NHacker Next
- new
- past
- show
- ask
- show
- jobs
- submit
login
▲USB Cheat Sheet (2022) (fabiensanglard.net)self.__VINEXT_RSC_CHUNKS__=self.__VINEXT_RSC_CHUNKS__||[];self.__VINEXT_RSC_CHUNKS__.push("2:I[\"aadde9aaef29\",[],\"default\",1]\n3:I[\"6e873226e03b\",[],\"Children\",1]\n5:I[\"bc2946a341c8\",[],\"LayoutSegmentProvider\",1]\n6:I[\"6e873226e03b\",[],\"Slot\",1]\n7:I[\"3506b3d116f7\",[],\"ErrorBoundary\",1]\n8:I[\"a9bbde40cf2d\",[],\"default\",1]\n9:I[\"3506b3d116f7\",[],\"NotFoundBoundary\",1]\na:\"$Sreact.suspense\"\n:HL[\"/assets/index-BLEkI_5r.css\",\"style\"]\n")an>
Rendered at 03:56:00 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
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)
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.
So it's happening, unfortunately I'm not paying for new RAM. So I'm planning a new rig around my existing CMK64GX4M4A2400C14 sticks.
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.
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.
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"?
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
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.
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.
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.
Imagine the following naming:
Isn't that much clearer? I think USB 4 is finally going to the right direction.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...
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.
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.
I connects via USB4 to the host, and has the following markings on its ports:
- Power in/USB 10Gbps
- USB 10Gbps
- USB 10Gbps
- 8K HDMI
Pretty happy with this one so far.
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.
Higher number = better
- 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
Update: USB-PD is a requirement, but manufacturers are allowed to have their own proprietary charging solution.
A Thunderbolt 5 cable will always support 80Gbps, DisplayPort 2.1, PCIe, USB4 and power of up to 240 watt.
Except active optical cables. None exist yet that I'm aware of though.
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!
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.
>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).
USB Cheat Sheet - https://news.ycombinator.com/item?id=31271038 - May 2022 (168 comments)
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+/-.
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).
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.
- 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)
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...
Not completely true: Thunderbolt 5 demands some capabilities that are optional for USB4v2.
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.
I think most of those cables will also support USB the protocol.
Where I've had good and bad "luck" with USB cabling is video cameras, which require a fairly high bandwidth data stream.
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?