I can't imagine Apple is dropping audio/video etc. support over the connector. All existing accessories would be obsolete. Not a big deal if it's a dock for $50 but I wouldn't like to see support for an almost new car gone
By the way, what do we need USB3.0 for? Most people don't sync over cable anymore and I guess apple will be guiding us towards the cloud anyway.
It's not necessarily impossible to extract video signals over the same pinouts, but for what?
If you use your iPhone in your car, it makes sense to use Bluetooth, or, if you're really old fashioned, the headphone jack.
The point isn't that the new connector is ONLY USB 3.0, but that is likely NEWLY supporting USB 3.0.
Also, it's very useful to sync over USB already, and that is only getting more important as HD videos get larger. And do you suppose this might change over the next five-ten years? How fast is bandwidth going to get? LTE is 40mbps in cities. That's good, but its no 5,000Mbps.
But you wouldn't plug an iPhone into another iPhone, you'd plug it into another "PC". If the iPhone had the required TB chip and there was a TB dock connector cable, and you plugged it into a Mac's TB port why isn't any different than taking an external HDD with a TB chip and with a TB cable and plugging it to a Mac's TB port? it's not! There are cost, size, adoption, power, usefulness and many other reasons why it's not feasible but to say that you can't connect the iPhone to TB as a peripheral because it's not a "PC" running a modern Intel CPU and chipset is bullshit.
So when you plug a USB Dock accessory into an iOS device, is it a host then? Or does in invent some new peer to peer form of USB that doesn't exist?
What exactly are you arguing? That it makes sense for a iPhone to be a Thunderbolt peripheral, even though Thunderbolt makes no sense on a mobile device? I'm saying that it makes no sense for Apple to move from USB 2 to Thunderbolt, but it does makes sense to move to USB 3.0 at some point, and that the new cable could facilitate that.
So when you plug a USB Dock accessory into an iOS device, is it a host then? Or does in invent some new peer to peer form of USB that doesn't exist?
What exactly are you arguing? That it makes sense for a iPhone to be a Thunderbolt peripheral, even though Thunderbolt makes no sense on a mobile device? I'm saying that it makes no sense for Apple to move from USB 2 to Thunderbolt, but it does makes sense to move to USB 3.0 at some point, and that the new cable could facilitate that.
Sounds like you're just being argumentative.
I've clearly stated many times that I don't think TB on an IDevice is feasible but you're not seeing what is possible because one part of the total device has an ARM ASIC. As my examples should clearly show you can have a device that supports TB that doesn't have the Intel chipset that is required to host TB. Apple could use the same setup as every other TB connected device to allow data transmissions between a Mac/WinPC with a TB port and a peripheral device.
edit: Here is an image from a teardown of an ATD.
RED: Pericom PI7C9X440SL PCIe-to-USB 2.0 host controller
ORANGE: L129NB11 EFL, which looks to be the Thunderbolt port controller
Where is the Intel Nehalem microarchitecture and Core processors it's suppose to have to be able to access data from a host device that supports TB?
Maybe. But USB would greatly complicate both iPod and iOS device accessories - docks, speakers, chargers, car accessories, etc. (This same smaller dock connector has to trickle down to the iPods too don't forget.) It would also increase Apple's unit manufacturing costs, particularly for the lower-margin iPods, and hurt profit margins. USB 3.0 isn't as cheap as 2.0 yet and even 2.0 probably costs more than the current 30-pin dock connector.
Currently building accessories is relatively simple -- a license from Apple and no real software or drivers required (just listen on and talk on the required subset of the 30-pins needed for the accessory in question). If they go USB 3.0 suddenly the accessories will need USB drivers and other software built-in as well as more expensive and complicated circuitry and chipsets. The cost to design, build and maintain them will be substantially higher and likely the variety of accessories will be much smaller than now. Why Apple would want to kill the vibrant 3rd party accessory market that has helped sell so many iPods, iPhones and iPads I don't know.
I've clearly stated many times that I don't think TB on an IDevice is feasible but you're not seeing what is possible because one part of the total device has an ARM ASIC. As my examples should clearly show you can have a device that supports TB that doesn't have the Intel chipset that is required to host TB. Apple could use the same setup as every other TB connected device to allow data transmissions between a Mac/WinPC with a TB port and a peripheral device.
