Back to Article

  • zanon - Friday, May 11, 2012 - link

    Back during your CES 2012 coverage in January you reported on one of the initial planned external expansion boxes, the MSI GUS II. Things like that (I know there are a few others now) seem to be where the story gets truly interesting for mobile users IMO. Even though the bandwidth available is only equivalent to a few lanes, testing like HardOCP's 480 x16 vs x4 article indicates that at single screen resolutions (ie., no more then 2560x1600) graphics cards can perform shockingly well even with severely restricted bandwidth. So more then merely having a hub there's the potential of being able to plug an ultrabook into a hub and have a mid-range full GPU ready to go, along with a screen and other ports. If it all goes smoothly it could really expand the desirability of ultrabooks even farther, making them more and more no-compromise, although it'll probably take the next-gen 40/100 Gbps TB standard to push it farther or handle GPGPU applications. Still exciting stuff. Reply
  • Guspaz - Friday, May 11, 2012 - link

    Sadly, the MSI GUS II is not that useful due to only supporting bus power (75W, no external power connectors supported). The fastest (nVidia) card it can support is the GT 640, which isn't that much faster than the integrated graphics it would be intended to replace. Anything faster can't be bus-powered.

    It would give you a small graphics upgrade over the Ivy Bridge integrated graphics, but not a big enough one to warrant all the cost and effort... and a notebook with discrete graphics could easily outperform it.

    If they add support for externally powered GPUs (anything that draws over 75W), then it could be something special.
  • zanon - Friday, May 11, 2012 - link

    If they add support for externally powered GPUs (anything that draws over 75W), then it could be something special.

    That's extremely trivial though, I was only using it as an Anandtech featured example of one initial expansion, not as the ideal solution itself. Not exactly a big deal to stick a tiny external power source in there, in fact it's odd MSI didn't do it in the first place (or who knows, maybe it'll get revised before release), but as the chips themselves become cheaper and get plenty of supply it seems likely other solutions will appear. It feels like one of the real possible killer apps for the interface after all, something that can't be easily replicated through other means.
  • DerPuppy - Friday, May 11, 2012 - link

    they are (villagetronic+others in the egpu community) working on a thunderbolt system..supposedly development support is limited to large partners though because the thunderbolt team at intel is overworked? someday it'll come out, external gpu solutions on >x1 PCIe2.0 connections, that is. the power issue is somewhat of a nonissue if you don't mind bringing around any sort of 12v psu, be it normal atx, sfx, or w/e and just hooking it up Reply
  • Roland00Address - Friday, May 11, 2012 - link

    The product tag says the device can support up to 150w cards. See product tag picture here.

    I also have heard it has been reported there is no 6 pin power thus the device can only do 75w.

    If the device can do 150w you can put a 7850 in there, or possibly a gtx 660 (the gtx660 is a personal guess since the gtx 670 tdp was 170w and to do the gpu boost on the gtx 670 the tdp has to be 141w.)
  • yyrkoon - Saturday, May 12, 2012 - link

    Thing is though, even more important than power. Is the technology bandwidth.

    10 Gbps (one-direction ) really is not that much in the way of graphics card bandwidth. As I recall. a 7600GT from a few years back, could chew through about 20Gbyte /s under intensive situations. So 10 Gbit/s is hardly going to put a dent into that. So with that in mind, we're basically stuck with integrated graphics performance again.

    That is, at least until the next iteration of the technology. Maybe.

    Still, from a modular system approach, I like the idea. However I doubt it would be practical any time soon ( like you pretty much already said ).
  • DerPuppy - Saturday, May 12, 2012 - link

    No idea where you're getting these metrics from? PCIe 3.0 x16 is 16GB/s so there's no way a 7600GT could have used that much bandwidth. maybe you're referring to the internal memory bandwidth?? It has been proven that an x4 2.0 connection is sufficient for about ~80-90% of the performance of an x16 link in many gaming situations Reply
  • yyrkoon - Sunday, May 13, 2012 - link

    Of course it was the memory bandwidth.

    " It has been proven that an x4 2.0 connection is sufficient for about ~80-90% of the performance of an x16 link in many gaming situations"

    You would be very lucky to see half that. It is very likely, that you would see 25% or less of that.

    People have been working on this problem for years now. Through other means. Partially, they have succeeded, using the MXM laptop graphics connection. At a cost that begs to wonder why they did not just buy a $2000 laptop to begin with.

    You can buy an external graphics enclosure, for laptops, right now. If you're willing to spend ~$800 for it. Then, only if you have the right laptop.

    So in ending I will say this. You're dreaming. You're dreaming a dream I have had myself. At some point however, you're going to have to come back to reality.

