Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
So I wonder why Apple didnt use Titan Ridge from Intel?

they did. Most of the assertion that they didn't isn't based on facts.

That XDR hooks up to any of the Thunderbolt ports on the Mac Pro 2019 (if feed from the GPU sufficiently) is illustrative that they did use Titan Ridge. ( need to be able to pass through two DisiplayPort 1.4 streams onto the TB network. ).

The chips being pointed at are really more "Alternative Mode" assist chips (and/or pragmatically being used as re-drivers to cover longer distances. ).
 
  • Like
Reactions: thisisnotmyname
According iFixit's teardown, the mp 7,1 instead Intel Titan Ridge tb3 headers it implements 2x Diodes Incorporated PI3DBS16 PCIe Thunderbolt 3 signal mux ( functional equivalent to Intel Titan Ridge).



Digging whiting it's specifications, it's notorious this part is full pcie 4 compatible, which makes more sense seing it inside an Ryzen motherboard.

It could means either Intel titan ridge too expensive, Apple depart Intel as supplier, AMD Zen 2 on the way.

This Chip could implement full tb3 with just 2x pcie4 lines, supports full dp.1.4 (6k/8k) etc.

What you think?

When Ice Lake / Tiger Lake Xeons get released next year with ( allegedly ) in-silicon TB3 , I bet Apple will take advantage of it . They are unlikely , though , to be socket compatible with the current MP7,1 . They're scheduled to be LGA 4189 socket size . Users already are struggling with processor upgrades right now with Cascade Lake LGA 3647 installations . What's a few extra pins ?
 
When Ice Lake / Tiger Lake Xeons get released next year with ( allegedly ) in-silicon TB3 , I bet Apple will take advantage of it . They are unlikely , though , to be socket compatible with the current MP7,1 .

Intel has only put "integrated' Thunderbolt into systems with integrated GPUs. There isn't really even a "socket" as so far those as so far they are all ball grid array (BGA).

The "ice Lake" Xeon SP and likely W CPU packages won't have an integrated GPU and also have much bigger transistor budgets for other stuff. So probably won't have integrated Thunderbolt. . Xeon spans a large range though Xeon E for Tiger Lake might (if there is an 'E' version with that microarchitecture).


They're scheduled to be LGA 4189 socket size . Users already are struggling with processor upgrades right now with Cascade Lake LGA 3647 installations . What's a few extra pins ?

Those 4189 socket size growth is highly likely to get two CPU dies into the package ( e.g., something like two 10 core dies to get to 20 cores or two 14's to get to 28 ). If they are stretching to do big dies, it is not very likely they are going to be adding something like Thunderbolt or GPU subsystems to the dies. )

Intel isn't going to completely get out of the discrete TB controller business at all. With DisplayPort 2.0 going to Thunderbolt base line tech it doesn't hurt to get a jump start coupling TB controllers now to your GPU for practice. Intel probaby a bit ahead of the curve on that. 3 more years it will be probably be more common. (just not in the server and high end workstation CPU packages. )
 
Note that I emphasized *require*….

If Thunderbolt didn't help propagate DisplayPort then they wouldn't have joined up to support it.
Same with USB. To get USB-IF buy in Thunderbolt v3 controllers had to add in USB controllers and minimally provision USB 2.0 to every port as a requirement. Guess what? 2 years later USB-IF is using more Thunderbolt.

It is about relationships with standards bodies and their dynamics. Rouge folks who come along and pilfer the connector or parts of a standard and don't give nothing back typically get shunned over time.

But if deeply wedded to 1980's dogma about the limitations of add-in-cards ... yeah probably would miss that.
 
Rouge folks who come along
Ah yes, rouge.

Chanel_02[1].jpg
 
  • Like
Reactions: Mago
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.