Please read my previous post which answers this. Yes it'd be possible to have an iPhone with Thunderbolt, but it would be pointless and power hungry. Who needs 10gbps to their phone? It'd need a chip to convert the TB signal to PCIe, then from PCIe to USB to be understood by the ARM CPU, for no advantage over USB at all.
As I wrote earlier in the day, I think Apple will use "smart" cables that support USB 3.0 today, and have the potential to support ThunderBolt in the future, much like today's wired TB cables and gear will be supported by tomorrow's optical TB cables and gear.
I can't imagine Apple is dropping audio/video etc. support over the connector. All existing accessories would be obsolete. Not a big deal if it's a dock for $50 but I wouldn't like to see support for an almost new car gone
By the way, what do we need USB3.0 for? Most people don't sync over cable anymore and I guess apple will be guiding us towards the cloud anyway.
Well, I think you'll be able to support most of the existing docks and devices with the 30-pin to 9-pin-and-phono-plug adaptor. That will give you almost everything you need for now except video, and the future of audio/video is streaming via AirPlay and low-power Bluetooth 4.0.
Heck, one has only to look at the external speaker market to realize that everyone is going Bluetooth, as that gives you iOS, Android and Windows 8 phones and tablet support.
Please read my previous post which answers this. Yes it'd be possible to have an iPhone with Thunderbolt, but it would be pointless and power hungry. Who needs 10gbps to their phone? It'd need a chip to convert the TB signal to PCIe, then from PCIe to USB to be understood by the ARM CPU, for no advantage over USB at all.
I did. You stated what I've stated in this thread and in countless others while at the same time sounding like you're disagreeing my very clear statement making "corrections" about how it is technically possible. I didn't reply to it because I didn't know how to respond to that.
USB 3.0 makes logical sense the new Macs have it, Windows based PC's have it which is the common speed benefit for both. Windows based PC's do not however have any Thunderbolt ports (at least not that I have heard about).
For me though I am looking to go more wireless syncing and skipping my USB 2.0 on my current MBP, perhaps when I upgrade in the future I can benefit from USB3.0.
The chips in the HDDs and monitors are client chips, relatively dumb compared to the host chips in the Mac. If the iPhone was a client device, it'd have to turn the PCIe comms from TB to something the ARM CPU could understand, which'd waste a lot of CPU cycles. Right now the iPhone couldn't be a TB host, the TB controller chip is about half the width of the phone itself. But in any case, the ARM CPU doesn't have PCIe by any stretch of the imagination, so it just wouldn't make any sense at all. Also the TB to Dock Connector cable would cost the same as 20 USB to Dock Connector cables.
I don't think the iPhone 5 will have USB 3. That doesn't mean the Dock Connector 2 won't support USB 3 in the future, but it's probably more likely on the iPhone 5S, and most likely on the 6. AFAIK, the ARM chips don't support it yet, and I'm pretty sure Apple won't waste board space with a dedicated USB 3 chip when it won't provide any advantage. Especially as current USB 3 chips are huge, and drink juice. Apple's not in the business of adding X or Y feature just coz they can, they'll only add something that provides a real use. When ARM/Apple build a USB 3 controller into the CPU itself, where it can just sip power and will use little room, we'll see USB3. Right now the bottleneck isn't the external communications, it's the NAND, which tops out at around 10MB/sec. USB 2 is ~20-25MB/sec.
Tegra 2 supports PCIe, I am 99.9% certain Tegra 3 does. Hard to say with the A5/A6 but there is no reason to believe they could not support PCIe. The real problem with implementation (that could have been spelled out better in this article) is Displayport support. While they could probably make the iPhone behave like a thunderbolt hard drive, it would be a kludge. The device should be a host. If it can't be a host it should just walk away. There is no compelling reason to get a 5Gbps faster connection to the flash memory. It will not come close to maxing out USB3 as it is.
I've clearly stated many times that I don't think TB on an IDevice is feasible but you're not seeing what is possible because one part of the total device has an ARM ASIC. As my examples should clearly show you can have a device that supports TB that doesn't have the Intel chipset that is required to host TB. Apple could use the same setup as every other TB connected device to allow data transmissions between a Mac/WinPC with a TB port and a peripheral device.
edit: Here is an image from a teardown of an ATD.
