Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
I don't believe that alternative mode has been announced - only DisplayPort. Which is not to say it couldn't come, but I don't believe another signal types have been announced (EDT: in an earlier AnandTech article they mention it only as a possibility and this announcement was that the first alternative mode is DisplayPort, there may be more coming). Also that's not quite the same thing as Thunderbolt which encapsulates other protocols and allows direct access by PCI. And in the case of eGPUs, even Thunderbolt 2 has trouble sending enough bandwidth (currently) - though tests on Thunderbolt eGPUs showed it wasn't as bad as you might think.

If they gave USB 3.1 a PCIe passthrough ability, it would basically be integrating thunderbolt into the USB 3.1 standard. That would be a good thing, but I doubt it would happen. USB prides itself on better security over thunderbolt/firewire, as data has to pass through the CPU as a peripheral instead of getting direct access to the PCIe bus.
 
If they gave USB 3.1 a PCIe passthrough ability, it would basically be integrating thunderbolt into the USB 3.1 standard. That would be a good thing, but I doubt it would happen. USB prides itself on better security over thunderbolt/firewire, as data has to pass through the CPU as a peripheral instead of getting direct access to the PCIe bus.

That's my understanding as well (also you probably couldn't get the full bandwidth potential of Thunderbolt), but until something is announced it's hard to say what their plans are or what form it might take ... (EDIT: re-reading that last statement: well I guess that's almost tautologically true! :p)

----------

I love how we're getting yet another standard. My plethora of USB cables in my desk at home would disagree with the "universal" point of the name. But at least they're cheap, unlike lightning or thunderbolt.

To be fair, even standards have to evolve :)
 
So before Thunderbolt gained traction it has serious competition?
They should have priced it reasonably while they had the chance, I wonder if TB will be the next FireWire as far as consumers are concerned.

I've given up on that port on this Mac.

Maybe the next one, then again, my next Mac will have something better than USB 2.0 which Apple deemed a solid choice for the iMac in 2011. :rolleyes:

Glassed Silver:mac
 
I think the problem with most of you guys is that you need to realize Thunderbolt is not for you. It's for professionals. It doesn't have the overhead that USB does and it's twice as fast as this new USB standard, with optical versions on the way that go 50Gbps and 100Gbps bi-directional. It's about having direct access to the motherboard externally. You run things like USB on top of Thunderbolt.

I call BS on that. Why?
Because they included 2 Thunderbolt ports in my iMac in 2011, when I already had a USB 3 HDD that just happened to be the best bang-for-buck Western Digital external HDD some time before that already.

So I was USB 3 ready, yet they decided to give me TB instead of USB 3.

Now granted, I COULD shell out the money for Thunderbolt solutions like docks etc...
But I'm also the kind of person who only got an iMac over the Mac Pro because of the convenience of a compact design and the Mac Pro being very outdated back then.

PS: Stop telling people what they need. That's just smug and you don't know them well enough, let alone their use cases.

Glassed Silver:mac
 
I call BS on that. Why?
Because they included 2 Thunderbolt ports in my iMac in 2011, when I already had a USB 3 HDD that just happened to be the best bang-for-buck Western Digital external HDD some time before that already.

So I was USB 3 ready, yet they decided to give me TB instead of USB 3.

Now granted, I COULD shell out the money for Thunderbolt solutions like docks etc...
But I'm also the kind of person who only got an iMac over the Mac Pro because of the convenience of a compact design and the Mac Pro being very outdated back then.

PS: Stop telling people what they need. That's just smug and you don't know them well enough, let alone their use cases.

Glassed Silver:mac
I assume you are aware that pretty much all Macs got TB in 2011 and USB 3 in 2012, ie, it wasn't for a very long time. And USB 3 only become part of the Intel motherboards in 2012, before that USB 3 required an extra chip. One can argue that Apple didn't want to redesign their laptops to implement USB 3 in 2011. Or can argue that they were slow as usual (Apple tends to be a fast mover in a few select big things: TB, retina screens and a slower mover in more mundane things). Or can argue that Apple wanted to give TB a head-start of one year.
 
