Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Leaked Info on Third-Generation Thunderbolt Points to 40Gbps Transfer Speeds
- Thread starter MacRumors
- Start date
- Sort by reaction score
I think you may be right in that USB 3.0 will simply be another signaling mode of the new connector, rather than being transported over Thunderbolt via a protocol adapter. The new power handling capabilities are probably based on the USB Power Delivery Specification, and the connector might be a derivative of the proposed USB Type-C connector. Otherwise I would be surprised that Intel would even allow the Thunderbolt controller package size to be increased by the addition of connections for USB.
Since Thunderbolt ports were always required to support DisplayPort Dual-Mode, and Thunderbolt 2 already supports DP 1.2 and High-Speed HDMI 1.4 via DP Dual-Mode 1.1, the HDMI 2.0 part isn't that surprising. Likely Alpine Ridge will just support DP Dual-Mode 1.2 or whatever rolls HDMI 2.0 into the mix. It would be really nice if the DP to Dual-Mode DP redriver was finally included on the Thunderbolt controller die though.
Thanks for pointing this out as wellThunderbolt has always been "faster" than PCIe, it just isn't designed to be as wide. PCIe 2.0 is 4 Gbit/s and PCIe 3.0 is 7.877 Gbit/s per lane (without including encoding overhead) but scale to x16 lanes vs. Thunderbolt at 10 Gbit/s per channel with a single cable carrying two channels, but only Thunderbolt 2 being able to bond them together as an x2 link.
If you're discussing the PCIe transport capabilities of Thunderbolt though, the controllers will continue to be limited by the back end, which probably won't go past x4 any time soon due to the costs associated with scaling up the on-die PCIe switch, protocol adapters, Thunderbolt crossbar switch, and package to accommodate connections for more PCIe lanes. That means we're stuck at 16 Gbit/s of PCIe throughput for Falcon Ridge, and 31.5 Gbit/s for Alpine Ridge. In reality, that's probably just fine though. It's not like the platforms that are likely to be graced with Thunderbolt have a ton of PCIe capacity to spare.
Thunderbolt does impose a pretty hefty PCIe latency penalty though, something like 1.5 microseconds per hop, or 9 μs of round-trip latency for the last device in a 6 device chain. This is much worse than typical PCIe switches which only induce maybe a couple hundred ns of latency.
OK, sorry, totally missed the rather valid point that you were calling out the sequential write limitations in particular. I think the iPhone 5s might be able to do close to 40 MB/s, but you're right, still hardly worth switching to USB 3.0 for.
Although I'm probably not the typical user here, that might have gone over my head because I generally plug my phone into my PC to back it up or transfer photos / videos to the PC, so I'm always supremely annoyed that I can't get much better than 25 MB/s due to feeble USB 2.0.
I was hoping for a new connector type to FIX the current one, which is too loose. It should have some type of "locking" mechanism. Many engineers have told me it would have been easy to add in a "lock" but I wonder why Intel didn't do it?? Right now, it's way too easy to come off.
God, ANOTHER new connector type! Thunderbolt is so recent, Intel really couldn't foresee a "~3mm" slimmer FINAL and PERMANENT (and why the hell not REVERSIBLE) connector from the get-go?!?!
Dudes, I never said Thunderbolt is PCIe. If I did, I'd say that instead I said it is based on PCI-e, as in that's its backend. As in there's a PCIe SSD, PCIe GFC card. They all use PCI-E as its backend. If based on PCIe is not the proper term, please let me know how to say it right. From now on I'll say TB currently depends on PCIe to carry its data through to PC, it cannot be faster than PCIe.
Thunderbold intermixes the PCI-E and DP lanes into the same controller, so that it can carry both traffic via its cable into its controller and that gets pushed into the PCI-E lanes on the motherboard.
Thunderbolt does not have a special line path to the CPU.
Here's the graphic from Intel itself:
Unless Thunderbolt has direct paths to the CPU, it CANNOT go faster than the entire bandwidth of PCI-E. Thus, the point I was trying to make.
Stop being lazy and watch the keynote. Wireless sync was announced BECAUSE nobody was doing wired sync even BEFORE there was an alternative. That's right, people would rather not sync at all than do wired sync and that was back in 2011. There is nobody left doing wired sync three years later except for the odd loudmouth.
I have a career as a freelancer in the creative field and handle large files and large volumes of files on a regular basis and FireWire 800 is only so fast these days.
Considering that I'm a working professional and not just some average tech geek, yeah, I do need Thunderbolt.
What I was trying to get at is the fact that Thunderbolt accessories were very expensive when they were first released and remain so to this day. There are freelancers like myself that need to own this type of hardware in order to keep up with the rest of the field but have to make decisions whether it makes financial sense to make this type of investment at the moment.
Thunderbolt 3 is nice but the Thunderbolt 1 ports in my 2011 iMac are still virgins.
The get something for it like the CalDigitThunderbolt Station. This way you can add another cheap monitor for a second monitor.
Last edited:
That $150 Belkin dock is most likely the first and last piece of thunderbolt accessory I would buy for personal use. I just need it to extend my mid-2011 MacBook Air with USB3 ports and gigabit ethernet. My next Mac probably won't even need this dock at all..
