Thursday, September 1st 2022

USB Promoter Group Announces USB4 Version 2.0 Specification: 80 Gbps Over Type-C

The USB Promoter Group today announced the pending release of the USB4 Version 2.0 specification, a major update to enable up to 80 Gbps of data performance over the USB Type-C cable and connector. The USB Type-C and USB Power Delivery (USB PD) specifications will also be updated to enable this higher level of data performance. All of these specification updates are expected to be published in advance of this year's series of USB DevDays developer events planned for November.

Protocol updates are also being made to enable higher performance USB 3.2, DisplayPort and PCI Express (PCIe) data tunneling to best use the higher available bandwidth. "Once again following USB tradition, this updated USB4 specification doubles data performance to deliver higher levels of functionality to the USB Type-C ecosystem," said Brad Saunders, USB Promoter Group Chairman. "Solutions seeing the most benefit from this speed enhancement include higher-performance displays, storage, and USB-based hubs and docks."
Key characteristics of the updated USB4 solution include:
  • Up to 80 Gbps operation, based on a new physical layer architecture, using existing 40 Gbps USB Type-C passive cables and newly-defined 80 Gbps USB Type-C active cables.
  • Updates to data and display protocols to better use the increase in available bandwidth.
    • USB data architecture updates now enable USB 3.2 data tunneling to exceed 20 Gbps.
    • Updated to align with the latest versions of the DisplayPort and PCIe specifications.
  • Backward compatibility with USB4 Version 1.0, USB 3.2, USB 2.0 and Thunderbolt 3.
USB Developer Days 2022 will include detailed technical training covering the latest updates to the USB4, USB Type-C, and USB PD specifications. Registration for the two scheduled events, November 1-2 in Seattle, WA and November 15-16 in Seoul, South Korea, will open shortly on the USB-IF website.

This update is specifically targeted to developers at this time. Branding and marketing guidelines will be updated in the future to include USB 80 Gbps both for identifying certified products and certified cables.
Add your own comment

25 Comments on USB Promoter Group Announces USB4 Version 2.0 Specification: 80 Gbps Over Type-C

#1
randomUser
Not again...

USB4 gen 1, gen2?
USB4.0, USB4.1?
or is it gonna be USB4 version 1.0, USB version 2.0

They just can't learn from their own mistakes, can they?
Posted on Reply
#2
Zaqq
And then after "USB4 Version 2.0" will logically follow "USB4.1"...
Posted on Reply
#3
Valantar
Oh for f***s sake. Are you serious, USB-PG?
Posted on Reply
#5
Nanochip
So usb4 is usurping thunderbolt ? Who would’ve thought we’d see this happen. So what controllers will be released to facilitate this? Is intel maple ridge and goshen ridge (40 gbps) now doa ?
Posted on Reply
#6
R0H1T
randomUserNot again...

USB4 gen 1, gen2?
USB4.0, USB4.1?
or is it gonna be USB4 version 1.0, USB version 2.0


They just can't learn from their own mistakes, can they?
Mistakes? You think those were mistakes :laugh:
Posted on Reply
#7
Nanochip
randomUserNot again...

USB4 gen 1, gen2?
USB4.0, USB4.1?
or is it gonna be USB4 version 1.0, USB version 2.0

They just can't learn from their own mistakes, can they?
To be fair, usb4 cables are now coming with good labeling informing the user of the supported data speed and wattage. I just picked up an Anker USB4 cable that is labeled “40 Gbps / 240W”. Which is clear.

And I can verify that the cable (when paired with the appropriate GaN USB-c charger) does indeed charge one of my devices at 140W which is 40W greater than more expensive thunderbolt4 cables. perhaps thunderbolt will die and usb4 will take off. Making the solution cheaper (but perhaps more confusing in terms of nomenclature) for all involved.

I think they should call 80 gbps usb4.1 given that it is using a new physical layer.
Posted on Reply
#8
defaultluser
so finally we can "theoretically" send pcie 4 x4 over thunderbolt? so just enough to not castrate a 6500 xt? :rolleyes:
Posted on Reply
#9
Dragokar
Once again the have no clue how to name a product in consumer friendly way, but this ist probably on purpose.

Also that most of the stuff inside usb4 is optional, not mandatory.......
Posted on Reply
#10
Nanochip
DragokarOnce again the have no clue how to name a product in consumer friendly way, but this ist probably on purpose.