    Oh, one last thing. I should point out that Gigabytes, and Gigabits are not universal with one another. 10 Gigabit == 1.25 Gigabyte. That is, 25% more, than a PCIe 1x 1.0 connection can handle. Under ideal circumstances.

    How do you like them metrics ?
  • repoman27 - Sunday, May 13, 2012 - link

    Well, connecting a GPU via Thunderbolt doesn't affect the memory bandwidth, so that's not really relevant.

    Plus your math is way off. A single Thunderbolt controller can provide 10 Gbps of PCIe bandwidth. A PCIe 1.0 x1 connection provides 250 MB/s (i.e. 2 Gbps) which is 1/5 of what you get from Thunderbolt. Or put another way, Thunderbolt can currently provide the equivalent of a PCIe 2.0 x2.5 connection.

    DerPuppy is correct. If driver support with PCIe compression for Thunderbolt connected GPU's was available, we could achieve better than 80% of the real-world performance of a PCIe 3.0 x16 connected GPU with an external solution.

    Note the x2 (2 GB/s) performance in these charts:
  • Jaybus - Monday, May 14, 2012 - link

    Yes, more bandwidth is needed to support external high end GPUs, not to mention the simultaneous use of multiple external PCIe devices. This is why the initial Intel project was named Light Peak. Intel Labs silicon photonics researchers never intended the interface to use an electronic PHY. I believe the electronic PHY version (Thunderbolt) was due to Apple's collaboration along with Intel hitting snags in the development of an on-chip optical PHY.. Ultimately, there will be an optical PHY, since the ability to scale the electronic PHY is limited. Reply
  • repoman27 - Monday, May 14, 2012 - link

    This is a persistent misconception regarding Thunderbolt—that it would somehow be able to offer greater bandwidth if it had only used optical media.

    Light Peak was developed from the outset with a target link rate of 10 Gbps (and using conventional fiber optic technology, not silicon photonics.) While optical cables have much less of a problem with signal attenuation than copper, which makes them more efficient for longer runs, when it comes to distances of less than 10 ft, which are the norm for typical PC use cases, copper is currently much cheaper, more power efficient, and user friendly.

    Look around any data center and you'll most likely see lots of direct attach twinax copper for short runs. The highest bandwidth optical interconnects I'm aware of are rated for 56 Gbps (4x 14 Gbps lanes, full-duplex), but you can buy 3.0 m passive copper cables that can do that as well and cost way less.

    Since you'd need a PCIe 3.0 x8 connection to fill the largest pipes we currently have at our disposal, I don't think copper's ability to scale is a real issue yet.
  • yyrkoon - Monday, May 14, 2012 - link

    I stand corrected. For some reason I was thinking the bandwidth provided for each PCIe lane was half of what is indicated in the charts you provided a link to.

    Sorry about that - everyone.

    I do not think anyone here would like to see this come into fruition more than myself. Having been let down by the industry many times in the past however. I just do not think it will ever happen. So I suppose I tend to get worked up over it.
  • BrightCandle - Saturday, May 12, 2012 - link

    What I really want to see is 300W capable external cases for GPUs that support thunderbolt. Then we can have laptops that can be connected to modern high end GPUs that can game well when connected up. It also really needs to be able to drive the laptops screen. Reply
  • JKflipflop98 - Sunday, May 13, 2012 - link

    That would be cool. Maybe we'll start seeing monitors with built in video card expansion slots or something. Reply
  • RamarC - Friday, May 11, 2012 - link

    VGA and a PS/2 mouse/keyboard connector on a mobo with Thunderbolt? echos of 1988 still resound in the most modern of hardware... Reply
  • hechacker1 - Friday, May 11, 2012 - link

    PS/2 is still best for n-key rollover, and it's an interrupt device rather than polling.

    But vga, yeah, that seems like a waste of a port, it should be dvi and just include the dongle.
  • DerPuppy - Friday, May 11, 2012 - link

    PS/2 keyboard connectors still have advantages over usb keyboard connections...
    vga is legacy though, no doubt
  • tjoynt - Friday, May 11, 2012 - link

    For home computers ps2 and vga don't make sense anymore, but for businesses (especially data centers) ps2 and vga are critical. USB + DVI KVM switches for more than 4 computers get *very* expensive, while 16-32 port ps2+vga KVMs are rather inexpensive and rock solid reliable. Reply
  • Zoomer - Friday, May 11, 2012 - link

    They could've included a DVI-I port and included a dongle, as some said. It's possible that the cost of utilizing the space freed up was too much; or there was a lack of suitable ports besides serial that are available to use that space. Reply
  • Exodite - Friday, May 11, 2012 - link

    I suppose I'm the one that has to go against the grain here then but I'd consider VGA more useful than PS/2 as I still rely on a fair few projectors that use VGA.