Yeah, that's a good use of my time considering now you're changing what you thought it said.
So let's get this straight. At that point wireless sync didn't exist. So millions of people were buying iPods and other iOS devices. And nobody was doing wired sync...which meant that since that was the only way to sync, all of those millions of people were walking around listening to iPods with no music loaded on them.
Yeah, I really wish I would have wasted my time watching a whole keynote just to confirm that you are an idiot who pulled statistics out of your ass.
This isn't about other people being lazy, this is about you making claims with nothing to back them up. If you really think there was something in that keynote that makes your point then provide a direct quote. It's time to put up or shut up.
Wish I knew about this one before I bought my Belkin. I only used it a few times to transfer large files to my USB3 drives..
How. Have I changed what was said? Jobs pointed out that most people never sync after the day they buy their device and never back up or load additional content other than what can be bought on the device, nor do they ever take their photos off their device. He had a problem with that, so he announced iCloud backup, wireless sync, etc. to make life easier for those who do and encourage those who don't. As for wires, they have no future and lighting will be the last wired connector for iOS.
You should try using IP over Thunderbolt for that. Connect two Macs with a Thunderbolt cable, enable File Sharing in System Preferences > Sharing, and thats it. (If both Macs are simultaneously connected via a Wi-Fi network, make sure that Thunderbolt Bridge is above Wi-Fi in System Preferences > Network; if not, drag it higher.)
IP over Thunderbolt will soon be available for Mac-to-PC and PC-to-PC Thunderbolt connections as well: http://www.engadget.com/2014/04/07/thunderbolt-2-networking/ .
You keep insisting that 99% never use the cable with nothing to back it up, now the best you can do is "most". And you still haven't given the Jobs quote from WWDC 2011, not sure if it's because you're lazy or because he didn't say it.
My guess is that you're thinking of the report that came out just before WWDC 2011 about iPhone users not syncing. And that report said the number who don't sync is less than half. But that was almost three years ago, since you didn't bother to look it up before spouting off I'm not surprised you didn't really remember the details.
You totally have the opportunity to prove me wrong and make me look like a moron, you just need to post that Jobs quote. But you won't. Because he didn't say it. I know, but the bad man is so lazy and won't look it up himself (pounds tiny fists on the ground!!!!).
As for wires having no future, one thing that will be required for that to happen is either for iPods to have wireless functionality added or for Apple to discontinue the iPod completely. Neither seems likely to happen any time soon.
It's also kind of hilarious to say that Apple thought the cable was dead in 2011...and then 15 months later they went to the trouble of creating a new standard of cable for those very same devices. Yeah, that makes total sense for something that nobody was using.
I see your point. But, it is always better for the technology to improve. Thunderbolt 3 is backwards compatible. So you should still be ok with it.
Because USB 2.0 did get stuck for 10 years, it created a room for other technologies, like Thunderbolts to come in and take over a portion of pie.
Don't blame the world for spinning too fast. Blame yourself for not being able to catch up.
Are you sure? I thought I read an article that said retina TB display won't come yet because 2880 x 5120 at 60 Hz is just over TB2's capabilities.
----------
That's just awful. Why are prices so high still for it? I always assumed flash was fast on ios devices because of how speedy and how fast apps open on them. Do you know if there is going to be an improvement in the near future on phone/tablet storage speed?
I prefer to let you dig yourself a little deeper first. Let's here some more silliness along the lines of your second last paragraph. The iPod is dead, you must be the only person who didn't hear? Apple stopped updating it some time ago now.
You're referring to latency, not performance, they're not the same thing.
You can have both hard drive and flash drive perform at maximum of 25MBps but the Flash drive get there sooner for you. Opening app quicker is usually about latency but if both apps starts transferring data, they'll be at the same speed.
An analog would be two cars, a Ford Focus and a Porsche. Let's say you put a limiter on the Porsche at 80MPH. Both cars can go up to 80MPH but you will feel the acceleration faster in Porsche than Focus, that's what the latency is about.
However in the end, both cars will be going at 80MPH with no differences between then except when you break, turn corners and speed back up. Again, that's the latency, not the performance.
The performance in the mobile device's flash chips are increasing every year, they'll get better for sure. Sandisk and Toshiba have working in this area and they plan to have something out later this year or next year that should boost the performance beyond 100MBps in mobile devices.
The question is, can those chips perform at an acceptable power and price levels.
You must be talking about SLR camera lenses. The lower limit on cost of a top-quality lens is fairly high no matter how many you make.
Actually, when you are talking about silicon, mass produced and cheaper go together-- it is that simple. Build millions of supercomputers on a chip, and, they can be had for $500-$2000, or less, depending. Now, we build supercomputers out of those cheap chips. And you carry around a supercomputer in your pocket and call it a cell phone.
The average consumer, or, a higher-end, "prosumer" consumer? Grandma may not have that big an appetite for I/O, but, even people using Aperture and storing and backing up tens of thousands of pictures need I/O. Oh, and Grandma probably has a iPad and may be doing more computing than you realize.