I call BS on that. Why?
Because they included 2 Thunderbolt ports in my iMac in 2011, when I already had a USB 3 HDD that just happened to be the best bang-for-buck Western Digital external HDD some time before that already.

So I was USB 3 ready, yet they decided to give me TB instead of USB 3.

Now granted, I COULD shell out the money for Thunderbolt solutions like docks etc...
But I'm also the kind of person who only got an iMac over the Mac Pro because of the convenience of a compact design and the Mac Pro being very outdated back then.

PS: Stop telling people what they need. That's just smug and you don't know them well enough, let alone their use cases.

Glassed Silver:mac

Apple switched to Thunderbolt first because Intel was slow to adopt native USB 3 on their CPUs and Apple didn't want to add support themselves to the CPUs. Apple was Intel's partner in bringing Thunderbolt to market and Thunderbolt could effectively replace ExpressCard slots and Firewire ports in a thinner, smaller port. Thus, Apple adopted Thunderbolt before USB 3.0. Once Intel added native support the following year, Apple adopted USB 3.0.

Almost all rotational hard-drives are best served by USB 3.0 unless you have a lot of them in a RAID setup. Thunderbolt is overkill as USB 3.0 is quite capable of saturating even the fastest rotational drive's read/write capability. So I wouldn't recommend moving to Thunderbolt for that anyway.

I think a better way of phrasing it is: if you need a replacement solution for an ExpressCard slot or really fast RAID/SSD transfer speeds, go Thunderbolt. If not, go USB. Before this announcement I would've added if you want a simple, convenient dock for a laptop, go Thunderbolt. But when devices and docks supporting the new USB standard become available, one might as well get what will presumably be a cheaper option for that. Now if eGPUs become a reality in the market (besides the few practically experimental chassis out currently) then there might still be a reason to Thunderbolt here too but I suspect that will be a niche for awhile yet unless someone makes a big push for it.
 
This will be interesting. Let's see if USB still has the same pull if they lose compatibility. Having the same plugs made migrating to USB 2 and USB 3 a cheap and easy choice. Not doing that is what pushed FW800 out, and what made Thunderbolt so weak outside the Apple world (where it was a very natural migration for the mini DP).

Compatibility is what makes or breaks these standards. Most people don't care about a few Gb speed difference. But every new connector means less room on a laptop for connectors the customer already uses.

so again: This will be interesting

(and yes, if Thunderbolt had come in USB shape and with compatibility, I bet there wouldn't be a USB a 3.1!)
 
I think you mean DVI. I haven't seen a monitor with VGA (that didn't have DVI or HDMI as an option) in years.

I don't know about you, but at my office almost all projectors and monitors still work on VGA. Not to mention, most of the netbooks there come with a full VGA port as well, so sadly VGA is still quite relevant.

––––––

Seems to me that Apple would fancy adopting this, perhaps trying to make all ports reversible (such as their lightning cables, and rumored new USB cables)
 
Usb - C + DisplayPort 3 it's the right solution for future ARM based Macs (either a server or a desktop/laptop) as well for low end Intel Macs w/o Thunderbolt capability. It's smarter since an usb3 header maybe shared by n terminal while having only one capable of Display Port, not to account much cheaper than HDMI.
 
Lightnight port replacement???

Could this be a potential replacement for future lightning port on iPhones, iPod Touches, and iPads?
 
I call BS on that. Why?
Because they included 2 Thunderbolt ports in my iMac in 2011, when I already had a USB 3 HDD that just happened to be the best bang-for-buck Western Digital external HDD some time before that already.

So I was USB 3 ready, yet they decided to give me TB instead of USB 3.

