USB Cheat Sheet (2022) | Hacker News Reader

USB Cheat Sheet (2022)

fabiensanglard.net

514 pointsgwerbret87 comments

Excellent article.

If I could offer one correction, it would be that SBU (as specified by the USB 3.0 Promoter Group[1]) means "Sideband Use" rather than "Secondary Bus".

On some devices, it is used to carry UART; on others, audio.

[1]: https://www.usb.org/sites/default/files/USB%20Type-C%20Spec%... (pdf)

Thank you for taking the time to report the issue. It has now been addressed.

Oh wow, thank you!

Their email address is under the Contact link in the header :)

Tangent: Author has this fabulous post I'd highly recommend: https://fabiensanglard.net/mjolnir/index.html

I read it once years ago and I come back to it every now and then wishing my current PC (10+ years and going) would gently die so I could finally build something small and tiny.

You know, accidents happen. If you were to trip over the carpet and that venerable PC falls in the dumpster.

Ah that's where your wrong, see it's called Mjolnir. Only the worthy can move it.

It's totally not because its density is probably close to lead given the concentration of parts in such a tiny space, with some rubber feat that thing ain't going nowhere.

Ha! No chance, my case comes in at a beefy 18kg (purchased circa 2008). Some of the components are slowly starting to give up the ghost. My GPU purchased in 2011-03 died a few weeks ago.

So it's happening, unfortunately I'm not paying for new RAM. So I'm planning a new rig around my existing CMK64GX4M4A2400C14 sticks.

This sounds kinda premeditated though...

Maybe walk back and forth a bit dragging your feet on said carpet and open your pc for ‘maintenance’ and accidentally fry some critical component?

I actually like the 3.2 naming. Gen is speed, "by" is width. It puts it very roughly on par with PCIe's naming which nobody complains about. I just don't like that USB 3, USB 3.1, and USB 3.2 are the same things. And that sales people don't seem to understand that saying a chip supports 3.1 or 3.2 tells me it's anywhere from 5-20gbps which isn't ideal.

PCI-E has had the same standard since its inception: 1.0, 2.0, 3.0, etc. USB has changed multiple times and has remained confusing for the vast majority of people. What was 3.0 is now not 3.0. Even 3.1 has changed. There is no reason to use this naming convention they currently have but for some reason they stick with it..

The problem with USB, is the revisions are all encompasing. So USB 2.0 includes the 1.5 Mbps and 12.0 Mbps modes from usb 1.x as well as the new 480 Mbps mode, so you could have a USB 2.0 device that only did 12 Mbps (high speed!). It might actually be your old usb 1.1 device with a new label.

PCI-e requires hosts and devices to be backwards compatible, but the interface speed is a required part of the spec. Nobody makes a PCI-E 2.0 device that only works with PCI-E 1.0 encodings/speeds, or anyway, it wouldn't be acceptable.

PCIe also had things like "1.1", "2.1" and "3.1" - that fixed issues and added functionality - but there wasn't the same crossover between "feature sets and spec revisions" and "speeds" we see in USB today.

Manufacturers of mainstream consumer motherboards never used 1.1, 2.1, etc. for PCI-E though. What is 4.0 on the spec sheet will be 4.0 to the buyer. My old 2016 motherboard has a slew of 3.0 labelled USB ports that are now not 3.0, hence the conundrum. It just doesn't make sense why they changed established naming conventions. Is this something that causes me sleepless nights? Not in the least. But it's still an annoyance for consumers and even advanced users as detailed in that latest Geerling video et al.

1.1 was very much commonly used in consumer marketing, to the level where there's many instances today of people referring to pcie1.x speeds as "1.1". And I'm pretty sure I've seen 2.1 in consumer marketing contexts. But you're right I didn't know 3.1 existed until I looked it up :p

But USB 3.0 is pretty much the only "speed" that hasn't changed - it always required the extra connectors for 5Gbps from the start - but no more. What about those ports is now not "3.0"?

Possibly they stick with it because it's usable (ish) and it was driving everyone up the wall when they'd change it?

And not only the sales people. Windows doesn't report anywhere what your motherboard is capable of, and even if you connect with a device it will not tell you the speed it agreed on.

Out of curiosity, what do you use the higher 20gbps transfer speeds for? Video production?

I use USB-C displays, but they run in DP Alt mode. I don't have many (any?) storage devices that can max out a 20gbps connection, and usually don't exceed 5gbps

This goes back to another point I've historically made which is that except for storage devices, pretty much nothing supports those speeds. I think there are some USB adapters that don't use alt mode and that can have some advantages on some hosts but usually that's a disadvantage.

