USB4 Version 2.0 to Offer 80Gb/s Transfer Speeds

[constructor]

So how will this USB 2 v2.0 standard compare to Thunderbolt 3 (or are we at 4?) in terms of real-life speed rather than theoretical?

I expect Thunderbolt to also get the doubled speed so the relationship should remain the same as it is now, just at double the data throughput.

 

[boyarka]

Year of our Lord 2022. New Thunderbolt standard to be adopted - same as the Fruit company pioneered and promoted. In, perhaps, 30 more years time - it may come to the iPhone as well.
For the Fruit is bitter-sweet, and too many Cooks in the kitchen pepair this Apple pie.

 

[femike]

The confusion is deliberate. Confusing brings in more money.

 

[citysnaps]

Yes, I'm sure you understand completely the simplistic and ever-so-straightforward differentiations between USB 3.2 Gen 2, USB 3.2 Gen2x2, USB-C, Thunderbolt 3, Thunderbolt 4, USB4, USB4 Version 2.0 and their varying levels of throughput, Alternate Mode support, power delivery limitations, cross-compatibilities, display support, etc.

Off the top of my head from memory in 100% precise detail, no. In general, broad terms, yes.

But if I were to need a USB cable for a certain bandwidth I certainly know how to get that information and read before making a purchase decision.

That's how I came to purchase the proper cables for driving four displays (one 5K, three 4K) from my Mac Studio Ultra. Not a big deal. It's certainly nothing (for me) to whine about. Unless that's what gets you feeling better starting your day.

It's not rocket science.

 

[ajfahey]

No thunderbolt announcement?

 

[mcbainsloafers]

Lightning already charges about as fast as Apple probably wants (up to 27 watts). There are phones that charge at 50+ watts but it's not likely Apple wants to get there. USB-C iPhones I predict will charge up to 30 watts which is where the iPads with USB-C are at, only 3 more watts.

Battery health is okay with fast charging, they just make sure to check the temperatures and ensure the charging rate isn't high when the phone is hot. And they slow down charging as the battery percent gets higher.


Usually there are labels but you have to understand the labels. If they say SS and they have the USB logo, they are probably 5 Gbps USB 3.x gen 1 cables. If they say SS10, they are probably 3.x gen 2 cables (3.1 or 3.2, they are the same when gen 1 or gen 2). If they say SS20, then it's 3.2 gen 2x2.

Thunderbolt cables need the Thunderbolt logo: https://en.wikipedia.org/wiki/Thunderbolt_(interface)#/media/File:Thunderbolt_Symbol.svg

So if you see the Thunderbolt logo you will know it's a Thunderbolt cable.

The PD rating is split between 60 watts, 100 watts, and 240 watts cables. You pretty much have to look at the manufacturer's website, packaging, or some review somewhere to see its PD rating, or you just test it out yourself. A cable might also say on it how many watts it has. 240 watts cables are brand new and just came out and not too many devices use more than 100 watts for the time being so you can discount needing 240 watts yet.

The big problem is how you can't easily get all the features in one cable, including Thunderbolt, and you sacrifice something to get all the features or you pay a lot of money. Apple's passive .8 m Thunderbolt cable has all the features (and being passive supports the new 80 Gbps mode) but it's only .8 m, Apple's 1.8 m cable costs $130 and it appears it won't support 80 Gbps because it's active and not certified for the new 80 Gbps mode, same for the 3m cable which is more expensive at $160.

Why does this need to be so difficult? Every cable should be labeled with two numbers, Gbps (G for short) and watts (W) for short. That is all we really care about. Then I could just by a 20G / 100W cable and know exactly what I am getting. When USB-C 5.7 Rev J comes out I won’t care. I will just look for 80G/140W or whatever specs I need on the label.

 

[Muzzakus]

Do we know if iPhone 14 is moving away from lightning yet?

 

[fromgophonetoiphone]

As a techie, the list here is insane and even worse for regular consumers. It allows for deceptive marketing by companies getting to make their cables seem like the newer/better ones, and is also a mouthful for no good reason.

The problem is that table is confusing too. 3.2 Gen 1, Gen 2 etc didn't exist when only 3.1 was out, so there were 3.1 Gen 1 and 3.1 Gen 2 products (now 3.2 Gen 1 and 3.2 Gen 2 respectively). So you may get even more confused if you find products that have not had their labels updated.

And then I'm sure there are products and descriptions out there that just lump all of 3.x as 3.0. This table, while helpful is only as good if all the marketing, product packaging, and retailer teams properly label their products.

 

[Nicky G]

USB4 Version 2.0, eh?

Yeah.

Someone needs to take the USB-IF folks on the naming committee out behind the shed to be dispensed with -- proverbially, of course!

No but seriously, anyone involved with naming at USB-IF needs to be removed from their positions. I will gladly offer free consulting to effin' fix things already, OHMYGAWD.

 

[ArkSingularity]

USB4 Version 2.0, eh?

Yeah.

Someone needs to take the USB-IF folks on the naming committee out behind the shed to be dispensed with -- proverbially, of course!