    Granted, it's an old and dying standard but personally I'd find a lot more use for VGA than I would PS/2, Firewire or, uh, Thunderbolt really. :P
  • DerPuppy - Friday, May 11, 2012 - link

    lots of businesses might say the same about IE6...not helping your case lol Reply
  • Exodite - Friday, May 11, 2012 - link

    That's really not relevant.

    Any computer capable of running IE6 can run a more modern browser as well.

    A VGA projector won't grow the digital connector of your choice through a firmware upgrade, however.
  • DerPuppy - Saturday, May 12, 2012 - link

    you can have adapters just the same way you can have more modern browsers Reply
  • Zoomer - Friday, May 11, 2012 - link

    And you'll be carrying a desktop around to said offices using these vga ports for your presentations? Reply
  • MrPickins - Saturday, May 12, 2012 - link

    As others have mentioned, DVI-I and DVI-to-VGA adapter would have worked for older displays. Reply
  • DerPuppy - Friday, May 11, 2012 - link

    Question regarding requiring reboot upon adding thunderbolt devices. people have been adding external pcie devices via expresscard etc. (see: for forever. supposedly you can reconnect a disconnected egpu just by sleeping-> plugging-> waking a PC...would that work for thunderbolt? Reply
  • cjb110 - Friday, May 11, 2012 - link

    Using thunderbolt you could put the slower advancing stuff into monitors (which also have stagnated at 1080p)

    So my gigabit network, large mechanical drives, wifi, bt etc is all in the monitor along with the usual periphery stuff connected printers, card readers etc.

    My desktop then has the cpu/gpu and a much much simpler motherboard, no network, no need for lots of slow booting sata controllers, its the cpu, memory, gpu and a boot ssd.

    A Alienware x51 type pc with a thunderbolt 1080 40/50" display would make a 'pc' console that would also be easy to maintain.
  • Zoomer - Friday, May 11, 2012 - link

    Is that integration necessary? It seems like it'll be out of date soon. Would you want to replace your whole monitor just to get 11ac, or bt 3.0 lp? Reply
  • DerPuppy - Saturday, May 12, 2012 - link

    well considering it's basically porting over PCIe...if you had mini PCIe connectors in your monitor you could "upgrade" your "dock", so to speak Reply
  • Metaluna - Friday, May 11, 2012 - link

    It's good to see more Thunderbolt options showing up in the PC market, but it seems to me that a full-sized ATX motherboard that already has 7 PCIe slots is about the last place anyone would need TB.

    Totally unrelated topic: Is it just me or does anyone else find this whole "Military Grade" thing that has been showing up on motherboards lately to be one of the most embarassingly cheesy marketing gimmicks ever?
  • sphigel - Friday, May 11, 2012 - link

    "does anyone else find this whole "Military Grade" thing that has been showing up on motherboards lately to be one of the most embarassingly cheesy marketing gimmicks ever?"

    It's been around for at least 10 years that I know of but, yes, it is very cheesy.
  • A5 - Friday, May 11, 2012 - link

    Military/defense grade is a real thing when dealing with component rating, though. It makes absolutely zero difference (besides the cost) on consumer-grade stuff, but it is a different set of specifications. Reply
  • jabber - Saturday, May 12, 2012 - link

    Well if there were real 'military grade' they would cost $5000 each as you'd have £4900 worth of 'recertification costs' over the standard OEM issue the defence contractor buys in and then stamps 'military grade'. Reply
  • StormyParis - Friday, May 11, 2012 - link

    It's nice to see CPU usage... 20% seems high, but I guess for high throughput it's bearable, but what CPU was in your setup ?

    Did you run into any compatibility problems. I'm still having those with USB, both 2 and 3.. which is very disappointing .
  • Death666Angel - Monday, May 14, 2012 - link

    He said it in the Youtube video: Core i7 3770K. :-) Reply
  • Stanly.ok - Friday, May 11, 2012 - link

    "pushing the Thunderbolt Display at its native 2560 x 1600 resolution"

    "drive the Thunderbolt Display at 2560 x 1600"

    looks like you miss 30" Cinema Display a lot ((=
  • coder543 - Friday, May 11, 2012 - link

    This review was really unfair in my opinion. This is one of the first reviews of a non-Apple usage of Thunderbolt, and we don't even try Linux on it? You spent pages reiterating over and over and over about how Windows doesn't support hot plugging of those devices.

    And you didn't see what Ubuntu (or Fedora, or whatever else) did when hotplugging? Come on.

    slightly frustrated at the complete absence of Linux on Anandtech.
  • Metaluna - Friday, May 11, 2012 - link

    Yeah, this is one of the problems with Thunderbolt that doesn't get a lot of coverage, though kudos to Anand for touching on it in this article.