USB interface chips are, as far as I've seen, a Cypress/Infineon FX3 or a bit more rare FTDI FT600/FT601. I even talked with the FTDI guys at s conference and they said nobody's asking for higher than 5gbps. Infineon just recently, after I think 10+ years, came out with 10 and 20gbps chips. But only for receive. Seems to be for cameras mainly. So surprisingly yes, video production.

But I want it for other reasons professionally. For example, if you look at the signalhound (which uses the fx3) series of products, they often cap out at 40 Msamples/sec for USB. This is a classic 5gbps limit. To compete with the big boys they need 250 MHz if not more. That's 8 gbps before protocol overhead. It doesn't help that USB is extremely dependent on host compute capability to keep throughput up but assuming your PC is up to the task, 20 gbps could interface some serious data to the real world.

Besides storage devices, i.e. external SSDs, which are very frequently used and they need a USB port as fast as possible, the other frequent application that needs the fastest USB ports is the use of USB Ethernet interfaces.

Also eGPU. I have a tiny NUC-size system with decent internal GPU and a (physically much larger) game system with a slower CPU that idles at only a bit under twice the maximum power of the NUC. It would be handy to be able to just plug in an eGPU when needed. The power and cooling requirements of fancy GPUs are so much higher than that of CPUs that large cases designed around the CPU don't make much sense. Even the pysical stability of a large GPU in an ATX style case is not ideal.

> Out of curiosity, what do you use the higher 20gbps transfer speeds for?

Images, videos, movies, file transfer/backup. 50 megapixel RAW images from a DSLR that can capture up to 20 images a second get big. My daughter is a much better volleyball player than I am sports photographer, so I have to spray-and-pray to capture those high-speed hits at the net.

Transferring a few hundred such photos via a card reader was so glacially slow it was worth adding a 20Gbps USB3.2 2x2 port to my home server (Ryzen 5600x) via a dedicated PCI-E card. The USB3 ports on a good enthusiast-class mobo for that generation (only 4 yrs ago) max out at 5Gbps (theoretical). I would have added a 40Gbps Thunderbolt port instead but then I'd have to take a hit on the top speed of my second NVMe drive due to sharing PCI-E lanes.

While the increasing deployment of true USB4 ports is wonderful, it's not quite a panacea. Just because a port is labeled USB4 doesn't mean you necessarily get 40Gbps performance. USB bandwidth is shared across multiple ports via internal hubs and then the PCI-E lanes the hub is connected to might be shared with other peripherals (GFX, NVME, I/O cards). And different USB ports have different trade-offs depending on how they're internally connected, which isn't always documented well by mobo makers.

Sadly, in consumer systems the lack of PCI-E bandwidth can still be an issue if you want your expensive GPU to run maximally fast and have multiple fast NVMe drives. You have to spec your system carefully, get the latest generation hardware or pay 3-4x for HEDT/Enterprise chipset motherboards. Getting even 10s or 100s of gigs of data in or out of a PC reasonably quickly and conveniently has always been a bottleneck that's only getting somewhat better quite recently.

Any external NVMe SSD from the last 7-8 years easily saturates a 20 Gbps connection, because already from that time the NVMe SSDs were able to saturate a 32 Gb/s PCIe 3.0 4-lane connection.

For at least 7-8 years I have been using USB external enclosures for M.2 Key-M NVMe SSDs, which always saturated whatever kind of USB port they were connected to, i.e. 5/10/20 Gb/s.

I do not remember when I have last used a SATA SSD, which is slower than 10 and 20 Gb/s USB, but I think that this was about a decade ago.

I don't know what short-distance data communications will be like in 2050, but we know it will be called USB.

USB-G 4.6 SuperSpeed Plus, but the cables will still just be used for charging your random electronics and won't even work for that half the time.

…because Staples has replaced Amazon and Temu, which both went bankrupt in 2042, and all usb-c cables are made in Papua New Guinea with only 1/4 of the pins connected to wires. Some things never change.

I wouldn’t be too surprised if they rebrand to AI Bus.

Wait, really? Where can I invest?

I know not with what technology 2030 will use, but 2040 will use USB sticks and stones.

Well, obviously. What do you think the U stands for? When we make contact with aliens, we’ll find them using it too.

I once heard that the USB naming is misleading by design so that vendors could still sell older generations accessories they had in stock. The USB-IF just rebrands the old ones to make them sound current.

