I’m a huge fan of Thunderbolt. A single wire carrying both DisplayPort and high-speed PCIe data is an incredibly elegant approach to minimising cable clutter even if you don’t need the blistering speed, especially when you can use an Apple Thunderbolt Display as a hub for your USB devices.

I also admire clever tech. The reason you can daisy-chain up to six separate devices is because Thunderbolt automatically multiplexes and de-multiplexes the signals as needed. Thunderbolt 2 takes this approach one step further, combining two 10Gbit/s channels into a single 20Gbit/s connection, with the the Thunderbolt controller again doing all the work. It’s impressive stuff.

A fast, clever technology developed by Intel and enthusiastically marketed by Apple ought to stand a fighting chance at mass-market adoption. Sadly, there’s so far not much sign of this happening. It’s all looking rather reminiscent of Firewire … 

Firewire was far superior to USB. While USB 2 claimed peak speeds of 480Mbps, it rarely delivered them. Firewire 400’s delivery of 400Mbps was consistent, making it significantly faster in real-life use. And Firewire 800 left USB for dead.

But we all know that technological superiority is no guarantee of commercial success. The mass-market went with USB. Partly because consumers buy numbers without necessarily knowing what they mean, but mostly because it was cheaper.


Firewire didn’t die. It’s still in use in many high-end setups, and you can still buy Firewire drives today. But it definitely became a niche product, with even Apple eventually dropping the port from its MacBooks.

Thunderbolt currently looks to be heading the same way. That wasn’t the plan. Intel intended it to be the new USB. Optical thunderbolt was supposed to take over from copper, and there was supposed to be a Thunderbolt port in every PC. Neither has happened.

The reason is USB 3. It’s not as good. It’s not as fast. But it’s cheap and it’s familiar. You can virtually count on one hand the number of Windows PCs out there with a Thunderbolt port. I think it’s no exaggeration to say that without Apple, Thunderbolt would already be dead.

The new Superspeed USB protocol matching the 10Gbps throughput of standard Thunderbolt adds to the pressure, and Thunderbolt 2 – like Firewire 800 before it – is unlikely to be enough to sway the market.


But there are two glimmers of hope. The first is the new Mac Pro. That has not just one Thunderbolt 2 port, but six. Given that each one can drive six Thunderbolt devices, Apple clearly believes that at least one segment of the market is going to share its enthusiasm for the interface.

So far, that segment looks to be video editors. They want a machine that is lightning-fast, that will drive multiple 4k displays and which provides high-speed access to lots of external drives. The new Mac Pro seems to be that machine, and Thunderbolt 2 one of the key elements of its appeal, supporting both the displays and the drives.

Of course, one expensive, niche machine is never going to turn Thunderbolt into a mainstream protocol, or create a flood of mass-market peripherals. But I do think it’s enough to ensure that it will hang in there, the way Firewire did – and that may buy Thunderbolt the time it needs.


Time, because that second Thunderbolt trick – driving multiple 4K displays – is going to grow in appeal. Sure, today, when they cost $5k a pop, that’s as niche as it gets. But in a couple of years’ time, when 4K is the new HD, all those ordinary business professionals and home enthusiasts with twin-monitor setups are going to have twin 4K monitors, and that’s when Thunderbolt is going to come into its own.

It’s still not a done deal: updated versions of both DisplayPort and HDMI will do the same, and those, like USB, are built on familiar standards. The fact that Thunderbolt is one cable that does all of it – displays and drives and more – doesn’t mean it will emerge victorious.

What’s my view? I’m confident Thunderbolt will still be around in five years’ time. I’m just not 100 percent convinced the port will still be there in MacBooks by then. I’m hoping I’m wrong.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

58 Responses to “Opinion: Is Thunderbolt doomed to be the new Firewire, or can the new Mac Pro save it?”

  1. “Thunderbolt currently looks to be heading the same way. That wasn’t the plan. Intel intended it to be the new USB.” — I’m not sure you’re right on this. Intel never intended thunderbolt to compete with USB 3.0…they always said that thunderbolt *compliments* usb in PR, it doesn’t compete with it.


  2. The main reason I haven’t adopted a thunderbolt setup is clearly down to cost. I’m pumping out multiple HD videos weekly for my work and I still can’t justify £35 on a cable and a £50-100 premium over a Firewire 800 hard drive.


