Intel has made good on their promise from 2017 to release the Thunderbolt specifications to the industry so that upcoming products can offer that connection without being tied to an Intel license and the possible limitations included therein. Today Thunderbolt 3 was released to the USB Promoter Group, who promptly undid the insanity that the USB-IF released upon us last week by promptly announcing it will be the basis of USB 4.
Thanks to their lack of an obsession over stringing letters and numbers to the back of USB 3 we will end up with a certified standard that provides "two-lane operation using existing USB Type-C cables and up to 40 Gbps operation over 40 Gbps-certified cables" (pdf link). It will maintain backwards compatibility with previous Thunderbolt generations as well as the various flavours of USB 2 and 3. It may or may not be compatible with the new ones, such as USB 3.2 Gen 2×2 … indeed one might hope they refuse to accept such things into their specifications.
Considering that the USB-IF and USB-PG are closely related, this new nomenclature will be the new standard and last weeks announcement just a memory.
"Releasing the Thunderbolt protocol specification is a significant milestone for making today’s simplest and most versatile port available to everyone. This, in combination with the integration of Thunderbolt 3 into upcoming Intel processors is a win-win for the industry and consumers."
Here is some more Tech News from around the web:
- AMD Ryzen 3000 CPU specs and prices outed by Singaporean retailer @ The Inquirer
- The Current Spectre / Meltdown Mitigation Overhead Benchmarks On Linux 5.0 @ Phoronix
- Microsoft will use Google's Retpoline to mitigate Spectre in Windows 10 @ The Inquirer
- WannaCry-hero Hutchins' trial date set, Microsoft readies Google's Spectre V2 fix for Windows 10, Coinhive axed, and more @ The Register
- Get your first look at the OnePlus 7 and its crazy pop-up camera @ Ars Technica
- Facebook is using your 2FA phone number for 'look up' search feature @ The Inquirer
- DRAM price falls to slow, says Nanya @ DigiTimes
- Sony Officially Ends Production of PS Vita @ Slashdot
- Are Quad-core CPUs Finally Dead in 2019 @ Techspot
- Nvidia RTX 2050 mobile graphics card shows up on Dell website @ The Inquirer
The same people that need
The same people that need this ^ are the same ones that worked on the USB-IF naming scheme
The surreal thing is, they
The surreal thing is, they kinda are.
It looks like the swan song
It looks like the swan song from Intel… L O L
USB 4 is no good, I suggest
USB 4 is no good, I suggest USB 3.4 Hyper championship edition 3×4 superMegaSpeed.
Sebastian and I decided
Sebastian and I decided USB4TB3 2×2 E* Type Cx1 would be most appropriate for the initial launch.
So Intel opens up the TB3
So Intel opens up the TB3 standard by contributing the TB3 siginaling/tunneling portocol to the USB Promoters Group/USB-IF.
And currently Intel has the only controller hardware that supports the TB3 siginaling/tunneling portocol.
So how long will it take any third parties to engineer their own in-house TB3 controller hardware. I guess most USB 4.0 compatible controllers will not have already been in development and that USB 4.0(With all the Bits Working) will take time.
Just look at how many laptops even support the USB 3.1 Gen 2(10Gbs, now renamed to USB 3.2 Gen 2) standard! And that’s almost no Laptop OEMs currently. And what about the recently announced USB 3.2 Gen 2×2(20Gbs over dual 10Gbs pairs) USB-IF standard! Will any Laptop OEM’s go for that or will they look at USB 4.0 that rolls up TB3 and all that other stuff that includes the USB Standards and Alt Standards support into one Swiss Army/Swiss Roll of everything subsumed by the USB Type-C BLOB.
None of this is going to be very practical without Laptop MBs supporting at least PCIe 4.0/5.0 levels of per PCIe lane bandwidth!
And TB3 also supports Tunneling(1) of PCIe 3.0 siginaling via the TB3 tunneling protocal! And that TB3 also can tunnel 10Gb Ethernet and DP 1.2 and up to USB 3.1(I think only Gen 1 as the TB3 PDF is not saying specifically).
So now there needs to be a primer published by the USB-IF that describes the TB3 tunneling protocol and just what siginaling can be tunneled via TB3 and what can be natively siginaled over any of the USB-Type-C cable’s pinout’s native USB signalling pairs
I want to see a graphic that’s a 3D look at a USB Type-C Port/Cable pinout with arrows showing just what differential cable pairs are able to transmit USB, DP, whatever, natively and what USB Type-C differential cable pairs will be transmitting the TB3 tunneling Protocal with all its ability to encapsulate PCIe 3.0, 10Gb Ethernet, DP(1.2), USB(3.0/3.1 Gen 1, 3.2 Gen 1), Whatever siginalling protocols.
That USB 4.0 root hub on the motherboard is sure going to be something to behold what with all the PCIe traces that will have to be run into that complex of protocol controller IP and whatever link layer topoloies that have to be enumerated in the firmware/OS by PC/Laptop/MB OEMs for their specific hardware.
So really that’s loads of encapulated siginalling that will be going/coming on/off of the TB3 contollers ends that will have to be routed and fed to the proper end point consumer of any of that myriad of siginalling protocols that TB3 can Tunnel and then there are the native Protocol controllers that can push their signalling natively across the very same USB Type-C cable pairs.
Laptop OEM’s are sure known for advertising any and all theoretical maximum bandwidths able to be pushed via whatever Protocol/Ports the laptop will be supporting. But Laptop OEMs are sure not very good a supplying the Needed PCIe MB connectivity to properly feed all the very very bandwidth hungry Protocal controllers if they should actually all be needing their bandwidth at once.
So Let’s ask all of the Laptop OEMs just when, in actual time, do they expect to be supporting PCIe 4.0, or 5.0, motherboards in order to properly support all that USB 4.0 bandwidth that includes TB3 and its 40Gbs needs also. And I guess the laptop OEMs will be laying on the Processor Makers legs and waiting for all that USB support to end up coming from the chipsets that the Processor makers will be providing to the MB makers. And the Laptop OEMs all make their own MB, or subcontract the Laptop MB’s custom design to some third party subcontractor.
Really, the CPU/APU chipsets will need to be supporting PCIe 4.0, or 5.0, for any of this USB 3.2 Gen 2×2, or 4.0 standards, to be possible on Laptops.
“Thunderbolt 3 Technology Breif”
https://thunderbolttechnology.net/sites/default/files/HBD16235_Thunderbolt_TB_r05.pdf
Best news I’ve heard this
Best news I’ve heard this year.
I don’t understand what the
I don’t understand what the problem is. I have four kids named Joseph, Joey, Josephine and Josie and a 5th on the way that I’m naming Joe, to avoid confusion. It’s more convenient for me, because all the monogrammed hand-me-downs can be shared, so I don’t have to rebrand them. The shirts and towels, not the kids.