Now granted, I COULD shell out the money for Thunderbolt solutions like docks etc...
But I'm also the kind of person who only got an iMac over the Mac Pro because of the convenience of a compact design and the Mac Pro being very outdated back then.

PS: Stop telling people what they need. That's just smug and you don't know them well enough, let alone their use cases.

Glassed Silver:mac

It wasn't Apple that decided. Intel instruction sets didn't support usb 3 until 2012 with ivy bridge. You should have waited or resold like the rest of us
 
Usb - C + DisplayPort 3 it's the right solution for future ARM based Macs (either a server or a desktop/laptop) as well for low end Intel Macs w/o Thunderbolt capability. It's smarter since an usb3 header maybe shared by n terminal while having only one capable of Display Port, not to account much cheaper than HDMI.

Oh FFS.

Why is it that every rumor on this site attracts ARM fanatics hailing whatever the rumor is about as some sign from on high that ARM is coming to the Mac? I would think they'd have given up by now, or maybe listened to the people explaining for a variety of technical and market reasons why Apple is not going to make ARM Macs.
 
It wasn't Apple that decided. Intel instruction sets didn't support usb 3 until 2012 with ivy bridge. You should have waited or resold like the rest of us

Do you always pretend to know what you're talking about?

Instruction sets have nothing to do with USB 3.0, it was a lack of support baked into the chipset and not the cpu.
 
When I first heard about a reversible smaller USB C, there was a part of me that hoped it would be Lightning.

It's sure small enough and sturdy. We could have Lightning ports on the side of our MBP and just use Lightning to Lightning cables to sync or charge.

After reading through this thread it looks like Lightning wouldn't be able to handle some of the heavier stuff like 100w of power.
 
You're right about everything except Thunderbolt. Thunderbolt is now DEAD. This just killed it. It's a total replacement for Thunderbolt 2 while retaining USB connectivity and cable standards AND is reversible. It gets rid of ALL major problems and ALL major complaints and it's a HUGE standard (unlike Thunderbolt which hardly anyone supports or uses except Apple). It sucks to be people who bought into Thunderbolt and it sucks to own a Mac Pro. You've just been PWNED! :eek: :D

I called this a couple of years ago. No one gives a crap about APPLE standards and Intel or not, it's ONLY been really used by Apple. You CANNOT create "standards" when you represent 8% of the market. Making matters worse, Apple wanted to have exclusive access to Thunderbolt the first year. WHY???? There was NOTHING available for it and it only meant that PC adoption would not occur for at least another year to even start. They helped doom their own connection standard. Call it Steve's dying folly if you like, but it was a stupid move.



Dream on. They can't even hardly get any devices to support Thunderbolt I several years later and they're all universally OVERPRICED. Don't give me the "Pro" line of defense because Apple has done everything in its power to KILL its own Pro market and hand it lock, stock and barrel to Windows. The Mac Pro is really a Prosumer machine (given the lack of internal expansion, rack mount capability and even software now that most have left the market), except most prosumers can't afford it. I doubt you'll even see Thunderbolt III. For goodness sake, look how long it's taken to get significant USB3 adoption by consumers (given the lack of desktop sales and the lack of mobile support). By the time Thunderbolt III makes an appearance, USB 4 will be out in short order and that will be the last you will EVER see of Thunderbolt, virtually guaranteed and IMO good riddance. We don't need competing standards and we certainly don't need overpriced products for those standards. Sometimes competition is BAD (look at HDDvd vs. Blu-Ray and how that only slowed adoption).


You seem way too excited about this. You say Apple can't create standards, but it sure looks like they drive others to create better ones. Is it a coincidence that this comes after Thunderbolt and Lightning? I doubt it.

PS: The original plan was for light peak (thunderbolt) to use the USB connector, but the USB consortium squashed that. Funny how they've come around to the idea.

PPS: Based on the earlier rumour, it sounds like Apple will be an early adopter of this. How does this equate to being pwned?
 