No but seriously, anyone involved with naming at USB-IF needs to be removed from their positions. I will gladly offer free consulting to effin' fix things already, OHMYGAWD.

Don't worry, they'll find a way to make it even worse. USB 4.1 version 2 gen 3, coming to a city near you. 😂

 

[theorist9]

At first I thoght this might work the same way as Display Port AltMode 2.0 (which gives USB a run for the money in cryptic naming conventions), which achieves 80 Gb/s with Thunderbolt 4 by taking both the incoming and outgoing 40 Gb/s data lanes TB4 provides, and combining them to get the bandwidth for a single 80 Gb/s outgoing signal—i.e., by turning a duplex system into a (mostly) simplex one (I think there is still provision for the device to communicate back to the host to provide status info.).

But the article says they instead achieve this by adding another physical layer. Does this mean this actually provides 80 Gb/s duplex? I.e. 80 Gb/s in both directions?

 

[ArkSingularity]

At first I thoght this might work the same way as Display Port AltMode 2.0 (which gives USB a run for the money in cryptic naming conventions), which achieves 80 Gb/s with Thunderbolt 4 by taking both the incoming and outgoing 40 Gb/s data lanes TB4 provides, and combining them to get the bandwidth for a single 80 Gb/s outgoing signal—i.e., by turning a duplex system into a (mostly) simplex one (I think there is still provision for the device to communicate back to the host to provide status info.).

But the article says they instead achieve this by adding another physical layer. Does this mean this actually provides 80 Gb/s duplex? I.e. 80 Gb/s in both directions?

I'm curious about this as well. Thunderbolt only carries the signal with two physical pairs of wires (out of 24 pins in the connector), and I have always been curious if they would ever utilize other pins and use them to provide more channels. Then I looked at the pinouts and realized that USB-C has these pins designated for very specific purposes and my hopes haven't been too high ever since.

The language they're using seems to suggest it's not totally out of the realm of possibility. My hope is that they are at least exploring it, because trying to run data at ever increasing clock rates just increases the cost of cables and thunderbolt-supported devices as a whole and makes it even harder to make inexpensive passive cables that perform on par with expectations.

 

[dwaite]

why usb 4 2.0. they are starting it again 😭.
USB
USB 2.0
and USB 3.0. Good old time.

They painted themselves in the corner. The reason it is USB4 and not USB 4 - it isn't backward compatible. Older USB support runs over the same port, not the same protocol.

So if they want to come out with a backward compatible protocol based on USB4.. its USB4 version 2.

But this still should _only_ be a problem for the implementers, because there's completely different, sane branding for USB cables and ports that describe their capability (which absolutely nobody uses.)

 

[UltimateSyn]

Off the top of my head from memory in 100% precise detail, no. In general, broad terms, yes.

But if I were to need a USB cable for a certain bandwidth I certainly know how to get that information and read before making a purchase decision.

That's how I came to purchase the proper cables for driving four displays (one 5K, three 4K) from my Mac Studio Ultra. Not a big deal. It's certainly nothing (for me) to whine about. Unless that's what gets you feeling better starting your day.

It's not rocket science.

Good for you. There is a clear consensus that the naming conventions, as well as the distinction between the various parallel standards which all use the same port, and the documentation that details which devices and cables allow each of the available features, are confusing. Very confusing. Can it be figured out in most instances given enough time and research? Sure. Is it easy, straightforward and clear as these industry standards should be? Absolutely not.

Ironically my day job includes a bit of rocket science. And the USB-IF standards are still a mess.

 

[eifelbube]

Could just call it USB 5 and move away from this daft situation we are in with multiple things being called the same thing.

Shouldn‘t it be USB 9?
USB1-1.0
USB2-1.0
USB3-3.0
USB3-3.1 gen 1
USB3-3.1 gen 2
USB3-3.2 gen 1
USB3-3.2 gen 2
USB4-1.0
USB4-2.0
…
USBM
Universal Serial Bus Mess

 

[Bustycat]

Shouldn‘t it be USB 9?
USB1-1.0
USB2-1.0
USB3-3.0
USB3-3.1 gen 1
USB3-3.1 gen 2
USB3-3.2 gen 1
USB3-3.2 gen 2
USB4-1.0
USB4-2.0
…
USBM
Universal Serial Bus Mess

You missed these relations.

USB 3.0 = USB 3.1 Gen 1 = USB 3.2 Gen 1
USB 3.1 Gen 2 = USB 3.2 Gen 2
USB 3.2 Gen 1x2
USB 3.2 Gen 2x2
USB4

Be careful about the spaces.

 

[constructor]

At first I thoght this might work the same way as Display Port AltMode 2.0 (which gives USB a run for the money in cryptic naming conventions), which achieves 80 Gb/s with Thunderbolt 4 by taking both the incoming and outgoing 40 Gb/s data lanes TB4 provides, and combining them to get the bandwidth for a single 80 Gb/s outgoing signal—i.e., by turning a duplex system into a (mostly) simplex one (I think there is still provision for the device to communicate back to the host to provide status info.).

But the article says they instead achieve this by adding another physical layer. Does this mean this actually provides 80 Gb/s duplex? I.e. 80 Gb/s in both directions?

