USB4 Version 2.0 to Offer 80Gb/s Transfer Speeds

[profcutter]

Great. Now get new routers to use these instead of those ancient ethernet cables.

Umm. Why? Ethernet cables are super flexible, they have locking connectors, they’re much cheaper at least up to CAT 7 than usb cables, and you can run them 300 meters. And you can make them at home. Ever tried to make a USB C cable, especially an active one? Why would anyone want to swap for a teeny, delicate, non-locking USB C connector for routing?

 

[dave070]

Which would be great if USB ONLY transferred data. None of those designations would help if a person’s using USB for power delivery.


Yes, but, most importantly, it’s optional AND requires certification. A company making a cheapy cable that they want on the market next week? NO label required!

Couldn't that be handled with USB4 80 (PD240)? USB4 80 would be without PD.

People would have to learn what the numbers meant but a simple way to get message across.

 

[jaytv111]

Apple it's not the only company in the world that makes cables (including TB) 🤦‍♂️
u can get a TB3 1,2m cable on aliexpress for 15$, that's less than a genuine lightning cable. and no, it doesnt explode, no, it doesnt break (at least not more than apple cable does), and no, it doesnt release toxic substances. But, yes, it doesnt fill Cook's wallet

I meant to say that you lose a feature with those Thunderbolt cables. If you get a 2m active TB cable with 40 Gbps, you get no Superspeed USB or DP capability, you only get Thunderbolt or USB 2.0 support. If you get a 1m passive cable, you do get Superspeed USB and DP and 40 Gbps, but it’s only 1m or less. You can get a passive 2m cable with SS USB and DP, but only with 20 Gbps Thunderbolt at 2m lengths. Only Apple’s cable seems to have all the features (USB Superspeed, DP, 100 watts of PD, and Thunderbolt 40 Gbps), while being 2m (Apple might be the only seller of 3m cables as well, except for optical Thunderbolt but those are even more expensive and niche and don’t deliver any power or alternate protocols).

The problem with Apple’s is of course the extreme cost to their cables.

 

[ArkSingularity]

I meant to say that you lose a feature with those Thunderbolt cables. If you get a 2m active TB cable with 40 Gbps, you get no Superspeed USB or DP capability, you only get Thunderbolt or USB 2.0 support. If you get a 1m passive cable, you do get Superspeed USB and DP and 40 Gbps, but it’s only 1m or less. You can get a passive 2m cable with SS USB and DP, but only with 20 Gbps Thunderbolt at 2m lengths. Only Apple’s cable seems to have all the features (USB Superspeed, DP, 100 watts of PD, and Thunderbolt 40 Gbps), while being 2m (Apple might be the only seller of 3m cables as well, except for optical Thunderbolt but those are even more expensive and niche and don’t deliver any power or alternate protocols).

The problem with Apple’s is of course the extreme cost to their cables.

The USB-C cable mess has somehow managed to be even more convoluted than I imagined possible. This is just downright terrible.

I don't necessarily mind the fact that not all cables support all standards (USB-C is used for charging cables too, and nobody wants to spend $40 on a charging cable). But if someone spends money on an active fully thunderbolt-compatible cable, it should at least support better USB standards than USB 2.0. There isn't really a viable solution for the consumer that doesn't make this just a plain mess.

 

[Damian83]

I meant to say that you lose a feature with those Thunderbolt cables. If you get a 2m active TB cable with 40 Gbps, you get no Superspeed USB or DP capability, you only get Thunderbolt or USB 2.0 support. If you get a 1m passive cable, you do get Superspeed USB and DP and 40 Gbps, but it’s only 1m or less. You can get a passive 2m cable with SS USB and DP, but only with 20 Gbps Thunderbolt at 2m lengths. Only Apple’s cable seems to have all the features (USB Superspeed, DP, 100 watts of PD, and Thunderbolt 40 Gbps), while being 2m (Apple might be the only seller of 3m cables as well, except for optical Thunderbolt but those are even more expensive and niche and don’t deliver any power or alternate protocols).

The problem with Apple’s is of course the extreme cost to their cables.