Imagine the following naming:

  USB 3.0 / USB 3.1 Gen 1 / USB 3.2 Gen 1 -> USB 3 5Gbps
  USB 3.1 / USB 3.1 Gen 2 / USB 3.2 Gen 2 -> USB 3 10Gbps
  USB 3.2 Gen 2x2 -> USB 3 20Gbps

Isn't that much clearer? I think USB 4 is finally going to the right direction.

> I think USB 4 is finally going to the right direction.

USB 4 is actually going into an even worse direction. USB 4 = Thunderbolt 4, except everything is optional. e.g. USB 4 might not even support DP Alt mode. Thunderbolt 4 always will.

Even backwards compatibility is optional in USB4. There are USB4 devices (SSDs at least) that will not function when connected to USB 3 ports.

That’s not backwards compatibility

That is exactly how the USB IF has been branding it for consumer use. They explicitly tell[0] implementers to not call it "USB 3.2 Gen2x2", but "USB 20Gbps".

The problem is just that the manufacturers and the tech press keep ignoring it...

[0]: https://www.usb.org/sites/default/files/usb_performance_logo...

Which is why I honestly believe they should have fixed this in the design stage itself. Post-facto reframing/renaming never seems to go well.

Especially once the mass produced cheap stuff starts being churned out, and there's no cost incentive to go back and fix wrong messaging. USB-IF constantly drops the ball around this ngl, feels like they're a pure scientific community that doesn't think about consumer adoption and UX.

I have a USB hub that I bought recently, that has very nice markings on it that are almost like you say :)

I connects via USB4 to the host, and has the following markings on its ports:

- Power in/USB 10Gbps

- USB 10Gbps

- 8K HDMI

Pretty happy with this one so far.

I got a jCreate5 hub at clearance from an Office Max (rip) and the ports are labeled just like this, no futzing on which port is the PD

Maybe we even have the same one :)

Mine's this one:

j5create USB4 8K Slim Hub - 8K60/4K144 HDMI, 1 x USB-C 10Gbps with PD charging, 1 x USB-C 10Gbps, 2 x USB-A 10Gbps | Compatible with MacBook, Windows Laptops (JCH453)

https://www.amazon.co.uk/dp/B0C8NDXH8S

I bought it because unlike many other USB hubs, the host connection is USB4 / Thunderbolt 4, instead of just USB3 for the host connection.

> Ultimate Connectivity: JCH453 combines the latest USB4 controller with Thunderbolt 4 and USB compatibility, ensuring seamless connectivity with various devices. Experience the power of multiple ports in one hub.

Manufacturer product page:

https://info.j5create.com/products/jch453

> A USB4® multi-port hub incorporates the latest USB4® controller offering compatibility with Thunderbolt™ 4 and older USB™ specifications.

> With up to 40Gbps of throughput, dynamic data, and display bandwidth allocation for efficient display data flow, you can easily create a high-definition monitor setup.

I think this practice is rather blatantly what you say. The same thing with HDMI forum folding HDMI 2.0 into HDMI 2.1. They made the new 2.1 features optional, therefore manufacturers were able to call their 2.0 devices 2.1 without actually supporting the 2.1 features. AMD has been recently doing similar things, releasing “new” generation of mobile processors where half of them are just rebrands of the older generation.

Or it could be: 5 Gbps --> USB 3 10 Gbps --> USB 3.1 20 Gbps --> USB 3.2

Higher number = better

Good sheet. Worth adding:

- Female vs male crossover naming and pinouts for Type-C connectors

- Actual voltage, modulation and signaling schemes (USB4v2 uses PAM3 11b/7t encoding)

- PD generations and profiles

... and the bunch of proprietary voltage schemes like Quickcharge.

Thanks to the EU those are now forbidden, all phones and laptops should be compatible with USB-PD.

Update: USB-PD is a requirement, but manufacturers are allowed to have their own proprietary charging solution.

The simplicity of Thunderbolt. Versions 1 and 2 used mini DisplayPort, 3 and upwards USB-C. Version 1 was 10Gbps, 2 was 20Gbps, 3 was 40Gbps, 4 was 40Gbps, 5 is 80 or 120Gbps with boosting.

A Thunderbolt 5 cable will always support 80Gbps, DisplayPort 2.1, PCIe, USB4 and power of up to 240 watt.

> and power of up to 240 watt

Except active optical cables. None exist yet that I'm aware of though.

I'd guess that most people who use optical Thunderbolt cables are aware that they do not carry power.