    • This is exactly it. Active cables crippled adoption. USB is dirt cheap. Cables are dirt cheap. If they wanted Thunderbolt to succeed it would have to compete not just on technical merit but match price or even undercut it.


    • I did. I went Thunderbolt 2 years ago with a Promise 12TB array, which puts my SSD in my 2011 MBP even to shame with 500MB/s throughput. There is no other way for that amount of storage and throughput to happen without Thunderbolt.


  3. For a product to be successful in the mass-market it needs to offer something that people really WANT and at a price that seems affordable. Thunderbolt does not meet that criteria whereas USB does.

    Thunderbolt is great technology but face it….where else other than in Apple’s ecosystem do you pay out the rear for what people perceive is a simple cable. Sure, you and I both know that it’s not a simple cable but to users unfamiliar with this and who just don’t care, this is one of the issues. $30-$50 thunderbolt cables are ridiculous.

    Throw in to this that you just don’t see much in the way of basic hard drives in thunderbolt enclosures. The products you typically see are raid systems many with ssd’s.

    In other words it’s been a niche item all along and will likely continue to be. It’s great that Apple included it throughout their line. It certainly hasn’t been perfect in it’s use for the features it offers though. Target Display Mode definitely needs some work so in retrospect it almost makes you wonder if Apple held off on bringing it to the Mac Pro(it’s target audience) just so it could try to perfect it first.


  4. I feel that the first mistake with Thunderbolt was Apple’s exclusive inclusion of the technology on its Macs. If that was a nifty video card or processor, it would have been a great boost for Apple, but a brand new interface needs to be in as many peoples’ hands as possible to encourage peripheral manufacturers to support the standard.

    If it had been built into PCs from day one as well as Macs, with strong cross-platform support, things would be looking better now.


  5. This article is missing some vital historical context. Without this context you cannot adequately understand how Firewire and Thunder differ and how they are alike.

    Firewire failed because it was an ancillary chip that needed to be put on the motherboard. Apple, not being the cash enriched company it is today, looked at Firewire as a better connection than USB and a way of generating revenue/profit from per port licensing. So if you were a motherboard manufacturer you had to pay a certain fee (usually cents per port) to include Firewire. This chaffed many of the motherboard vendors and thusly they left Firewire off their products. Intel integrated USB right into their controller hubs meaning that these vendors had a relatively free solution albeit lower performing.

    Consumers don’t really have a choice if Firewire is absent from their product yet USB is present. Thunderbolt is different. It is the external manifestation of PCI Express. This has many advantages but the largest are that eventually Intel will likely be able to integrate Thunderbolt controllers into their Platform Component Hub reducing cost. Because Thunderbolt is PCIe it means that drivers do not have to be modified extensively. Everything is speaking the same language as opposed to Firewire which required unique drivers.

    So Thunderbolt has three distinct advantages that make it different than Firewire. 1. It’s an Intel product that could move to the motherboard 2. It doesn’t require different drivers 3. Licensing is more transparent and will get cheaper.

    Intel’s marketing has never suggested that Thunderbolt would be a mass market item. The chipsets (Light Ridge, Cactus Ridge, Eagle Ridge etc) are expensive yet getting cheaper but there’s no way Thunderbolt was expected to become a mass market item when Intel was integrating USB 3 into Ivy Bridge PCH.

    Thunderbolt is designed to create new product categories. External GPU, Displays like Apple’s Thunderbolt display and more. Things that USB simply was not designed to connect. It’s going to take time and all too often people want to cast off technology that they think should be cheaper without looking at why said technology is expensive.


    • Ben Lovejoy says:

      You make some well-argued points. My personal view is that Intel had great ambitions for Thunderbolt, while recognising that it would never fully replace USB.


    • Love your post except for the external GPU thing — so many companies (mainly MSI) have promised thunderbolt eGPUs and they have been delayed and or cancelled, and Intel I think still to this day refuses to certify an external GPU as a thunderbolt device….so from what I have heard, Intel won’t even allow a company like MSI to make an external thunderbolt GPU because they won’t certify it. I guess they must really hate AMD/Nvidia, because we all know that’s who’s GPU chips would be used.


  6. Only if prices decrease quick as for the buyers as for the devs and only if we will see some new
    external GPU adapter with good power adapter.Otherwise its already dead.


