I wouldn't know. Maybe an expert will be able to judge it? I can imagine it depends on whether the components are clearly marked and identifiable. By the way, it is also entirely possible that the chip contains an additional USB 2.0 controller or that it is routed differently. Earlier models used to connect the keyboard etc. though a separate internal USB port for example, but I don't think this is the case anymore.
MacBook Neo is the Chosen One
- Thread starter RockRock8
- Start date
I wouldn't know. Maybe an expert will be able to judge it? I can imagine it depends on whether the components are clearly marked and identifiable. By the way, it is also entirely possible that the chip contains an additional USB 2.0 controller or that it is routed differently. Earlier models used to connect the keyboard etc. though a separate internal USB port for example, but I don't think this is the case anymore.
Doesn't Apple use a proprietary internal connection system called apple fabric? and not standard PCIe? They do have support for PCIe NVMe right for iPhone, but it's not a standard bus, so they don't expose PCIe busses directly to the rest of the logic board especially on A chips? Does this change the likelihood of being able to easily add a USB controller chip for the port and connect to a standard traditional PCIe bus?
What Is Apple Fabric? Does My Mac Use Nan… - Apple Community
Where I got this info from.
Apple has been using SPI for internal keyboards (and the trackpad too I think?) for some time. When you control the silicon on both ends, don't need to send the signals long distance or through external connectors, and want to save power and eliminate some components, it can make a lot of sense to switch from a relatively complex and heavyweight bus design like USB to something bone simple like SPI.
(Here's how simple: SPI consists of just four signals, serial data in, serial data out, chip select, and clock. If you don't already know how SPI works, you can probably make a reasonable guess just from knowing that pinout.)
Why would Apple do this? Well, there's something annoying about USB which might not be obvious to most. USB1 (and USB2 because it inherits a lot of things from 1.x) was designed back in the 3.3V logic era, and thus uses a 0V to 3.3V signal swing. This is caveman stuff today. In fact it's kinda impossible, if you're in any halfway modern high performance logic process node. None of them can do 3.3V IO or power rails.
So, whenever a modern SOC needs a USB1/2 port, it has to have a little outside helper chip to act as a fancy kind of voltage level converter. (I think it's more complicated than that, because USB, but logically that's what it's doing.)
By substituting SPI for USB, Apple can cut out the middleman silicon. SPI started out at 5V iirc, but there's not really formal specs for it (it's one of those ad hoc standards), so nobody cares if you make a SPI link which uses 1.0V CMOS signal levels, or whatever.
So what do you think this means for my question?