    The problem I'm referring to is that TB just pushes the driver availability problem out to the external device, which is particularly problematic for storage devices. Want to take that nice TB external drive and move it between OSX, Windows, Linux, FreeBSD, etc? Well, now you not only have to make sure all those machines have a TB port and a compatible filesystem, you also have to make sure they all have a driver for whatever controller chip the enclosure has in it. With USB, eSATA, SAS etc, each of those machines can just have whatever controller works best for that OS.
  • Zoomer - Friday, May 11, 2012 - link

    It's a PCIE bridge chip on the phy layer, so it should be transparent to the OS as mentioned.

    The driver issue would be the same as if the external device was plugged into an internal PCIE slot.
  • Klimax - Saturday, May 12, 2012 - link

    One thing: Windows support hotplugging(it doesn't care what type of device it is in general), but drivers have to support it too and if they don't provide proper callbacks to I/O manager then hotplug cannot work as IOM has no way to inform driver about changes. Reply
  • ananduser - Friday, May 11, 2012 - link

    I recall MS stating that TB will never be supported by them due to security issues. TB support will have to be provided by 3rd parties. Reply
  • A5 - Friday, May 11, 2012 - link

    And the article pretty clearly states that TB is transparent to the OS because it is just a bunch of PCIe lanes. MS doesn't really need to do anything to enhance support. Reply
  • embeddedbill - Friday, May 11, 2012 - link

    The security issues have nothing to do with 3rd party support. Pcie in every other instance was an internal device, now via TB, someone can develop a piece of hardware that has nefarious firmware and gain system level access - without opening up the computer. Reply
  • DerPuppy - Saturday, May 12, 2012 - link

    pretty much: see firewire. direct memory access "issues" i guess. but if someone has physical access to your computer you're screwed anyways. Reply
  • ka_ - Friday, June 01, 2012 - link

    "pretty much: see firewire. direct memory access "issues" i guess. but if someone has physical access to your computer you're screwed anyways. "

    I must say I disagree to this to a large extent even though I would agree if we add: Physical access without anyone else present.

    It is something entirely different to require someone to force power offs, use screw drivers, or do some other highly unusual things to a machine than simply plug it to a display or NAS or something with two tb connectors and hack it from the machine connected on the other side. That or a device with chained TB connectors i.e. projector sharing the connection with a Trojan device.

    Expect to see something like this on a security conference soon... "We tested this projector on all the presenters machines and..."
  • embeddedbill - Friday, May 11, 2012 - link

    Microsoft will also have an aversion to anything that allows you to boot using Windows on an external disk. Pcie is a threat to that because they might feel some loss of control. Reply
  • dagamer34 - Saturday, May 12, 2012 - link

    Em... Windows 8 To Go allows you to boot a full version of Windows from a USB stick. Reply
  • Conficio - Friday, May 11, 2012 - link

    And why again would I pay top dollars (for the mother board, the peripheral and the cables to connect those) to put the storage outside of my full size desktop? I don't see the use case.

    If it would be a nice small form factor running silently and still is expandable through thunderbolt, that would make sense.

    Or if I can access the storage and the display in my desktop through thunderbold from my laptop (and auto backup/sync/drop box) that would make sense as well. The standby desktop becoming the hub, sounds feasible to me as the interface is symmetrical.

    By the way w/o hot plugging this is broken. As you report those issue on Mac OS X as well, I'll go and wait until this becomes stable.
  • embeddedbill - Friday, May 11, 2012 - link

    The most compelling use cases will be in the mobile market for sure. However for someone who covets the bandwidth for frivolous stuff, I can see this in a niche. Reply
  • ggathagan - Friday, May 11, 2012 - link

    Given the ability to daisy chain up to six devices, this could represent an awful lot of storage capacity.

    From the manufacturer's standpoint, I think it makes a lot of sense for MSI to use ATX for their 1st motherboard with TB.
    ATX has lot more real estate to work with when deciding how to design the motherboard.

    It's also their flagship Z77 motherboard, with all the bells-and-whistles.
    At the price point of their flagship board, it helps recoup R&D costs.

    Given the scarcity of TB devices at this point in time, I doubt anyone wants to dive in.
    I expect more of a "sticking-my-toes-in-the-water approach from board makers.

    Once there are more devices made to utilize TB, I think you'll see a more widespread adoption of TB.