I still don't understand why MacBooks support USB4/Thunderbolt 4/5, but NOT USB 3.2 Gen 2x2. So you can get 20-40Gb/s speeds with more expensive external disks, but only 10Gb/s with the cheaper, more commonly available ones that advertise 20Gb/s.

I believe it’s that MacBooks support Thunderbolt primarily and USB only where absolutely necessary beyond what’s coded into one of the TB specs; and I assume TB doesn’t define 3.2x2x2 as part of any TB spec <=5?

This article is why I replaced all the usb dock cables in the office to make sure the usb cable connected to the laptops was transferring enough power so the laptop wouldn't silently lower its frequency for the lower power draw. 10-30% speed bump just because.

I’ve been a tech guy for 45 years and I still can’t figure out USB and Thunderbolt and what goes with what and how fast it’s supposed to run.

If you buy Thunderbolt 5 cables: every USB standard is compatible and then some.

It wasn't until last year that I finally purchased my first USB-C device/cables – and after years of solid DisplayPort and Thunderbolt2 connections I absolutely hate USB-C (it's too delicate, physically).

Not until 2023 did I even have a computer newer than 2012, so I missed almost all of USB3's hayday — including nomenclature disputes — but the speeds sure are an improvement!

While USB Type C can be broken much more easily by brute force and it is more prone to accidental disconnects than Type A, the Type C connectors are guaranteed to survive much more cycles of plugging/unplugging than Type A connectors.

Type A connectors are typically guaranteed only for around 1000 cycles, with some better connectors rated up to 1500 cycles and some worse connectors rated only for a few hundred cycles.

If you have a device with a Type A connector that you plug and unplug at least once per day, there is a non-negligible risk that the connector will become defective before other components of the device.

On the other hand Type C connectors are guaranteed for at least ten thousand mating cycles, with the best guaranteed for at least twenty thousand cycles, so you should not be able to wear them out through normal usage.

It is true however that you must handle Type C connectors much more delicately than Type A, otherwise you can break them before they are worn out by mating cycles.

During the last few years, high-endurance Type A connectors have also appeared, which can survive a limit between 5 thousand and 20 thousand mating cycles, matching Type C connectors, but most equipment with Type A connectors does not use such more expensive connectors.

Thanks for the details on mating cycles – although I've never had a TypeA connector "fail" from mating cycles (and have some in daily use for decades). I've only used USB-C for about a year, but have already broken one (which completely disables the plug, unlike the sculptable TypeA).

>It is true however that you must handle Type C connectors much more delicately than Type A, otherwise you can break them before they are worn out by mating cycles.

I would suspect that on a large enough datasample, TypeA connectors will out-survive TypeC (durability-wise), for your above reasoning alone. Have you ever worked hardware techsupport in an academic environment (or have children, or wives, or husbands)?

----

As an electrician with tons of realworld experience resolving burned-up installations, I also doubt the 240W-rating™ across top-end USB-C connectors is safe (I know theoretically it is... just like all those burnt-up outlets I've replaced in the real world). If I breath on my main display's USB-C connector (<1 year old!), it often re-sync's (a few seconds of annoyance).

Obviously USB-A could never approach these power ratings, but I suspect USB-C cannot either (in realworld == electric fires). I love & use PoE (via Cat5e/6): it has much lesser-rated ampacity (despite higher cross-sectional area of wire).

Related. Others?

USB Cheat Sheet - https://news.ycombinator.com/item?id=31271038 - May 2022 (168 comments)

> SBU1 and SBU2 are secondary bus wires, for the DisplayPort AUX channel and hot plug detection (HPD).

Correction - HPD signal is translated into vendor message and carried over CC lines - same ones that are used for PD and AltMode negotiation.

In DP-Alt mode SBU1/2 basically becomes AUX+/-.

I just wish product listings were clear and actually followed the specs.

IMHO USB 3.0 was the last sanely-named version. Then again, if you're familiar with Ethernet, the proliferation of variants isn't unexpected.

Where does TB5 come into all of this?

- Thunderbolt 3 is a superset of USB 3.1

- USB4 is built on Thunderbolt 3's protocol, implementing a subset of its mandatory features

- Thunderbolt 4 is a strict profile of USB4 (all optional features made mandatory)

- USB4 v2 introduced 80 Gbps signaling

- Thunderbolt 5 is a strict profile of USB4 v2 (again, optional features made mandatory)

I don't see why it would. Thunderbolt is not a USB standard

> Thunderbolt is not a USB standard

Concerning Thunderbolt 3: USB4 is based on the Thunderbolt 3 protocol [1].