I expect them to switch from pure digital (0/1) to multi-voltage (0/1/2/3) signaling – this requires tighter (adaptive) calibration but does not raise the signal frequencies while still doubling data throughput.

This also fits with existing passive cables still working (because multiple levels are not a problem there, but higher frequencies would be) but active cables needing to be replaced (because the existing active cables have digital signal amplifiers in each end which can't cope with multilevel signaling, so new active cables with multilevel-enabled amplifiers will be needed).

I have no actual inside information, but that would be the most feasible path to take and it would be consistent with the cable compatibility they have announced.

 

[ian87w]

Naturally the required cables look exactly like all the other cables so you never know what's going to work.

This is the problemm with USB-C. So many USB-C cables out there are basically just dumb USB2.0 cable with USB-C connector. It's increasingly difficult finding an actual USB-C USB3 unless you just get the pricey Apple ones or know to simply search for Thunderbolt 4 cable.

 

[Tagbert]

Do we know if iPhone 14 is moving away from lightning yet?

All of the rumors about switching to USB-C are in the context of the iPhone 15. I could see Apple doing a surprise switch for the 14 Pros this year, but since we haven’t heard anything about USB-C for the 14s, I doubt they are doing it. It’s a pretty obvious physical change that probably would have been picked up in the supply chain if it was happening. I wish it would though. Just took a trip and would have loved to simplify on just a USB-C connection for all devices.

 

[basehead617]

Isn't that faster than the fastest M2/SSDs?

 

[iPhoneFan5349]

That’s incredible

This is definitely going to make a huge difference on future Macs that will have USB 4 Version 2.0 with regards to transferring video files

Apple has definitely staked a huge portion of their future business on the Mac lineup and an improvement as immense as this has massive potential to be a game changer for dominating the computing industry in the years going forward

Definitely am excited to see Apple continuing to innovate and improve upon products that are already unbelievable

You should apply for a spokesperson position.

 

[Tiberius2000]

That’s just to confuse you so you buy the wrong thing first so you have to spend money twice.

Amen. It’s difficult enough with USB-C available on:
v. 3.0
v. 3.1 gen 1
v. 3.1 gen 2
v. 3.2 gen 1
v. 3.2 gen 2
v. 3.2 gen 1x2
v. 3.2 gen 2x2
v. 4
v. 4 2.0
Thunderbolt 3
Thunderbolt 4

And that’s before you go cable shopping and find out your new, braided, kevlar cable is power only.

The effort to unify around one connector lead to a more confusing market than when each had its own connector.

 

[robertcoogan]

USB-C is awesome for speed, but has one of the worst hardware designs out there. It's flimsy and doesn't always sit in the port securely, and is worse when you travel.

 

[deconstruct60]

The USB implementation forum may have the worst branding and marketing department in the history of branding and marketing departments.

It isn't the marketing department solely. Part of this is because USB-IF is a somewhat unwieldy alliance with multiple agendas. That leaks out in some of this hodge podge of branding.

The full details of the specification isn't out yet, but given the older track record of USB-IF there is a pretty good chance that a large fraction (if not almost all ) of the "hot new features" here in USB 4 version/gen 2 are optional. That is why it still has the "USB 4" prefix because the majority of what the system vendors are going to deliver is going to mainly be just USB 4 (version 1). There will be some narrow , relatively inexpensive doo-dad option can add to get to allow the "version 2" branding , but allowed to skip anything that would incur the highest component cost increase.

Similarly it won't be surprising if Thunderbolt 4.2 turns out to be USB 4 version with the optional stuff required.

For example, will not be surprising if they probably allow a slim-shady way of saying you hit the 80Gb/s bandwidth mark by just allowing DisplayPort 2.0 alt mode. That happens to be 80Gb/s in only one direction ( by 'reversing' the 40Gb/s inbound 'lane' to both going outbound; so aggregate to 80Gb/s). The 80Gb/s bidirectional spec will be an optional subsection of "USB 4 version 2" that folks don't have to implement.


The split between passive and active in bandwidth... driven by implementers and subsection of USB-IF that don't want to see cable cost (bill of materials) go higher. It isn't some coherent marketing plan. It is two factions getting what they want. There are "users" , "implementors" , system vendors on the USB-IF committee. It is a broad set of folks with diverging interests.

The name is a joke. Even worse though is that we now have yet another cable that looks identical to existing cables but has different capabilities. 🤦‍♂️

If you don't look at the end points most folks couldn't tell most cables apart at all. 10Gb/E or 1Gb/E cable difference is looking at the middle of the cable is what? (some happen to have writing on them , but some do not. )

80Gb/s without active is going to be far more problematical. Very high bandwidths and long copper cables has physics problems/issues.

 

[Nermal]

Not likely as the macOS assumes the GPU is on the SOC sharing memory with the CPU on Apple Silicon systems.

It assumes that now, but it doesn't necessarily have to assume that in the future. The OS itself can handle separate memory (as that's how it works on Intel systems) so Apple would just need to enable the same/a similar code path for external GPUs.

Whether Apple wants to do so is another matter entirely.