    As others have pointed out, TB has a better fit in the mobile market at this point.
  • Zoomer - Friday, May 11, 2012 - link

    Yeah, imagine 3TB drives on sata controllers on each of these. That's like 24 * 3TB = 72 TB. And that's without resorting to PCIE switches or SATA port multiplied. Reply
  • DanaGoyette - Saturday, May 12, 2012 - link

    The HP Microserver is one device that could make Thunderbolt extremely useful. It has 4 drive bays (and 2 more SATA ports), but only has two low-profile PCIe slots.

    I'm using one of the slots for the remote access card, and the other one for a serial port card, of all things, because the remote access card is actually quite buggy.

    If the Microserver had a Thunderbolt port, you could chain massive storage off the thing.

    Now, what happens if you try to connect two computers together via Thunderbolt?

    As for hotplug, I'd imagine the BIOS just needs to properly mark that PCIe port hotpluggable -- even Apple's own implementation doesn't do that properly.
  • dagamer34 - Saturday, May 12, 2012 - link

    External SSD RAIDs that can easily push 1GB/sec (yes, that's gigaBTYE) which, with modern SSDs only requires 2 high end drives at this point.

    And while the 10Gbps v1.0 product is rather limiting, when you get to 100Gbps, that's when Thunderbolt will really shine.
  • repoman27 - Sunday, May 13, 2012 - link

    How about if you were building a box for video production and wanted to use one of the Thunderbolt interfaces available from AJA, BlackMagic or Matrox?

    Although many of the Thunderbolt products that came to market in the past year were storage devices, Thunderbolt really isn't about external storage except for corner cases. You don't look at your USB ports and think that they're just there as a way to attach external disk drives, do you?
  • embeddedbill - Friday, May 11, 2012 - link

    I'm curious, which disk was the windows OS installed on, how did that work out? Reply
  • apspeedbump - Friday, May 11, 2012 - link

    Was an attempt made to make Windows re-initialize a hot swapped device with a utility like "Hotswap!"?

    It's worked, under Win 7 as well, for my computer to get drives I've plugged into external Sata to get recognized.

    Just wondering it that's a feasible workaround until the drivers get certified.
  • peterfares - Friday, May 11, 2012 - link

    I still don't understand why they needed to merge PCIe and DisplayPort, at least the way they did. If they kept everything separate, that's WAY less thunderbolt controllers everyone has to buy. If they really needed to combine with a video connector, couldn't they have just added some pins to DisplayPort or made a new connector that has both DisplayPort pins and PCIe pins? That way you could also have devices that only need the PCIe part of it without the video.

    Just seems like a scheme by Intel and Apple to sell unnecessary chips.
  • ggathagan - Friday, May 11, 2012 - link

    "If they really needed to combine with a video connector, couldn't they have just added some pins to DisplayPort or made a new connector that has both DisplayPort pins and PCIe pins?"

    Maybe I'm not understanding your comment, but it seems very contradictory to me.
    You've just described the TB connector and then have asked why someone hasn't made it.

    There is only one TB contoller in each device (motherboard, display, hard drive enclosure).
    You can run any protocol on it.

    The fact that you can run both PCIe and DP through a single connector is a positive thing.
    Why is simplifying down to a single connection for video, network, USB and firewire bad in your eyes?
  • repoman27 - Sunday, May 13, 2012 - link

    Mini DisplayPort connectors already pack 20 pins in a 33 mm^2 cross section. That's pretty dense, there's not really room for adding more pins. Plus, by keeping the same physical connector as mini-DP, it's easier to maintain backwards compatibility with existing DisplayPort gear. Besides, more pins in the connector generally means more conductors in the cable, and Thunderbolt cables are already complex and expensive enough as is.

    Apple was already building Macs with this tiny little mini-DP connector that worked quite well for the form-factors they were designing, and it was capable of pushing 17.28 Gbps worth of packetized data over a single cable. Someone probably looked at that and said, "Wait a minute, why just use this for display data? Why not make it full-duplex and use it for transporting PCIe packets as well?"

    As it is, you can plug DisplayPort gear into most Thunderbolt ports and just use it as if it was connected to a regular mini-DP port. Also, Intel is supposedly now shipping a much cheaper, single-channel, PCIe only variant of the Thunderbolt controller called "Port Ridge".
  • iSayuSay - Friday, May 11, 2012 - link

    One question:

    From the picture, I saw Thunderbolt cable being plugged into on board TB port (which fully make sense of course). But what does it mean? Which GPU used to control the TB display?

    Intel HD 4000? Or regular PCI-E GPU? If it's the latter, it means the port on GPU card become obsolete? And can we still use HDMI to be used as secondary display?

    It kinda confuse me about how we use Thunderbolt on a regular tower, iMac and Macbook are integrated system so things become simple. But desktop and next gen MacPro (if it ever be updated) bring another problem since you can use PCI-E GPU which has its own display ports.