These things are getting too complicated nowdays, with Apple getting patents for all sorts of cruff

A display port to usb that WITHOUT an adapter or anything ?? (even an adapter was possible)

They have display port to mini display port, usb to usb,,, whats the problem ?

Its only convenient for end users since they just need one cable.

But now days companies don't care.. it its possible, why the hell not ?

Why didn't Apple do a 30-pin to lightning without adapter required ? That would be possible to.
 
Oh FFS.

Why is it that every rumor on this site attracts ARM fanatics hailing whatever the rumor is about as some sign from on high that ARM is coming to the Mac? I would think they'd have given up by now, or maybe listened to the people explaining for a variety of technical and market reasons why Apple is not going to make ARM Macs.
Not a Mac desktop, but an Mac Server running a headless version of OSX Server for ARM has a lot of sense.
 
Not a Mac desktop, but an Mac Server running a headless version of OSX Server for ARM has a lot of sense.
What sense would that be? To reach the performance of what architecture a Mac mini has today, an new class of ARM processor would be made by or for Apple. It would need support for desktop class (or even server class, with EEC) RAM, PCIe buses, USB, SATA and other peripherals in the chip. And, the CPU cores would have to be clocked at least double, but more likely three times as high to be in performance parity of the current Core i processors in the Mac mini (not to mention next generation). It would have to be a 8 core version, or a new core entirely with hyperthreading (since that doesn't exist on ARM yet). What kinds of power would such a device draw? We can only look at some comparable devices from Freescale, or APM they will be devices with TDP of 30-50 W.
And thats about what we have for the complete system of the current Mac mini.. and we even get graphics in that one!

After that.. all software would have to be recompiled. Key software like VMWare ESXi wont run at all since there is no ARM version of it.
 
What sense would that be? To reach the performance of what architecture a Mac mini has today, an new class of ARM processor would be made by or for Apple. It would need support for desktop class (or even server class, with EEC) RAM, PCIe buses, USB, SATA and other peripherals in the chip. And, the CPU cores would have to be clocked at least double, but more likely three times as high to be in performance parity of the current Core i processors in the Mac mini (not to mention next generation). It would have to be a 8 core version, or a new core entirely with hyperthreading (since that doesn't exist on ARM yet). What kinds of power would such a device draw? We can only look at some comparable devices from Freescale, or APM they will be devices with TDP of 30-50 W.
And thats about what we have for the complete system of the current Mac mini.. and we even get graphics in that one!

After that.. all software would have to be recompiled. Key software like VMWare ESXi wont run at all since there is no ARM version of it.
1. If there are an arm osx server, the software will come soon, recompile to arm on xcode it's near a trivial task.

2. You don't understand Servers, servers require efficient cpu mote than power, right now server farm are moving it's LAMP Servers to arm power savings are so huge and software it's ready and abundant (Linux as OSX it's very portable), Intel cpu outperform ARM on single core performance, it super scaling processing it's deeper, but 4 64 bit arm comes does the work of a single Xeon core, now there are 32 cores x64 arm cpu deployed and 64 & 128 cores versions on the way (that why Intel released a 18 cores Xeon).

3. Apple more than a powerful Mac Server, needs one compact and cheap to compete with the cornucopia of NAS Devices offering almost all the services osx server does (just ask here how many has an Nas at their homes, not all but growing fast the personal home server), so consider there are engineering prototype of Mac running on ARM and the need of Apple to get in the home/soho NAS market where the timecapsule failed miserably, and

4. actually Apple don't need a new cpu an A7 on an appliance it's enough for a headless osx server (headless I mean to say having no display but only Web/console based management), enabling Apple to offer an robust NAS appliance loaded with 2 or more hard drives, and leaving demanding users the choice of an Intel based Mac mini for higher server needs.
 
PS: Stop telling people what they need. That's just smug and you don't know them well enough, let alone their use cases.