RED: Pericom PI7C9X440SL PCIe-to-USB 2.0 host controller
ORANGE: L129NB11 EFL, which looks to be the Thunderbolt port controller
Where is the Intel Nehalem microarchitecture and Core processors it's suppose to have to be able to access data from a host device that supports TB?
The iPhone should properly behave like a host, not a device. Yes being a device would be useful, but silly.
In any case, there is no point to thunderbolt on a USB3 iPhone it gets you nothing but added complexity. The media speed will be slower than the connection anyway.
Samsung's new Exynos 5 is a dual core ARM Cortex-15 and apparently the first SoC to include USB 3.0.
iPhone 5 is currently expected to use an "A6" quad core Cortex-A9, more closely related to the Exynos 4 Quad used by the Galaxy SIII.
It will likely have different graphics cores (Apple's A4/5 use PowerVR GPUs, while Samsung uses ARM Mali GPUs), and it could possibly include USB 3.0, but almost certainly will not include PCIe (making Thunderbolt impossible in addition to overkill).
Even if it only supports USB 2.0, it's still likely that Apple designed the new 9-pin Dock Connector to eventually support USB 3.0, and will (as the poster above noted) migrate to USB 3.0 as that becomes increasingly useful as a faster interface (and more widely available on Macs.
Note too that USB 3.0 is designed to be backward compatible, so whatever future iOS devices that support it won't require it to sync.
I did. You stated what I've stated in this thread and in countless others while at the same time sounding like you're disagreeing my very clear statement making "corrections" about how it is technically possible. I didn't reply to it because I didn't know how to respond to that.
Alright, I don't think we were quite clear on what you were getting at. Many things are technically possible, such as a 12-disk RAID on a MacBook, but there is little point in discussing things that are that infeasable
Tegra 2 supports PCIe, I am 99.9% certain Tegra 3 does. Hard to say with the A5/A6 but there is no reason to believe they could not support PCIe. The real problem with implementation (that could have been spelled out better in this article) is Displayport support. While they could probably make the iPhone behave like a thunderbolt hard drive, it would be a kludge. The device should be a host. If it can't be a host it should just walk away. There is no compelling reason to get a 5Gbps faster connection to the flash memory. It will not come close to maxing out USB3 as it is.
USB3 makes good sense. Thunderbolt is pointless.
The Tegra 2 is more based around a traditional motherboard architecture, with a northbridge/southbridge etc. Apple's design is somewhat different to this, it's more tightly integrated and doesn't use a PCIe style bus internally. I'm sure Apple could support PCIe, but adding PCIe just for TB would be somewhat pointless, and I agree completely, TB on an iPhone would be total overkill. There simply aren't enough pins on the connector for TB or DP in any case.
The flash in the iPhone can't even saturate USB2, let alone USB3. When the flash does speed up in the future, Apple'll be more likely to consider USB 3.
The iPhone should properly behave like a host, not a device. Yes being a device would be useful, but silly.
In any case, there is no point to thunderbolt on a USB3 iPhone it gets you nothing but added complexity. The media speed will be slower than the connection anyway.
1) The iPhone should act as both. when appropriate. If you are plugging it into some iPod Dock accessory it's the host, when you plug it into a "PC" it's not.
2) I've mentioned NAND speed as being a major bottleneck many many times, as well as many other reasons why it's not feasible or likely. Not likely or not feasible does not mean it's technologically impossible.
3) So in conclusion I've proven it's possible for any such device to connect to a "PC" that is using Intel Nehalem or better via TB.
Alright, I don't think we were quite clear on what you were getting at. Many things are technically possible, such as a 12-disk RAID on a MacBook, but there is little point in discussing things that are that infeasable
I don't think I could have been any more clear. I've looked at my comments and I've stated everything as plainly and thoroughly as possible.
To recap, I was responding to this comment...
While modern Macs have Intel chips and support a PCIe architecture, no iOS devices do. [...] That rules out any support for Thunderbolt and its blazing fast speeds that iOS devices couldn't make any effective use of anyway.
I even used examples that immediately disproved the comment that you can't have a device that supports TB if it doesn't have and modern Intel processor to be a host.
This article contains speculation I've had for awhile... ever since I heard it was going to be 9 pins. I'm a little disappointed AI wasn't able to confirm any of it with any sources. The dock connector had lots of options because it had so many pins-- they could add new meanings to pins, retire old pins, etc. With only 8 pins I suspect it's just going to be a straight-up USB 3 port and not a self-configuring port like bsimpsen suggests.