    Suggestion please? Thank you
  • ggathagan - Friday, May 11, 2012 - link

    Did you read the second page?
    Ivybridge chipsets utilize Virtu software to allow you to switch between the IGP and a discrete GPU without having to physically switch the display connections.
    The display can be attached to either the motherboard connection or the dGPU connection and Virtu will allow you to use use the dGPU for gaming, while utilizing the IGP for less strenous uses.

    As Anand clearly stated, you can use the TB connection for your display and that process still works properly.
  • Zoomer - Friday, May 11, 2012 - link

    HD 4000 as clearly mentioned in the article. Reply
  • Death666Angel - Monday, May 14, 2012 - link

    And GTX680 when firing up games.... Reply
  • Ujjwal Tiwari - Friday, May 11, 2012 - link

    I couldn't quite grab it from the article, Was it possible to play games from gtx 680 on mac display through TB port? Or was another normal display used to connect to graphics card? And how in the hell are we supposed to run gpu intensive apps through HD 4000 only?? (Assuming we use mac display as only display?) Reply
  • Zoomer - Friday, May 11, 2012 - link

    Yes with Virtu. Reply
  • ggathagan - Friday, May 11, 2012 - link

    Read the second page and read up on Ivy Bridge and Virtu!

    The only difference between this and a non-TB motherboard is that you have yet another connection that can be used for your display.
    Nothing else changes with regard to how the IGP and dGPU work on an Ivy Bridge board.
  • ggathagan - Friday, May 11, 2012 - link

    If there's one market besides mobile where ThunderBolt could really take off, it's in the corporate world.

    Just think how inexpensive it could be if someone like Dell or HP were to take the following approach:

    You put out a line of computers and monitors where all the connectivity is at the display.
    There would no longer be any need to build network, USB, Firewire or even sound into the motherboard.
    This reduces the computer itself to nothing more than the CPU, memory, GPU and storage.
    Even the storage could be separately housed.

    That makes for a tiny foot print and your customers can now upgrade their computers to the latest-and-greatest CPU/GPU with no more effort than it takes to plug in the TB cable and the power cord.

    This is also applicable to the HTPC community if someone ends up marketing TB-equipped TV's.
  • descendency - Saturday, May 12, 2012 - link

    Wow. Cool. I'm not really interested in this in the desktop area, but a laptop... yes. Just yes.

    If Lenovo reads this, an Thinkpad X2xx tablet with this is a day 1 pre-order. I can't imagine how awesome this will be when it gets into the market at large.
  • Wardrop - Saturday, May 12, 2012 - link

    Making the monitor the interconnect hub for a computer does make sense. You're more likely to move around a computer than a monitor. Portability doesn't just refer to laptops and using your computer away-from-a-desk. Portability for me means moving a computer from one desk to another. This is especially applicable at my place of work, where I have technical requirements that exceed the capabilities of a laptop, but where I need to move between offices which have a hotdesks with monitor, keyboard and mouse.

    Basically, all of the logic should reside in the computer, but the monitor should act as the consolidation point for all the copper connecting your devices to your computer. I wouldn't like to see the monitor have any intelligence, with thunbderbolt (the carrier) being the exception.
  • Impulses - Saturday, May 12, 2012 - link

    TB displays are great for laptop users but they're not a prefect solution for everyone... Any power user that has multiple displays probably doesn't care a whole lot for having a bunch of peripheral connections hanging off displays that are possibly wall or arm mounted. The holy grail for TB is still being able to offer that powerful external GPU that makes a thin and light laptop a gaming machine at home imo, but we seem to be a long way off from that. Reply
  • AnTech - Saturday, May 12, 2012 - link

    It should be possible to power on a computer from a Thunderbolt port, as was previously possible with Macs using a device like the i-Cue for USB
    or the former ADB Apple keyboards that also allowed powering on the Mac from them. That is really convenient when the computer is away and difficult to reach, below the table, etc.
  • rikm - Saturday, May 12, 2012 - link

    First page says: Co-developed by Apple and Intel, Thunderbolt is a tunnel that carries both PCIe and DisplayPort traffic to the tune of 20Gbps per channel (10Gbps up and down).

    Page 3 says: Total available bandwidth for a single-port Thunderbolt device is 20Gbps bi-directional (40Gbps total), so there's still additional headroom available.

    Which is it?
  • KitsuneKnight - Saturday, May 12, 2012 - link

    It's 10Gbps up or down per channel. So 20Gbps per channel.
    There's 2 channels, so there's a total, aggregate bandwidth of 40Gbps in both directions.

    So it's both.
  • buzznut - Saturday, May 12, 2012 - link

    This is a very vanilla looking board for around $200. And the PCI-express lanes are pathetic. They choose to implement Thunderbolt on a motherboard with how many lanes??