Experience shows that both amateurs and professionals have an appetite for I/O. The professionals can pay more for it, but, just like with higher-end CPUs, both amateurs and professionals benefit if they can share peripherals and enjoy economies of scale. Artificial market differentiation doesn't benefit anyone.
It's not gonna happen this year. The leaked charts themselves show TB3 with Skylake which is after Broadwell. Right now Intel's saying the manufacturing ramp of Skylake will begin in 2H2015 and that's if there aren't any delays. So I wouldn't expect Skylake, and by extension, TB3, until the end of 2015 or early 2016.
Strange. I just backed up and synced today, using USB. I never thought I was so special over such a thing.
Had mine a few years. Thanks to the Thunderbolt port, I upgraded to an SSD boot drive that runs at near internal-drive speeds without having to open the iMac
I did the same, very little overhead. I'm really excited to upgrade the ssd in my wife's 2013 air soon (whenever OWC brings to market) and then my boot drive will be flying at pcie ssd speeds! Try that with USB 3 (I did, the ssd performance was less than half and you're chewing through CPU cycles)
You are correct. 5120 x 2880, @ 60 Hz, 24 bpp would require at least 22.52 Gbit/s, which is more than a single DisplayPort 1.2 main link or Thunderbolt cable can handle. The DSL5520 Thunderbolt 2 controller has 2 DP 1.2 sink adapters though, so it could drive such a display as two 2560 x 2880 tiles (or four 2560 x 1440 tiles using MST), but that would require at least 23.2 Gbit/s and the use of two Thunderbolt ports / cables.
I kind of jumped all over pgiguere1 for that comment earlier because I was afraid people would have the same reaction that you are. The NAND used in smartphones is often quite good, it's just optimized for different aspects of performance. For instance, small random read performance might be 30x better than that of the fastest HDDs, thus the quick boot times and app launching.
SSDs achieve their crazy performance through parallelism, often using 8 or more channels, not intrinsically faster NAND. That type of design is currently not really feasible for ultra mobile devices where the whole storage system is packaged in a single chip. Through Silicon Vias (TSVs) and more advanced die stacking techniques may change that in the near future though.
As much as I like car analogies, I think this one has some issues. Mostly because although we take throughput in MB/s as a measure of storage performance, if we look closely at any given time interval, data is generally either flowing at close to the maximum rate allowed by the media or interface, or not at all. This is why we look at max sequential read and write speeds. The other corners of the performance equation are small random IOs, which introduce the most latency, or time spent not delivering any data, because they require the most overhead to service. Solid state will generally always beat mechanical spinning disks when it comes to small random IOPS, and can leverage parallelism much more readily to boot—it just costs way more per GB at this point.
Thunderbolt is a high-speed, packet based, serial I/O interface. A 4-channel Thunderbolt controller has a PCIe 2.0 back end (a 4-lane, full-duplex serial interface operating at 5 GT/s with 8b/10b encoding), two DP sink connections, and one DP source connection (each with 4-lane, simplex main links operating at up to 5.4 Gbit/s with 8b/10b encoding). Protocol adapters in the controller deserialize the data on each interface and forward the packets via a crossbar switch to the next hop. The packets are reserialized at 10.3125 Gbit/s with 64b/66b encoding before being sent over one of the channels comprising a Thunderbolt link.
The front facing Thunderbolt channels offer considerably more bandwidth than just the PCIe back end, but once you add DisplayPort into the mix, the back end swells to a potential 50.56 Gbit/s in and 33.28 Gbit/s out (or even more if you include DP AUX channels).
In a host situation, the DP sink connections are generally provided by digital display outputs from either an integrated or discrete GPU. The diagram you provided is slightly outdated due to changes in the role of the PCH for certain Haswell platforms, but the idea is still more or less the same. In practice (thanks to Apple), the majority of Thunderbolt controllers are actually connected via PEG lanes which come directly from the CPU instead of from the PCH. So with some Haswell systems, both the PCIe and DP back ends for the Thunderbolt controller are provided by direct connections to the CPU, and the bandwidth linking the controller to the CPU is equivalent to or greater than that linking the CPU to the PCH. However, just to clarify, DP packets are never transported via PCIe.
Edit: Of course you are correct in that the PCIe bandwidth of a system is ultimately determined by the aggregate PCIe / DMI bandwidth of the CPU itself these days. In the case of recent desktop and mobile CPUs from Intel, that would be PCIe 3.0 x16 (the PEG lanes) plus DMI 2.0 x4 (which is essentially PCIe 2.0 x4), or 142 Gbit/s after accounting for encoding. For the enthusiast and workstation SKUs it would be PCIe 3.0 x40, plus DMI 2.0 x4, or 331 Gbit/s. So even Alpine Ridge will only be able to provide 22% of a current system's total PCIe bandwidth over a single cable, with just one tiny little connector...
Last edited:
Correct but,am i wronging or TB2 came a little before it was stated to arrive?
Also,not every today Macs have TB2 yet isn'it?
Problem with that is that it will take 10 years for the technology to improve to the point where they can realise the 100Mb per channel in anything like a consumer usable form. This is why they said it would initially be 10Mb per channel and quickly ramp up to 100Mb as the tech is perfected.