I suspect being electrically compatible with USB 3 is going to mean a huge number of third party devices will appear quickly, sanctioned by Apple or not...
Comments
Quote:
Originally Posted by sirdir
I can't imagine Apple is dropping audio/video etc. support over the connector. All existing accessories would be obsolete. Not a big deal if it's a dock for $50 but I wouldn't like to see support for an almost new car gone
By the way, what do we need USB3.0 for? Most people don't sync over cable anymore and I guess apple will be guiding us towards the cloud anyway.
It's not necessarily impossible to extract video signals over the same pinouts, but for what?
If you use your iPhone in your car, it makes sense to use Bluetooth, or, if you're really old fashioned, the headphone jack.
The point isn't that the new connector is ONLY USB 3.0, but that is likely NEWLY supporting USB 3.0.
Also, it's very useful to sync over USB already, and that is only getting more important as HD videos get larger. And do you suppose this might change over the next five-ten years? How fast is bandwidth going to get? LTE is 40mbps in cities. That's good, but its no 5,000Mbps.
Don't I wish...
Quote:
Originally Posted by SolipsismX
But you wouldn't plug an iPhone into another iPhone, you'd plug it into another "PC". If the iPhone had the required TB chip and there was a TB dock connector cable, and you plugged it into a Mac's TB port why isn't any different than taking an external HDD with a TB chip and with a TB cable and plugging it to a Mac's TB port? it's not! There are cost, size, adoption, power, usefulness and many other reasons why it's not feasible but to say that you can't connect the iPhone to TB as a peripheral because it's not a "PC" running a modern Intel CPU and chipset is bullshit.
So when you plug a USB Dock accessory into an iOS device, is it a host then? Or does in invent some new peer to peer form of USB that doesn't exist?
What exactly are you arguing? That it makes sense for a iPhone to be a Thunderbolt peripheral, even though Thunderbolt makes no sense on a mobile device? I'm saying that it makes no sense for Apple to move from USB 2 to Thunderbolt, but it does makes sense to move to USB 3.0 at some point, and that the new cable could facilitate that.
Sounds like you're just being argumentative.
I've clearly stated many times that I don't think TB on an IDevice is feasible but you're not seeing what is possible because one part of the total device has an ARM ASIC. As my examples should clearly show you can have a device that supports TB that doesn't have the Intel chipset that is required to host TB. Apple could use the same setup as every other TB connected device to allow data transmissions between a Mac/WinPC with a TB port and a peripheral device.
edit: Here is an image from a teardown of an ATD.
RED: Pericom PI7C9X440SL PCIe-to-USB 2.0 host controller
ORANGE: L129NB11 EFL, which looks to be the Thunderbolt port controller
Where is the Intel Nehalem microarchitecture and Core processors it's suppose to have to be able to access data from a host device that supports TB?
Maybe. But USB would greatly complicate both iPod and iOS device accessories - docks, speakers, chargers, car accessories, etc. (This same smaller dock connector has to trickle down to the iPods too don't forget.) It would also increase Apple's unit manufacturing costs, particularly for the lower-margin iPods, and hurt profit margins. USB 3.0 isn't as cheap as 2.0 yet and even 2.0 probably costs more than the current 30-pin dock connector.
Currently building accessories is relatively simple -- a license from Apple and no real software or drivers required (just listen on and talk on the required subset of the 30-pins needed for the accessory in question). If they go USB 3.0 suddenly the accessories will need USB drivers and other software built-in as well as more expensive and complicated circuitry and chipsets. The cost to design, build and maintain them will be substantially higher and likely the variety of accessories will be much smaller than now. Why Apple would want to kill the vibrant 3rd party accessory market that has helped sell so many iPods, iPhones and iPads I don't know.
Please read my previous post which answers this. Yes it'd be possible to have an iPhone with Thunderbolt, but it would be pointless and power hungry. Who needs 10gbps to their phone? It'd need a chip to convert the TB signal to PCIe, then from PCIe to USB to be understood by the ARM CPU, for no advantage over USB at all.
http://www.isights.org/2012/09/inside-the-iphone-5-smart-cable.html
Quote:
Originally Posted by sirdir
I can't imagine Apple is dropping audio/video etc. support over the connector. All existing accessories would be obsolete. Not a big deal if it's a dock for $50 but I wouldn't like to see support for an almost new car gone
By the way, what do we need USB3.0 for? Most people don't sync over cable anymore and I guess apple will be guiding us towards the cloud anyway.