I'll do what I want. Why don't you stop telling me what to post? There is a serious problem on this forum where people complain about the prices and availability of professional-grade hardware, and even software. There is no way in hell that a $2000 Thunderbolt RAID solution is for the average person. It just isn't. There is no way that the average person would ever saturate that. I'm not being smug, it's just a fact. I'm not talking about some guy who does video production for fun but isn't a pro. If you do any sort of video production then you know that this stuff is expensive, and you likely aren't going to go on a forum to complain about it to a bunch of average users. You're missing the point that Thunderbolt isn't just a cable type, it's direct access to the motherboard. Just about anything can run on Thunderbolt, including USB 3.0, FireWire, eSATA, ethernet, digital audio, graphics cards. It's just external PCIe, and it's a way for professional users to extend their mobile system when they need to. You do know that the first computer to get Thunderbolt had "Pro" in the name, right? But I guess Apple is being smug by calling it that and pricing it higher than everything else. Sure. And now we have USB starting to copy some of the things that make Thunderbolt great. Right. It just sounds like you're mad because you bought a computer that didn't have the features you needed. But here I go again, telling people what they need because I'm apparently some kind of an a**hole: http://www.bhphotovideo.com/bnh/controller/home?O=&sku=1075056&gclid=CjwKEAjwtIShBRD08fKD1OWSik4SJAAuKLovyLEiazv7gqZ6ye1fyrUWw_0rIoBtrL0prFOur9xOjBoCodLw_wcB&is=REG&Q=&A=details
 
1. If there are an arm osx server, the software will come soon, recompile to arm on xcode it's near a trivial task.
There's only theoretical support for ARM as it stands now. Sore, everything is avery benevolent world is just a compile away and that's real easy.. but in the real world, it's not that simple at all. Look at the support for very mature platforms like PowerPC.. IBM and Freescale tried for years to get the x86-centric Open source community to just "recompile".. but hey, it wasn't that easy at all. And IBM does make servers. One of the problems they faces was that x86 applications does not just recompile to big endian plattforms, and hey.. ARM is big endian. And I'm sure that Apple developers are eager to support another plattform, which would be very low volume and quite high in support. The toolchain is there, but would they jump eagerly to ARM just because Apple tries to release a niche server plattform? Hardly.

2. You don't understand Servers, servers require efficient cpu mote than power, right now server farm are moving it's LAMP Servers to arm power savings are so huge and software it's ready and abundant (Linux as OSX it's very portable), Intel cpu outperform ARM on single core performance, it super scaling processing it's deeper, but 4 64 bit arm comes does the work of a single Xeon core, now there are 32 cores x64 arm cpu deployed and 64 & 128 cores versions on the way (that why Intel released a 18 cores Xeon).
You know nothing about what I know, thank you.
We can see how software vendors are jumping to the opportunity to support AMD's, TI's, APM's and Cavium's ARM plattforms. No, they are not.. some products are there in theory, but hey, most are just prototypes, because its easy to just recompile for prototypes.. So.. Oracle Java? Seems like Java v8 for 64-bit ARM is at least half a year away. Since that was a long time coming, this transition cannot really be considered a simple recompile. And why on earth would Oracle support Apple's tiny tiny platform?
Ohh, and can you get MySQL support on ARM, or even a download? Or should a just recompile it myself?
Dell and HP are really throwing themselves behind their ARM endeavors.. Those Moonshot servers are really flying of the shelves. They must be since the most important thing is power savings. Or.. it might be x86 support. I don't know.. My money is on the latter, in the 1-10 year timeframe. And that's why the Moonshot servers are x86 now, and not ARM.
And if recompile is such an easy thing to do.. why is the software support for mature server plattforms like POWER and Sparc nowhere near x86? And if we are talking performance/watt, single core performance and scale out parallelism.. neither ARM not x86 is near these guys. But still.. they are not particularly successful in the server space.. because of x86 legacy.