    I like msi so I really don't understand this. And for some reason, Intel motherboards in general don't provide enough PCIx lanes. As features are added, it seems like they limit the lanes even more. What gives?

    In my opinion, someone would have to be pretty desperate to have Thunderbolt in their system if this board is representative of most implementations. Sure, its a new technology but this is indicative of a larger problem-Intel had the same problem with their implementation of USB 3.0; not enough PCI express lanes. A person would be foolish to use this board with crossfire or sli.
  • ggathagan - Saturday, May 12, 2012 - link

    Why desperate ?

    I think many are not understanding ThunderBolt.
    It's not a device in-and-of-itself, it's a flexible bus mechanism that's fairly protocol-independent.

    You're viewing it as something that consumes x4 PCIe lanes, eliminating their use by another device.
    That's not accurate.

    In the case of this board, it allows you to use the 4x bandwidth for USB, firewire, SATA, DP, gigabit ethernet or any combination thereof.

    If a motherboard vendor chose to do so, they could link it to a PCIe 3 x16 lane and still not utilize TB's full bandwidth.

    Given the tests that show little or no difference in PCIe 2 x8 vs x16 performance, that opens a LOT of possibilities.
  • repoman27 - Sunday, May 13, 2012 - link

    Ivy Bridge LGA 1155 CPU's provide 16 PCIe 3.0 lanes for dGPU use, which is pretty much twice the bandwidth that came with Sandy Bridge and plenty for most setups.

    The PCH continues to offer 8 PCIe 2.0 lanes for additional features, which is about all that is reasonable given the 20 Gbps limit of the DMI 2.0 connection to the CPU. The more features a motherboard manufacturer adds, the more of these lanes they consume, and the fewer that are left available to the end user as open PCIe slots. By my count, MSI have used 6 of the 8 lanes branching off of the PCH by adding Thunderbolt, FireWire and an additional SATA 6 Gbps controller.

    If you really think 16 GB/s of bandwidth isn't enough for your particular CrossFire or SLI setup, you can always go with an LGA 2011 solution which would give you 40. The bottom line is that if you want more PCIe lanes, you need a bigger socket with a lot more pins to provide those connections, and that ain't cheap.
  • richard451 - Saturday, May 12, 2012 - link

    It's odd that the article states multiple times that Thunderbolt does not need any drivers for the OS to take advantage of it, yet the update from Intel states once Windows gets updated Thunderbolt drivers things will improve. Reply
  • ggathagan - Saturday, May 12, 2012 - link

    The comments about drivers from Intel revolve around extending hot-swap capability to TB-connected devices.
    Much like hot-swap AHCI is supported in Windows 7 while not in XP, the basic functionality for the connectivity in W7 is due to TB connections being seen as a PCIe device or a DP device in W7, while the hot-swap functionality has to be achieved through drivers.
    I suspect Windows 8 might have said capability built in.
    I doubt MS is interested in doing the same for W7, since they want us all to move forward on their 'enlightened' Metro path.
  • repoman27 - Sunday, May 13, 2012 - link

    To clarify, the Thunderbolt controller does not require any drivers, but the devices you connect via Thunderbolt may.
    Thunderbolt devices tend to include PCIe hardware for which Windows drivers already exist, but the drivers were not designed with hot-plugging in mind. Once the drivers are tweaked to support hot-plugging, the devices won't require a restart in order to be recognized by the system, and will become eligible for Thunderbolt certification for Windows.
    It's the drivers for the PCIe connected hardware in the Thunderbolt devices that need updating in order to work properly in the new Thunderbolt context.
  • SimoxTav - Saturday, May 12, 2012 - link

    May I suggest to take a look at the performance videos of the eGPU solution listed on the topic?

    Check the "1.2Opt-Int" videos (scroll down) that means "x1 PCI 2.0 + Optimus Compression on internal screen". The results are IMHO quite enjoyable with a GTX560. With Thunderbolt can only be better (and easier to be implemented). Currently the performance are about 75-80% of the x16 counterpart (obviously depending on the specific game).
  • repoman27 - Saturday, May 12, 2012 - link

    Overprovisioning is usually not a problem. What you have here is oversubscription, which definitely can be. The only chips on this motherboard that seem to be using the 8 PCIe lanes off of the PCH are the FireWire and SATA controllers, which use one apiece, and the Thunderbolt controller, which seems to be a 4 channel Cactus Ridge chip which uses four lanes. The audio chip does not connect to the PCH via PCIe, nor does the PHY chip for the Intel NIC, as far as I know. If they had just gone with 2 less PCIe x1 slots, they wouldn't have had any contention issues.