Concerning Thunderbolt 4: "In July 2020 Intel announced Thunderbolt 4 as an implementation of USB4 40 Gbit/s with additional requirements, such as mandatory backward compatibility to Thunderbolt 3 and requirement for smaller notebooks to support being charged over Thunderbolt 4 ports.[14] Publications such as AnandTech described Thunderbolt 4 as "superset of TB3 and USB4" and "able to accept TB4, TB3, USB4, and USB 3/2/1 connections"." [2]

Concerning Thunderbolt 5: Intel considers Thunderbolt 5 as an implementation of USB4 Version 2.0. [3]

[1] https://en.wikipedia.org/w/index.php?title=USB4&oldid=134742...

[2] https://en.wikipedia.org/w/index.php?title=USB4&oldid=134742...

[3] https://en.wikipedia.org/w/index.php?title=USB4&oldid=134742...

Thunderbolt 5 and USB4v2 are the same thing now. They both support 80gbps and pcie pass through.

> Thunderbolt 5 and USB4v2 are the same thing now. They both support 80gbps and pcie pass through.

Not completely true: Thunderbolt 5 demands some capabilities that are optional for USB4v2.

From a protocol/bandwidth level, it’s essentially the same though. Thunderbolt 5 has some more guarantees for power and display, but the data rate of the two is the same.

Doesn't it run over a USB-C shaped wire? If you're trying to understand things that plug into USB-shaped ports it seems at least worth mentioning.

To be fair: You should refer to these as Type-C cables, as they carry things that are not USB protocol.

The sole exception should be made for "charge only" cables, which can, and should, be referred to as "wired for USB 2.0". These cables "shouldn't" exist, but I also don't want to buy a $30 cable just to charge my phone.

Thunderbolt 5 is basically just PCI Express, power delivery, and DisplayPort over the same cable, which for reasons passing understanding is terminated with a USB-C connector.

I think most of those cables will also support USB the protocol.

Why do we constantly change this?

What technological advance was not available x years ago to dream up usb 4?

We already know we will use the bandwith, why not dream up what will be the usb 8 spec in 20 years now and have everything working without change for 20 years?

The wires count is suspect. USB 1.0-2.0 only use two wires for data (the other two are ground and power). USB 3.0 uses 4 for data (plus extra shield, 2 for USB 2.0 and 2 for power). I don't know well enough the others.

The wires count seems to be the number of conductors in the cable (i.e. the number of wires you'll find if you cut a cable in half, including ground and power).

It's true that the actual data is sent over a lower number of diffpairs.

I suspect the shield is not included in the number of wires, since all USB cables have a shield (not sure if usb 3.0 has an extra return ground wire for high speed).

It still would't be right. Full-featured USB-C has 8x superspeed (tx1p, tx1n, rx1p, rx1n, tx2p, tx2n, rx2p, rx2n), 2x high-speed (dp, dn), 2x power (vbus, gnd), 2x SBU, 1x CC. That's 15 wires.

A full-featured USB-C connector has 24 pins, as shown in a diagram in the parent article.

The "12-wire" count of the parent article refers only to the main wires, i.e. the 4 USB 2.0 wires + 4 differential pairs for USB 3 or 4.

Similarly, the "8-wire" count for Type A connectors refers only to the main wires, i.e. 4 USB 2.0 wires + 2 differential pairs for USB 3.

Yes, but not all of those 24 pins are used, and some of them are duplicated.

I also don't think you can make this "main wire" distinction: you shouldn't be counting the power wires if you only care about USB data in the Type A case, and you should arguably be counting the CC wire in the Type C case if you want to go for "strictly necessary to function", or the SBU wires if you want to go for "carries data at nontrivial speeds".

USB 3.11 for Workgroups.

this is great, does usb cable has some limitation? can i take any length? i saw you measure 4m any advices longer cable?

I've tried longer cables on the slower USB's. My experience is that it can be touch and go depending on the peripheral. As I understand it, cable length is limited by the round-trip time between the host and the peripheral.

Where I've had good and bad "luck" with USB cabling is video cameras, which require a fairly high bandwidth data stream.

I'm also wondering what the cable length figures in the cheat mean exactly.

This is generally good but it’s missing low speed (1.5 megabits/second), which is also under USB 1.1.

Great way of identifying the difference in types of USB

I'd love for someone who's part of the USB-IF to try and explain what the heck they were thinking with their naming conventions. They're indefensibly awful in every way.

nice work, thanks

Reply on news.ycombinator.com