Well, I think you'll be able to support most of the existing docks and devices with the 30-pin to 9-pin-and-phono-plug adaptor. That will give you almost everything you need for now except video, and the future of audio/video is streaming via AirPlay and low-power Bluetooth 4.0.
Heck, one has only to look at the external speaker market to realize that everyone is going Bluetooth, as that gives you iOS, Android and Windows 8 phones and tablet support.
I did. You stated what I've stated in this thread and in countless others while at the same time sounding like you're disagreeing my very clear statement making "corrections" about how it is technically possible. I didn't reply to it because I didn't know how to respond to that.
For me though I am looking to go more wireless syncing and skipping my USB 2.0 on my current MBP, perhaps when I upgrade in the future I can benefit from USB3.0.
Couldn't wait until tomorrow to find out for sure, eh DED?
Today is literally the last day you can speculate about the new connector rumors.
Tegra 2 supports PCIe, I am 99.9% certain Tegra 3 does. Hard to say with the A5/A6 but there is no reason to believe they could not support PCIe. The real problem with implementation (that could have been spelled out better in this article) is Displayport support. While they could probably make the iPhone behave like a thunderbolt hard drive, it would be a kludge. The device should be a host. If it can't be a host it should just walk away. There is no compelling reason to get a 5Gbps faster connection to the flash memory. It will not come close to maxing out USB3 as it is.
USB3 makes good sense. Thunderbolt is pointless.
The iPhone should properly behave like a host, not a device. Yes being a device would be useful, but silly.
In any case, there is no point to thunderbolt on a USB3 iPhone it gets you nothing but added complexity. The media speed will be slower than the connection anyway.
Samsung's new Exynos 5 is a dual core ARM Cortex-15 and apparently the first SoC to include USB 3.0.
iPhone 5 is currently expected to use an "A6" quad core Cortex-A9, more closely related to the Exynos 4 Quad used by the Galaxy SIII.
It will likely have different graphics cores (Apple's A4/5 use PowerVR GPUs, while Samsung uses ARM Mali GPUs), and it could possibly include USB 3.0, but almost certainly will not include PCIe (making Thunderbolt impossible in addition to overkill).
Even if it only supports USB 2.0, it's still likely that Apple designed the new 9-pin Dock Connector to eventually support USB 3.0, and will (as the poster above noted) migrate to USB 3.0 as that becomes increasingly useful as a faster interface (and more widely available on Macs.
Note too that USB 3.0 is designed to be backward compatible, so whatever future iOS devices that support it won't require it to sync.
Alright, I don't think we were quite clear on what you were getting at. Many things are technically possible, such as a 12-disk RAID on a MacBook, but there is little point in discussing things that are that infeasable
The Tegra 2 is more based around a traditional motherboard architecture, with a northbridge/southbridge etc. Apple's design is somewhat different to this, it's more tightly integrated and doesn't use a PCIe style bus internally. I'm sure Apple could support PCIe, but adding PCIe just for TB would be somewhat pointless, and I agree completely, TB on an iPhone would be total overkill. There simply aren't enough pins on the connector for TB or DP in any case.
The flash in the iPhone can't even saturate USB2, let alone USB3. When the flash does speed up in the future, Apple'll be more likely to consider USB 3.
1) The iPhone should act as both. when appropriate. If you are plugging it into some iPod Dock accessory it's the host, when you plug it into a "PC" it's not.
2) I've mentioned NAND speed as being a major bottleneck many many times, as well as many other reasons why it's not feasible or likely. Not likely or not feasible does not mean it's technologically impossible.
3) So in conclusion I've proven it's possible for any such device to connect to a "PC" that is using Intel Nehalem or better via TB.
I don't think I could have been any more clear. I've looked at my comments and I've stated everything as plainly and thoroughly as possible.
To recap, I was responding to this comment...
I even used examples that immediately disproved the comment that you can't have a device that supports TB if it doesn't have and modern Intel processor to be a host.
I suspect being electrically compatible with USB 3 is going to mean a huge number of third party devices will appear quickly, sanctioned by Apple or not...