I was talking of standard cables (1m). If u have special needs maybe u even need a 1km long cable, but it doesnt make it a reference point. Most phone cables are 1m so that’s the reference.

 

[deconstruct60]

I meant to say that you lose a feature with those Thunderbolt cables. If you get a 2m active TB cable with 40 Gbps, you get no Superspeed USB or DP capability, you only get Thunderbolt or USB 2.0 support. If you get a 1m passive cable, you do get Superspeed USB and DP and 40 Gbps, but it’s only 1m or less. You can get a passive 2m cable with SS USB and DP, but only with 20 Gbps Thunderbolt at 2m lengths.

That is older Thunderbolt 3 cables. Thunderbolt 4 cleaned that up.


"... but with Thunderbolt 4 it’s not as much of a consideration since all TB4 cables are backwards compatible with Thunderbolt 3, USB4, USB 3.0, USB 2.0 etc—unlike Thunderbolt 3 active cables that were backwards compatible with only USB 2.0 (at a rather pathetic 480MBps).... "

Best Thunderbolt 4 and USB4 cables

Thunderbolt 4 is the new connection standard in town and you need Intel-certified cables to take advantage of all its benefits.

www.techadvisor.com

OWC TBv4 2m cable does all the wide spread USB protocols. ( USB 3 gen 2 is 10Gb/s ) https://eshop.macsales.com/item/OWC/CBLTB4C2.0M/

Similar .. Caldigit

"... Now Thunderbolt 4 cables offer compatibility across all versions of USB-C including USB 2.0, 3.0, 3.2 and 4 regardless of the length of the cable. ..."

CalDigit | Thunderbolt 4 | USB 4 | Cable | 40Gb/s – CalDigit

www.caldigit.com

Length of cable is effectively means it covers both passive and active. ( active cables are necessary for the longer length. ). TBv4 stretched the maximum allowed passive length up to 2m (via better cable construction) , but safer to cover 2m (and up) with active. Thunderbolt 1-3 had shorter cut offs for passive max length.


There is no reason why USB 3 gen 2 can't travel over an active cable that has been developed and tested against the protocol. As long as the cable attach 'handshake' goes well and all the parties involve identify their limits , then it should just work. ( TBv3 cables built before USB 3 gen 2 + existed have an issue with 'talking to' something better than plain old USB 3.0 . That wasn't going to be permanent. ).

Only Apple’s cable seems to have all the features (USB Superspeed, DP, 100 watts of PD, and Thunderbolt 40 Gbps), while being 2m (Apple might be the only seller of 3m cables as well, except for optical Thunderbolt but those are even more expensive and niche and don’t deliver any power or alternate protocols).

Apple isn't the only one. But if looking for the discount super special Thunderbolt cable then features do tend to disappear to hit lower costs. There are "higher than USB cable" priced cables that do as much as Apple does , but leave off the super high Apple tax mark up. ( Apple tends to try to make their high cable costs 'disappear' but wrapping them up in a Mac system purchase order. $2000 for the Mac .. what's another $100 for a cable. )



It won't be too surprising if USB 4 version 2 introduces a new (from USB4/TBv4) split though. That the "passive max" length shrinks back off the 2m mark for 80Gb/s . Probably, the "Up to" means falling back to the 0.8 cables for maximum speed and sag back to 40Gb/s as get back out to 2m. If stick to the original TBv1-2 passive range than get the max speeds (with better transmitter/receiver in the ports over the relatively short wire. ).

That would get USB-IF a substantively large number of older cables that 'worked'. (most folks are buying more affordable , shorter cables). But the cable vendors that stretch the limit ( build 2m passive "max speed" ) cables probably don't cut it the new threshold.

 

[Unregistered 4U]

Couldn't that be handled with USB4 80 (PD240)? USB4 80 would be without PD.

People would have to learn what the numbers meant but a simple way to get message across.

The only way they could hope to fix the current USB-C connector situation is to mandate labeling. As long as the naming is “optional”, then those that benefit from the confusion won’t use labeling.

 

[theorist9]

Which would be great if USB ONLY transferred data. None of those designations would help if a person’s using USB for power delivery.