3. Apple more than a powerful Mac Server, needs one compact and cheap to compete with the cornucopia of NAS Devices
I'd certainly like to see that, but Apple really doesn't need to. Should Apple go don't this route, they could just pick an off the shelf ARM SoC, and run iOS on it. this is not the server we are talking about here, this is a slightly more capable Time Capsule, that's a far cry from Mac OS X Server..

4. actually Apple don't need a new cpu an A7 on an appliance it's enough for a headless osx server (headless I mean to say having no display but only Web/console based management), enabling Apple to offer an robust NAS appliance loaded with 2 or more hard drives, and leaving demanding users the choice of an Intel based Mac mini for higher server needs.
I thought we were talking about servers.. not NAS. If were talking NAS, I'm with you.. go ahead. I want Apple to open up an iSO based SDK and App Store for a slightly modified Time Capsule and AppleTV. I really don't know why they haven't.
 
Not a Mac desktop, but an Mac Server running a headless version of OSX Server for ARM has a lot of sense.

Apple got out of the XServe business a long time ago. Why would they go back, and if they did, why in the bloody heck would they choose ARM?

You talk about a sort of mini Mac mini, something less powerful for those who need a truly low end server, but Apple is not in the business of selling cheap super-low-end hardware. They can, and do, convince people to buy a full Mac mini all the time. The margins are far more better on a Mac mini than what you describe. Apple can just plug in Intel-made chips on a small underpowered board and go to town. That's a lot easier and fits their business model better than designing some custom ARM processor, porting over all their software to ARM, and then trying to sell it for pennies.

When will you ARM cultists give up?
 
Cool, this sounds like it will be a cheaper option than Thunderbolt for consumer docking stations. Though I wonder what resolutions it will be able to push? (EDIT: seems with DP 1.3 that won't be a problem, nice)

However, to people predicting the demise of Thunderbolt: No, USB and Thunderbolt are overall complementary not competitive technologies. USB is the cheaper and good-enough consumer option for connectivity to external devices. Thunderbolt is a replacement for Express Cards with DisplayPort added on. USB is not a replacement for Thunderbolt because it is not a replacement for PCI-E. The addition of DisplayPort on Thunderbolt was meant to drive adoption so that makers using ExpressCards would use Thunderbolt instead, not so that USB would be replaced. I'm sure Intel had hopes that Thunderbolt would also get wider adoption beyond that market, but that hasn't happened too much and indeed even replacing ExpressCards has been slower than they would've liked. However, connection adoption tends to be at glacial pace (the slow adoption of USB 3.0 from 2.0 is a case in point and that's on consumer products with higher turnover rates).

While Intel did make some mistakes that slowed adoption further, Apple exclusivity wasn't one of them - that was merely a rumor that got tossed around that Apple would have 1 year exclusivity. Both Intel and Apple denied it and indeed PC makers came out with machines with Thunderbolt ports long before that. Apple simply released devices with Thunderbolt soon after announcement as they helped bring it to market.

The Type-C connector is also capable of using PCIe signaling as an Alternate Mode, so it too is a completely viable replacement for ExpressCard. It would be great if that becomes standardized soon as well.

In general, you're right though. Thunderbolt is a honking big data pump that provides much higher serial I/O bandwidth than any of it's contemporaries, combined with protocol flexibility. There's a big difference between a physical layer that supports multiple signaling modes and a high-speed serial link that can transport multiple protocols simultaneously.

However, Apple absolutely did have an exclusive on Thunderbolt for the entirety of the Sandy Bridge era. The only exception was Sony, who were allowed to ship the VAIO Z and Power Media Dock using Intel Thunderbolt controllers, but only in conjunction with a proprietary USB Type-A + optical interface akin to those used by the Light Peak prototypes. An Anandtech article from shortly after the embargo lifted can be found here. This was not a rumor, and was never denied by Apple or Intel.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.