USB 3.2 Boosts Transfer Speeds To 2GBps Via Type-C

Get ready for more variance in USB support and performance: The USB Implementers Forum (USB-IF) released the specifications for USB 3.2, and the update increases maximum data transfer speeds to 2 GBps (that's gigabytes per second, not gigabits per second), thanks to the inclusion of two sets of SuperSpeed pins and wires in USB Type-C. You can get those speeds only via a USB Type-C connector.

USB-IF explained its goals for USB 3.2 in the protocol's finalized specs:

USB 3.2's goal is to enable devices from different vendors to interoperate in an open architecture, while maintaining and leveraging the existing USB infrastructure (device drivers, software interfaces, etc.). The specification is intended as an enhancement to the PC architecture, spanning portable, business desktop, and home environments, as well as simple device-to-device communications. It is intended that the specification allow system OEMs and peripheral developers adequate room for product versatility and market differentiation without the burdens of carrying obsolete interfaces or losing compatibility.

That's good news. People are using more devices and dealing with more data than ever. Combine that with humans' propensity for impatience, and it's clear that data transfer protocols, among other standards handled by USB-IF, will only become more important over time. The problem: Figuring out what exactly a specific device is capable of is made difficult by manufacturer branding (or lack thereof) and other issues.

In 2015, we detailed what exactly you can expect from products that claim to support USB 3.0, USB 3.1 Gen 1, USB 3.1 Gen 2, and USB Type-C. This isn't always an easy task—manufacturers are not required to use the USB-IF's branding on their devices, for example, which can leave people scratching their heads as they try to figure out if a product that boasts support for "USB 3.1" is compatible with Gen 1 (5 Gbps) or Gen 2 (10 Gbps).

The USB-IF came up with two brands: SuperSpeed (USB 3.1 Gen 1) and SuperSpeed+ (USB 3.1 Gen 2) that can be used with other logos to indicate which data transfer protocol a device supports, whether or not it's compatible with USB Power Delivery, and so on. Other marks can also be used to show if a particular port is also compatible with DisplayPort or Thunderbolt 3. All of these indicators are optional.

USB 3.2 will add another wrinkle to the process of sussing out a device's exact capabilities. As we stated above, the updated protocol supports up to 2 GBps data transfer speeds, courtesy of two sets of SuperSpeed pins and wires. Everything works over USB Type-C... provided you use a SuperSpeed+ USB 10 Gbps-certified cable, as indicated by the presence of a SuperSpeed+ logo on the cable or its packaging. Which will be present on sometimes.

That means device manufacturers can continue to use USB Type-C ports in their products while adding support for the USB 3.2 standard. We suspect that many companies will trumpet the standard's increased speeds—saying a product is twice as fast as the competition is a big selling point—to start. Calling this USB 3.2 instead of USB 3.1 Gen Whatever will also help. But we expect things to remain murky nonetheless.

Nathaniel Mott
Freelance News & Features Writer

Nathaniel Mott is a freelance news and features writer for Tom's Hardware US, covering breaking news, security, and the silliest aspects of the tech industry.

  • poochiepiano
    I'm sort of conflicted as this labeling system is really hard for consumers at large to grasp. People are still becoming aware of USB 3.0, much less 3.1. Now there's no 3.0, but rather 3.1 Gen 1 and 3.1 Gen 2, which is enough of a headache. Now there's USB 3.2? I haven't even mentioned Type-A and Type-C yet. This is easy enough for enthusiasts to get straight, but for non-tech-inclined consumers, it's a disaster. I'm only like, 80% certain everything I wrote above is correct.

    But what's the alternative, not getting faster transfer speeds? Guess we'll just have to deal with confusing nomenclature. Whatever... at the end of the day, if it fits in the port, you'll transfer the data by sometime.
    Reply
  • pabloxmena
    @POOCHIEPIANO 100% with you.
    Reply
  • Emerald
    If USB 3.2 will use two sets of SuperSpeed pins and wires, shouldn't we expect speeds of up to 20GB/s?
    Reply
  • deball14
    20gb/s mate.
    Reply
  • EdwardN
    Really not that complicated. A is rectangle, C is rounded rectangle and reversible. Everything else is revisional. 1.1 - 2.0 - 3.11 - 3.12 - 3.2 are just the protocol details. Most devices you buy come with the cable you need.
    Reply
  • velocityg4
    They should have made things clearer for consumers. Call it USB 3.1 Gen 2 Rev B.:sarcastic:
    Reply
  • collin3000
    3.1 Gen 2 Rev Be wouldn't clear things up for non techies.
    Call it "Superspeed ++" and slap a number with "2x as fast". That's the "easiest" way for consumers. And even then they'll be confused.
    Reply
  • jasonkaler
    The naming is really bad.
    In my opinion, type C should have been called USB 4 and the different variants of USB 3 should have simply been .1 .2 and .3
    Reply
  • derekullo
    I miss the easy way to remember them.

    USB 1.1 = Forever
    USB 2.0 = 40 times faster than forever
    USB 3.x = REALLY FAST

    As far as 128 gigabyte flash drives are concerned.


    Nobody mentioned it yet but, 20Gb/s = 2500 megabytes a second ... PCI-E SSD speed


    You would need to a 5 bay NAS full of Samsung 850s to fully exploit that kind of speed.

    Or a NAS with a single Samsung 960.

    I had to google that but m.2 nas's do exist.

    https://www.qnap.com/en-us/product/tbs-453a

    Not 3.2 or even 3.1 compatible but it does exist.



    Still praying for 10 Gb Ethernet to become standard.




    Reply
  • weilin
    @Emerald

    You are both correct. the 20gb/s you're quoting is 20 gigabits per second. Toms' is quoting 2gB which is 2 gigabytes per second. The former is the physical signaling rate (with overhead included). The latter is useful data payload rate (without overhead, also the bit vs byte factor).
    Reply