Also that most of the stuff inside usb4 is optional, not mandatory.......
That might be where thunderbolt5 comes in… make the optional stuff mandatory. I wonder if intel has new controllers on the way to succeed goshen ridge and maple ridge ?
Posted on Reply
#11
Valantar
defaultluserso finally we can "theoretically" send pcie 4 x4 over thunderbolt? so just enough to not castrate a 6500 xt?
Over thunderbolt? No. Thunderbolt is still 40Gbps max. This is USB 4, which is not the same as thunderbolt, though TB3 compatibility is baked into USB4 (and they share a lot of tech). It remains to be seen if this will result in USB4-to-PCIe bridges for eGPUs, but there won't be any 80Gbps TB3 or TB4 accessories simply because those are 40Gbps standards.
Posted on Reply
#12
Tek-Check
ZaqqAnd then after "USB4 Version 2.0" will logically follow "USB4.1"...
Anyway, what exactly will be the difference? What will USB4 2.0 bring?
- DP 2.0 Alt at 80 Gbps - ticked
- tunnel USB20 and USB40(?) - ticked
- PD 3.1 up to 240W
Unknown:
- tunnel DP 2.0 at 40 Gbps without Alt mode?
- tunnel PCIe 3.0 only or 4.0 too, x2 or x4 too?
Posted on Reply
#13
Dragokar
NanochipThat might be where thunderbolt5 comes in… make the optional stuff mandatory. I wonder if intel has new controllers on the way to succeed goshen ridge and maple ridge ?
Well It could have been done this time, but they screwed it up. I do believe there will be no real TB5, since not even the M-Macs does support it properly.
Posted on Reply
#14
trsttte
I don't know what's more cursed, the USB naming scheme or a picture that looks like a USB-C to micro USB cable.

Posted on Reply
#16
Valantar
JAB CreationsSo "USB 4.02.0" is basically USB 5.0.
No no, this is USB 4.0 v 2.0. Next up will be USB 4.0 v 2.0x2, then USB 4.0 v. 3.0, which will quickly be followed up by USB 4.0 v 3.1, 3.2 and 3.2x2, each of which has only minor differences. Then we'll move on to USB 5.0, which will launch at v 3.0x3.141592, and will be the exact same protocol as USB 4.0 for some reason. Then we'll revert to USB 1.1, as by then the world will have ended.
Posted on Reply
#17
Zareek
Is that the final naming convention? Please just call it USB4.1, this crap needs to end. How the hell can they expect consumers to understand if enthusiasts barely do? USB 3 is and was a train wreck.
Posted on Reply
#18
Flanker
It is still difficult (at least for me) to know whether a motherboard usb-c port can provide both the power and signal to an external monitor. Can I blame this on USB-PG as well?
Posted on Reply
#19
btarunr
Editor & Senior Moderator
Le random gadget co: "Our product will perform as advertised on a USB 4.7 4x4 type-C with 90W PD and DisplayPort 2.9z. Performance will be lower on other standards, and use not covered under warranty."
Posted on Reply
#20
Sybaris_Caesar
NanochipTo be fair, usb4 cables are now coming with good labeling informing the user of the supported data speed and wattage. I just picked up an Anker USB4 cable that is labeled “40 Gbps / 240W”. Which is clear.

And I can verify that the cable (when paired with the appropriate GaN USB-c charger) does indeed charge one of my devices at 140W which is 40W greater than more expensive thunderbolt4 cables. perhaps thunderbolt will die and usb4 will take off. Making the solution cheaper (but perhaps more confusing in terms of nomenclature) for all involved.

I think they should call 80 gbps usb4.1 given that it is using a new physical layer.
I think that's just Anker instead of a general market trend.
Posted on Reply
#21
watzupken
The naming convention cannot get more confusing than this. Whoever that came up with the name probably did not realize that he/she have 2 numbers (USD x.x) to play with to up the version, instead of adding another "version 2" to the name.
Posted on Reply
#22
Wirko
watzupkenThe naming convention cannot get more confusing than this. Whoever that came up with the name probably did not realize that he/she have 2 numbers (USD x.x) to play with to up the version, instead of adding another "version 2" to the name.
USB4, without a space, has been trademarked (as a result of zero forward thinking). So it can't be extended with .1 or .2 or anything else because then it would't be the same trademarked symbol any more.
Think of it like this:
USB4® ... O.K.
USB4®.1 ... K.O.
Posted on Reply
#23
Valantar
WirkoUSB4, without a space, has been trademarked (as a result of zero forward thinking). So it can't be extended with .1 or .2 or anything else because then it would't be the same trademarked symbol any more.
Think of it like this:
USB4® ... O.K.
USB4®.1 ... K.O.
There's nothing stopping them from trademarking USB4.1, so ... I don't see the problem?
USB4®
USB4.1®
Works just fine to me.
Posted on Reply
#25
Herbie Dragons
OK. I have two devices. I dig through my rats nest of cables and pick one. Will it work? Physical match is obvious. It seems to work, but I don't know the transfer rate. Maybe I could get a better connection. The collection of protocols is a mirror of my box of cables -- a mess.
Posted on Reply
Add your own comment
Apr 26th, 2024 04:07 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts