Making Sense of the USB Standards

Content:

USB ports are now ubiquitous. From humble beginnings, USB is now used to connect pretty much anything, with USB ports used for data transfer, charging ports, and even display outputs.

Unfortunately, the USB Implementers Forum (USB-IF), who control the USB standard, have seen fit to make the naming scheme used for the latest specs as confusing as possible. Let’s take a look at this naming scheme, and try to make sense of it all.

Ports vs Standards

Before doing so, let’s first take a look at the ports used for USB connections. It is important to understand that the physical port itself, and the USB standard, are independent of one another. There is no reason, for example, that a USB physical port could not be replaced with an HDMI port, provided the USB device on the other end had a complementary port and pinout. Both devices would still communicate using the USB protocol, through the HDMI port.

USB Type A

The above connector is a USB type A port. This port has existed since the very beginning, and has been used for all versions of the USB protocol until USB4. For USB 3.0, USB Type A often, but not always, features blue plastic inside the connector. This type of connector can be considered the standard USB port.

USB Type B

These ‘double deck’ connectors are often used for devices such as printers. Cables usually feature a USB A male connector on the other end, allowing them to be connected to a standard PC USB port.

USB 3.0 changed this connector, adding a second layer of pins above the original connector, as seen above. It is still possible to plug them into a USB 2.0 host port however – the port on the other end does not have to support USB 3.

USB Mini & Micro B

I’ve put these together, as the two serve a similar purpose. USB Mini B is rarely seen these days, as it was deprecated following the release of Micro B.

USB Micro B is extremely common, and since an informal agreement with the EU back in 2009, an increasing number of phone manufacturers made Micro USB the standard port for charging your mobile phone.

Again, USB 3.0 changed the shape of the Micro USB connector, while retaining backwards compatibility with the old version.

Both of these ports can be used with USB A hosts.

USB Type C

The newest member of the USB family, the USB C port is rapidly replacing the USB Micro B port in new devices. These ports feature more pins than previous ports, and while they can be used with USB A cables for some uses, others require cables supporting the full pinout.

They are also used as charging ports, owing to new power delivery standards allowing faster charging, and higher voltage through the port. The EU is in the process of requiring USB C to be used for charging ports in the near future.

USB C is also used as a display connector (in place of HDMI, for example), owing to its compact size. In this state, the port will utilise the DisplayPort protocol, rather than the USB protocol. This is a good example of the USB physical sockets, and the USB protocol, being essentially separate from one another.

USB4 requires the USB Type C connector, dropping support for all other types of USB connector.

So, that sums up the main types of physical USB port.

The USB Standards

On the other side, we have the USB protocol itself. This protocol governs the way in which data is transferred between devices, ensuring that the device on either end is able to interpret the data being transferred.

USB 1.0/1.1

USB 1.0 was released in 1996. Supporting data transfers of up to 12MBit/s (1 MBit/s = 0.125MB/s), it did not see much adoption, before it was replaced by USB 1.1. Version 1.1 added a ‘low speed’ 1.5MBit/s mode, with the 12MBit/s renamed to ‘full speed’ mode.

To enter ‘full speed’ mode, a device was required to supply 5V on the D+ pin, through a pull up resistor. This signalled to the host during enumeration that the connected device supports the higher speed.

USB 1.0 and 1.1 was paired with the USB Type A and USB Type B physical ports; no miniature version existed at this time.

USB 2.0

Released in 2000, USB 2.0 came with a major speed boost. Data transfers of up to 480MBit/s were possible in what was termed ‘high speed’ mode, with backwards compatibility retained with the previous 2 modes.

Device numeration was similar to ‘full speed’ mode. However, this time, an additional negotiation process takes place just after the bus is reset to switch to full speed mode. This extra negotiation lets the host know that ‘high speed’ mode is supported.

Over time, USB 2.0 evolved to support additional features, including:

• USB On The Go (OTG): Allows devices to connect to one another, without a host.
• Battery Charging Specifications: Allowing up to 1.5A of power draw, while also having an active data connection.
• Power Delivery: Allows a charging rate above the standard 0.5A.

Alongside USB 2.0, the Mini and Micro ports were released, adding smaller ports for the first time. At this point in time, the connectors were still solely used with the USB protocol.