    It's odd that MSI would use a 4 channel Thunderbolt controller and only provide a single Thunderbolt port. The PCIe throughput figures definitely point to this being the 4 channel version though. If it can drive 2 Thunderbolt displays, then it definitely is, and I'm guessing that the reason why only two of the display outputs work at the same time is because 2 of the 3 video signals from the IGP are going to the Thunderbolt controller.
  • Death666Angel - Monday, May 14, 2012 - link

    But then VGA+HDMI would not work, would it? Reply
  • repoman27 - Monday, May 14, 2012 - link

    Nope. I realized that after I posted, but then I hurt my brain trying to figure out how the display signals were actually routed from the PCH in order to cause this behavior. Reply
  • Jamiex - Sunday, May 13, 2012 - link

    could you get as high framerate when playing game through thunderbolt with the Lucid's Virtu GPU virtualization and an external GPU as you would when connecting directly to the external GPU? I don't fully understand how it works. Reply
  • Death666Angel - Monday, May 14, 2012 - link

    Pretty much, you lose a few percentage points going through the iGPU frame buffer. The way it works is that the GTX680 (or any dGPU) renders the frame and has it in its frame buffer. The Virtu software then copies said frame buffer over to the iGPU frame buffer and that iGPU has a direct connection to TB/HDMI/VGA off the mainboard. The act of copying the frame buffers has some overhead, but performance in the last review I saw of it was 97% to 100%. Reply
  • gtm - Monday, May 14, 2012 - link

    I find Thunderbolt interesting for the ability to put one big and loud pc in one room, and get all I/O in another room, saving noise, space and cable mess.

    But is it possible to wake an hibernating PC using for example an USB-Keyboard connected to a thunderbolt display?
  • Death666Angel - Monday, May 14, 2012 - link

    3m cable length for the copper version doesn't seem like enough for that usage case though. :-) Reply
  • repoman27 - Monday, May 14, 2012 - link

    That's why you need to get one of these:
  • gtm - Tuesday, May 15, 2012 - link

    Yes, I was thinking of those. Maybe even with a little more bandwidth, so that a Quad-HD display can be plugged, when they become affordable.
    But running upstairs to resume from suspend is no fun.
  • roaminggnome - Tuesday, May 15, 2012 - link

    Did you happen to test to see if the Display supported Audio IN/Out or Video over the built in speakers/iSight? Reply
  • Zclyh3 - Wednesday, May 16, 2012 - link

    Where can I actually BUY this board? I can't find it anywhere. Reply
  • slyck - Thursday, May 17, 2012 - link

    Not knocking the article, but just don't see any value in this. Reply
  • ChickenK - Saturday, May 19, 2012 - link

    Is Virtu necessary to run the dGFX? Without the software, would the Thunderbolt Display's output only be driven by Intel's processor graphics? Reply
  • timelapser01 - Tuesday, May 22, 2012 - link

    I may add, that while using a Pegasus R12 in a small render/graphics office Mac network (MBP, iMac and a recently added Hckintosh ;), now the limit for me on accessing that Pegasus is the Gbit Network itself.

    Since fibre channel is far out of reach for a small business, next accompanying development necessary for TB use would be more 10GBit network PCIe cards and 10Gbit ExpressCard available for prosumers. 1 GBit is a bottleneck, once you use TB devices.

    thanks to anand for his great reviews.

  • chaoflux - Thursday, May 24, 2012 - link

    Are they considered Windows certified Thunderbolt Devices? I'd love hot plugging and the ability to use both of my Thunderbolt Displays in Windows among other nuances.... Reply
  • Asterix007 - Thursday, June 14, 2012 - link


    do you guess it's possible to connect a GD80 MB to an IMAC27 (mid2011) with the apple Thunderbolt cable, and put imac as External display using CMD + F2 combinaison key ?

    I try to do so, but imac don't switch to GD80.

    My GD80 has FW 1.0.
    How did you put bios settings for TB part when connecting to the Apple Display 27 ?

    Thanks for your answer.

  • claptrap22 - Saturday, August 04, 2012 - link


    How did you folks get the iMac to work as a target display? I built a rig last night with the same motherboard. I tried to use my 21.5" iMac (mid-2011 with Thunderbolt) as a target display and couldn't get it to work. Please advise. Much appreciated.
  • mellertson - Wednesday, January 08, 2014 - link

    I setup an ASUS Z87-Expert. It has a Thunderbolt port. I can get the sound and other Thunderbolt connected devices to work. But, I can't get the display to work on Windows 7. Any ideas?

    Your help would be much appreciated since clearly you have it working in the video you posted. :-)

Log in

Don't have an account? Sign up now