

    • (Score: 2) by takyon on Wednesday September 04 2019, @01:55PM

      by takyon (881) <takyonNO@SPAMsoylentnews.org> on Wednesday September 04 2019, @01:55PM (#889521) Journal

      It's pretty bad, but I think Bluetooth 5 might be shaping up to be worse. The best features are optional, an old ass mandatory audio codec is still in use with no Opus.

      With USB, a hybrid cable delivering the max data rate and all the power the spec is rated for would solve (almost) everybody's problems. It's just that it would be too expensive I guess.

      --
      [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
    • (Score: 2, Touché) by nitehawk214 on Wednesday September 04 2019, @02:58PM

      by nitehawk214 (1304) on Wednesday September 04 2019, @02:58PM (#889550)

      When USB 4.1 comes out, they will retroactively rename USB 4.0 to USB 4.1 Gen a1 mk II

      --
      "Don't you ever miss the days when you used to be nostalgic?" -Loiosh
  • (Score: 1, Interesting) by Anonymous Coward on Wednesday September 04 2019, @01:51PM

    by Anonymous Coward on Wednesday September 04 2019, @01:51PM (#889518)

    If USB4 is essentially Thunderbolt, does that include DMA (and DMA exploits)?

  • (Score: 0) by Anonymous Coward on Wednesday September 04 2019, @08:21PM

    by Anonymous Coward on Wednesday September 04 2019, @08:21PM (#889679)

    10Gbit usb nic in your pocket for emergencies.

(1)