USB 3.0

Here’s where things start to get more complicated.

Compared to USB 2.0, 3.0 adds an extra 4 pins to each connector, to facilitate the SuperSpeed 5GBit/s transfer speed. The pinout is designed to still provide the 5 base pins required for USB 2.0, retaining backwards compatibility.

The extra pins allow two data paths, instead of 1 – one to send data, the other to receive data. In practice, this means in theory it is more feasible to obtain transfer rates close to the theoretical maximum.

Power delivery for charging devices is increased to 0.9A, with the standard power delivery rate bumped from 0.1A to 0.15A.

With the advent of newer standards, USB 3.0 also became defined as both USB 3.1 Gen 1, and USB 3.2 Gen 1×1. The three names all refer to this exact standard, there is no difference between them whatsoever.

USB 3.1

This is where the USB-IF really seemed to lose their minds.

USB 3.1 ups the speed again, to 10GBit/s, with the new standard referred to as ‘SuperSpeed+’. The standard is otherwise very similar to USB 3.0, the main difference being a change in how the data is encoded, 8b/10b to 128b/132b. Essentially, this means data is sent in larger chunks, reducing overhead. This will become important in a minute.

USB 3.1 retains backwards compatibility with all previous versions of the protocol.

With USB 3.1, a new naming scheme was introduced, with the new SuperSpeed+ standard known as USB 3.1 Gen 2. Later, with the release of USB 3.2, this standard also gained the moniker USB 3.2 Gen 2×1.

USB 3.2

Warning – headaches will ensue from reading this section.

The next change to the standard further boosted the transfer speed. While still known as ‘SuperSpeed+’, the new standard, supporting up to 20GBit/s, was also named USB 3.2 Gen 2×2.

Alongside this, a second USB 3.2 standard was released. Similar to 3.1, this standard allows transfers of up to 10GBit/s. However, in this instance, the encoding type is 8b/10b, rather than 128b/132b used in the previous 10GBit/s standard.

USB 3.2 is the final version (to date) in the USB 3 series. So, let’s summarise the USB 3 naming conventions.

I should also note, that the SuperSpeed name is also used alongside the transfer speed, which does make this slightly less confusing. Well, it would be, except USB 3.2 Gen 1×2 is not officially referenced in this manner. Anything claiming to support SuperSpeed USB 10Gbps, or featuring the USB 10GBit/s logo, refers instead to USB 3.2 Gen 1×2.

For hardware manufacturers, this naming system is a dream. A hardware manufacturer selling a USB 3.0 speed device can honestly claim their device supports USB 3.2. This is technically correct, as this mode is known as USB 3.2 Gen 1×1 in the USB 3.2 specification.

In practice, for the consumer, it means you may not be getting the device you are expecting, especially if trying to ‘upgrade’ from a USB 3.0 to USB 3.2 device. Therefore, it is best to assume you are getting a USB 3.0 speed device, unless a fully qualified USB 3.2 name is specified.

USB4

The first thing to note about USB4 is the new naming style. For reasons that only make sense to the USB-IF, the ‘USB4’ moniker omits the space between ‘USB’ and ‘4’. You can probably already see how this is going. :/

USB4 is where the protocol shifts away from offering data transfer modes. USB4 is designed as a tunneling protocol – instead of defining data transfer, it instead aims to allow other data transfer modes to run across the same protocol. In its first iteration, USB4 supports USB 3.2, DisplayPort, or PCIe data transfers.

USB4 can also support Thunderbolt 3, which is a requirement for USB4 hubs.

It is still backwards compatible with all previous USB standards, but only through the USB Type C physical port.

If this isn’t confusing enough, don’t worry! There’s more confusion to come!

Due to the success of the USB 3.2 naming scheme, a similar scheme is used for USB4.

The named USB4 standards only run in dual lane mode (hence the x 2 on the end), which doubles the theoretical transfer speed, though it is technically possible for it to run in single lane mode.

Conclusion

So, hopefully this page has been useful, and helped to clear up some of the confusion surrounding modern USB standards.

I’ll attempt to keep this up to date as more nonsense is released by the USB-IF.