Of course it would. Suppose you're buying a device that's currently labeled USB 3.2 Gen 1. If it were instead labeled USB5, that would provide a great improvement in clarity, irrespective of the power delivery.

And power delivery is qualitatively different from speed. There are currently a finite number of USB speed specs (5/10/20/40 Gbps). But with power delivery, it's messier to add it to the spec, since an infinite number are allowed--devices can deliver 66W, 87W, 123W, etc. Still, I suppose the manufacturer could further add power delivery info with USB5-66W, USB5-87W, USB5-123W, etc.

Yes, an unscrupulous manufacturer could decline official certification and label a device or cable USB10 when it's only capable of USB5, but they can do that now with the current system, by labeling it USB 3.2 Gen 2 when it's only capable of USB 3.2 Gen 1. Except with my system their lies would be more obvious, because it would be clearer to the consumer what the device or cable is supposed to do.

 

[ponzicoinbro]

The only way they could hope to fix the current USB-C connector situation is to mandate labeling. As long as the naming is “optional”, then those that benefit from the confusion won’t use labeling.

Or color coding like USB 2 and 3 connectors but it gets ugly sometimes.

 

[sideshowuniqueuser]

Shouldn't USB4 version 2, just be USB5. Unrelated I'm about to turn 35 version 4.

So you're around 47 or 48 then? I think we're running on a similar numbering system, it works well

 

[ivan86]

What if the data is inverted? Will it then show up tomorrow, but you won't see it until yesterday?

In such case you see it pop up right before “pressing send”.

 

[spicynujac]

The USB 4.x rev 1.1 version A nonsense needs to go, as does the anachronistic measuring of speed in bits.
80 Gb/s = 10 GB / s
We all measure data in gigabytes, and have for years.
Have 100 GB of photos ? They will transfer in 10 seconds at 10 GB/s. In bits it's just stupid and confusing.

It's something copied from the ISP marketing departments where they make their speeds sound faster than they really are by selling you in bits, which are 1/8 of a byte. 100 Mb for only $79.99 sounds a much better deal than 12 MB.

 

[Unregistered 4U]

If it were instead labeled USB5, that would provide a great improvement in clarity, irrespective of the power delivery.

I’m not saying it wouldn’t. Just saying that having speed listed on a cable won’t help someone that wants to know if this cable will charge their laptop at 100W or 20W.

And power delivery is qualitatively different from speed. There are currently a finite number of USB speed specs (5/10/20/40 Gbps). But with power delivery, it's messier to add it to the spec, since an infinite number are allowed--devices can deliver 66W, 87W, 123W, etc. Still, I suppose the manufacturer could further add power delivery info with USB5-66W, USB5-87W, USB5-123W, etc.

Yes, but as long as the vendors aren’t required to show the info, if it’s always “optional”, then there’s no labeling system that will solve any of the problems as the companies just won’t use it.

 

[Unregistered 4U]

Or color coding like USB 2 and 3 connectors but it gets ugly sometimes.

Whether it’s coloring/labeling it’s the same problem as long as it’s optional. Companies that want to use it, will use it. Companies that would rather not use it, won’t. Consumers looking to save money on cheap cables will find themselves looking at a load of unlabeled cables claiming “USB 4 compatible”.

 

[deconstruct60]

Or color coding like USB 2 and 3 connectors but it gets ugly sometimes.

USB Type A sockets are much bigger with a relatively large plastic "nub" to color that forces you to stick the cable in the 'right'/correct way. The two major points of Type-C are

1. No orientation to the cable. No up/down, the cable goes in either orientation.

2. much smaller. (suppose to be standard connector for devices like phones. So closer to the Type-B or micro B like connector).

There is a relatively much thinner , metal bar across the middle of the Type-C socket. If want to 'paint' the outer edge of the bar with a color then, not only color perception issues, lots less color to see issues. If want to paint around the outside of the socket (e.g., old school microphone jack colors on PCs) then, similar issue as the outside lip is thinner/smaller also.


Somewhat similar slippery slope with these every more USB-IF complicated logos . If the sockets are smaller and spaced closer together then there is less space to print the logo. If try to squeeze it in there anyway then how many folks can actually read it while holding the device at a normal distance.

