USB 3.2, which doubles the maximum speed of a USB connection to 20Gb/s, is likely to materialize in systems later this year. In preparation for this, the USB-IF—the industry group that together develops the various USB specifications—has announced the branding and naming that the new revision is going to use, and… It’s awful.
I won’t spoil it for you. It’s really, really bad.
Please get someone with even a little bit of product marketing experience involved in this process, please.
I think you got that backwards. Re-branding an old spec to make things more confusing for consumers is exactly the job of marketing. You gotta make the old tech still appealing, so why not fudge it a bit? Create some confusion! People will buy!
Get the marketing people out of the process.
The entirety of USB 3.x is a complete shitshow,
Relevent XKCD: https://xkcd.com/927/
This isn’t a new standard and it is backwards compatible and doesn’t use new connectors. That XKCD is completely irrelevant.
Still, is it sad that I guessed which XKCD it will be? 🙁
We definitely needed something to make USB 3/Thunderbolt 3 even more of a shit-show with respect to cables and ports.
Did a competing wireless standard (BlueTooth?) infiltrate the USB standards committee and sabotage it?
So, what is the purpose of USB C when compared to USB 3? Is it a mobile thing?
They’re functionally independent (except that USB 3.2 probably requires USB-C for the extra wire pairs).
USB 3 adds some extra signals to an existing USB connection (more specifically, it adds two extra differential signaling pairs and an dedicated ground line for the signal lines), and some extra stuff to negotiate the use of those lines. It also defined new Type A and Type B connectors of standard and micro sizes to accommodate those new data lines. In general, it’s only been a good thing, other than the rare devices that aren’t correctly backwards compatible with USB 2.0.
USB-C on the other hand defines a new connector with even more signals (it went from the 4 in classic USB or 9 in USB 3 to 24), together with a mechanism to negotiate use of the connection for non-USB signaling. It was at least partially an answer to Apple’s proprietary Lightning connector, and solves many of the same problems, which is why it’s so popular for mobile phones. It also, however, has caused a number of issues because of the extremely loose standards for what devices and cables are required to support. At minimum, you have four cable types (charge only, USB 2.0 only, USB 3.x capable, and Thunderbolt (which needs special active components in the cable for runs of more than 0.5m), and numerous operational modes (regular charging only, USB-PD only, USB 2.0, USB 3.0, USB 3.1, USB 3.2, DisplayPort 1.4, MHL (of almost any version), Thunderbolt 3, HDMI 1.4b, VirtualLink 1.0, and special debug and audio adapter modes). Any given device may support any combination of those, and most devices don’t properly label (or even document in some cases) what they actually support.
And this is why everyone needs to stop calling it “USB-C” when it’s more correctly called “Type C”.
The physical connector is separate from is transferred across it, and it’s up to the controller behind it to determine what it supports (USB 2.x, USB 3.x, DP, HDMI, Thunderbolt, etc).
We don’t say “USB-A” or “USB-B”, so why did everyone start calling it “USB-C”? It’s damned annoying and very, very, very incorrect.
Honestly, the official naming that I’d seen wasn’t _too_ horrible. What I had seen was generations listed using an XxY pattern, where the representation indicates Y channels of generation X across the link (because 3.2 is really just doing 2 gen 2 links in parallel bundled as a single link).
I’d argue that compared to the insanity that is USB Type-C, this is actually pretty tame.
The problem is that they are renaming what already exists.
First, we had USB 3.0 which was 5 Gbps.
Then came USB 3.1 which was 10 Gbps.
Now we have USB 3.2 which is 20 Gbps. Nice, easy to understand, no issues.
Except, instead of keeping it easy, they renamed existing things.
USB 3.0 –> USB 3.1 gen 1
USB 3.1 –> USB 3.1 gen 2
And they’ve done it again.
USB 3.1 gen 1 –> USB 3.2 gen 1
USB 3.1 gen 2 –> USB 3.2 gen 2
And, instead of following the pattern and using USB 3.2 gen 3, they came up with USB 3.2 gen 2×2 instead.
So, you can now have devices in your home that support USB 3.2 gen 1, USB 3.1 gen 1, and USB 3.0, which all have the exact same features, but different names.
Who actually uses the brand name? I didn’t even know they had those stupid names. I just use USB 1, 2, 3, 3.1
What is mind boggling is the connector mess. USB-C has the same connector as Thunderbolt 3 but the cross compatibility is poor when it comes to predicting what will happen if you connect one to the wrong port.
I didn’t think it would be _that_ bad.
…..It’s _that_ bad.
A much better approach would be… something like USB 480Mbs, USB 5Gbs, USB 10Gbs, USB 20Gbs.
I don’t know, perhaps they’re “too intelligent in their explanations and pedagogy” (wink wink to french politics).
Kochise,
Here here! That’s what we do for Ethernet and it’s fine.
I would note that there are physical and logical differences between USB 1/2/3 such that I wouldn’t totally drop the major version numbers.
“USB 3 5Gbps” “USB 3 20Gbps”
The USB version isn’t strictly redundant because some USB 3 devices run at USB 2 speeds. I only know this because I’ve been hit by some of these on ebay and sure enough they register as USB 3 devices but only support USB 2 speeds.
Or, even simpler: USB 3.0, USB 3.1, USB 3.2. Same as we’ve used for all previous generations of USB (1.0, 1.1, 2.0, 2.1, 3.0, 3.1).
No need to worry about the underlying speeds (we didn’t in the USB 1 or USB 2 or USB 3.0 days).
I’ve had a USB 3 issue for a couple years and since this is an article about the future of USB3 I’d like to inquire if anyone has a solution to my problem.
I have a number of HD webcams that I’d like to hookup to a server in a different room. I’ve got very long active USB3 extension cables (which work fine BTW), but it’s very impractical to hook up each camera using individual extension cables. The obvious solution of using a USB3 hub didn’t work, because it turns out that the USB3 spec only requires that USB2 devices plugged into a USB3 hub share the same 480Mbps bandwidth that a USB2 hub has. The result is that despite the 5Gbps bandwidth for USB3 hubs, USB2 devices will continue to share the same 480Mbps bandwidth and will not stack up to 5Gbps (such that I can run several USB2 cameras in parallel).
It’s essentially the same problem as this guy, although in my case each of the USB2 webcams saturates the hub’s bandwidth.
https://superuser.com/questions/590668/understanding-power-and-data-bandwidth-when-using-usb-2-0-devices-on-a-usb-3-0-c
The technical solution is known and it’s called a transaction translator.
https://en.wikipedia.org/wiki/USB_hub#Transaction_translator
The trouble is it’s only implemented by USB2 hubs to stack USB1 peripherals (yay). Needless to say, the lack of a transaction translator in USB3 is a severe limitation for anyone looking to plug in numerous USB2 devices into a USB3 hub or port. I looked at so many products but ultimately gave up on the project because I couldn’t find any hardware vendors who fixed it. If anyone knows of a way to stack USB2 devices, then I’d love to hear it!