 

[gaximus]

So you're around 47 or 48 then? I think we're running on a similar numbering system, it works well

I think we are on complete different numbering systems. 47 would 35 version 12

 

[deconstruct60]

PAM-3 has more confirmation than a fuzzy, deleted Intel twitter photo.

"... Also in April, Keysight released their PathWave Advanced Design System (ADS) 2022 Update 2.0 Software, where their website lists USB4v2 (PAM3) AMI Model Builder, and their W3081E PathWave USB AMI Modeler page “Builds USB4V2 AMI models for Tx and Rx“, “for USB with next generation PAM3 modulation“. ... "

USB4 v2

Has a trick up its sleeve

www.angstronomics.com

USB 4 v2 isn't coming any time soon though. But at least folks can test prototypes in the labs. Even if Intel gets it weaved into Gen 14 (Meteor Lake) if that is only the laptop version which may not roll out first. Could be well into 2024 when see any traction here. AMD trailing even further behind.

I don't see the discrete controllers coming on a super fast pace. Like previous evolutionary iterations if Intel releases discrete chips too far in advance of everyone else there is a large rukus. Apple isn't going to do discrete chips and push the envelope. The pragmatic start is going to be when the SoCs start rolling out with a incremental delay on discrete chips so they can talk to anything else. The USB4 v2 router/hubs are going to be a larger pain to do at the same price points.



This 120Gb/s mode is pretty likely a narrow corner case mode when doing Alt mode passthrough DisplayPort v2 with one transmit and receive group pair. So fall back to just one trans/receive pair for USB (Or Thunderbolt) bidirection. The video data is going out at higher bandwidth but may not be a "DP transported over thunderbolt protocol" output. For example, Direct connect DPv2 port to 8K display that can still up/down data from the other ports on the display at very good rates ( e.g., a USB hub in the box + high res webcam + etc. ).

That's probably why USB-IF likely isn't talking up that mode. It is likely largely for non USB traffic. Not really USB-IF's job to hype someone else's standard.

 

[theorist9]

We'll probably see DP Alt Mode 2.0 (aka DP UHBR = Display Port Ultra-high Bit Rate), which also gives 80 Gbps (though outgoing only) well before this, since that standard has been around for a while, and the first source and sink (host and device) controllers for it were finally certified by VESA in May 2022, in collaboration with AMD:

First VESA DisplayPort 2.0 Video Source and Sink Devices Complete DisplayPort UHBR (Ultra-High Bit Rate) Certification - VESA - Interface Standards for The Display Industry

Qualified VESA DisplayPort Authorized Test Centers Ready for End-Product UHBR Certification BEAVERTON, Ore. – May 9, 2022 – The Video Electronics Standards Association (VESA®) today announced that the first video source and display devices supporting DisplayPort UHBR (Ultra-high Bit Rate) – the...

vesa.org

DP UHBR, which is compatible with TB4, gives 80 Gbps outgoing (perfect for high-res/high refresh displays) by converting a duplex 40 Gbps incoming/outgoing pair into a single outgoing path (well, mostly outgoing; there must still be a provision for the device to send status info. back to the host).

If Apple's revision of their ProDisplay XDR is 7k or 8k@120 Hz, we could see Apple introduce this as well.

 

[joecomo]

I vote for USB 8 / Thunderbolt 8 to match the bandwidth speed

for sure they will not name it that way - much too clear and obvious

 

[dave070]

The only way they could hope to fix the current USB-C connector situation is to mandate labeling. As long as the naming is “optional”, then those that benefit from the confusion won’t use labeling.

This is what they should do then. It makes no sense to have a standard and then leave all these things optional. Creates confusion in the marketplace.

 

[Unregistered 4U]

This is what they should do then. It makes no sense to have a standard and then leave all these things optional. Creates confusion in the marketplace.

Yeah, as long as those companies making the cheap cables don’t want anyone to know the difference between what they sell and what USB-C customers need, they’re going to make sure the USB-IF always says that the labeling is optional.

 

[sideshowuniqueuser]

I think we are on complete different numbering systems. 47 would 35 version 12

Let's see how it looks in a few versions...

 

[carestudio]