  7. “The mass-market went with USB… but mostly because it was cheaper.”
    “The reason is USB 3. It’s not as good. It’s not as fast. But it’s cheap and it’s familiar”

    You answered why Thunderbolt won’t take over USB 3.

    Thunderbolt needs to be cheaper to implement and market just more than speed. I bet 90% of users don’t need that fast of data transfer, but if you market as a way to keep your desk clean, with increased speed, then you might have reason enough to buy.

    Also, they could have Bolt (from the movie) stand in for Thunderbolt. “When I need super fast speeds (showing Bolt being edited on 3-4K monitors) I turn to Thunderbolt.” Gimmick, yes, but affective?

    Anyway, I am sure the masses are more limited to internal specs than external, especially those on PCs where they get clogged with all kinds of junk!


  8. The principal mistake in this article, and the prevailing attitude of many of the commenters it seems, it that Thunderbolt has to be mass market to be successful. It doesn’t.

    This is no more true than specialty vehicles like sports cars or heavy-duty trucks need to sell as well as more popular basic cars.

    Or closer to home, it’s the same mistake that you make if you say Apple is loosing because they don’t have dominant marketshare in computers or cell phones. Apple doesn’t have marketshare dominance, it has profit-share dominance.

    I have NO DOUBT, USB will continue to hold the dominant marketshare position, and I’m thrilled it’s getting faster. But Thunderbolt will continue to outpace USB in terms of speed, and there is a segment of the Professional market (myself included) which will happily pay for that privilege because it means we can get our work done faster, do more, and be more profitable.

    Thunderbolt isn’t expensive if you consider what I charge for my time, and the fact that when the MacPro launches I’ll be able to transfer data at 4X the speed of USB3. That’s HOURS of my time every day on big jobs with terabytes of media.

    Thunderbolt doesn’t have to be cheap; any more than any better, faster, more capable peripheral does.

    It’s doesn’t exclude USB3 on Macs. Being combined with DisplayPort, it doesn’t take up extra port-space.

    If you don’t need it, don’t use it. But there’s a segment that does, and Intel’s near-term roadmap ends with optical 100Gb/s- which (guess what) isn’t going to be cheap either. But again, you don’t HAVE to buy in. This is not an either/or.

    This is a no loose scenario.


    • Ben Lovejoy says:

      The argument is not that in can’t be a niche product, rather than it would be a shame if that’s how it ended up, as the potential is there for much more


      • While USB3 has already become commoditized, Thunderbolt is on a much more aggressive development path- so it’s never going to be the “cheap” alternative.

        If the performance increases aren’t worthwhile for a particular user, then it’s no one’s “fault”, and it’s not Intel’s “mistake”. I never saw TB as a contender to take the crown from USB. I think that’s just denying what it is- a Pro-focused I/O that I think Intel (and Apple) very much want to replace any need for PCIe when the thru-put gets there. The MacPro is certainly evidence of that.


  9. degraevesofie says:

    I wonder if Apple could use its success with iOS devices to bolster the Thunderbolt ecosystem?

    Specifically, since Lightning is a configurable connection, they could in principle offer a Thunderbolt-to-Lightning option with greater bandwidth than the current USB2-over-Lightning situation. Of course, to be significant, that would require the iOS device to be able to take advantage of the improved bandwidth.


    • Problem is Thunderbolt is Intel. iOS device are ARM based so if it is possible at all it would take a Herculian effort to create a Frankenstein box.


      • Thunderbolt also requires a PCI express controller, which none of these ARM SoC’s have. Plus, no one needs 10Gbps speeds on a phone or tablet / handheld device. The best solution for a tablet or phone is Gigabit wifi; it’s faster than USB 2.0 and it’s more economical than implementing a USB 3.0 controller in the SoC…Why Apple left Gigabit wifi out of the 5S and 5C is beyond me…Samsung put Gigabit Wifi in the Galaxy S 4, released more than 6 months ago..


  10. Dave Park says:

    Thunderbolt’s success or failure is down to the devices available to plug into it. In this case, Intel has been very selective about the partners it shares the technology with. Two companies I have been involved with are still on the waiting list to get information and development tools after two years or more. For these companies, it’s simply impossible to enter the market with their mass market or novel ideas. You won’t see their products, because Intel just wants to deal with big name companies with lots of resources.

    The biggest side effect of this is high prices due to lack of competition.


  11. You know what was an “elegant approach to minimising cable clutter” PCI Slots. It was also a universal standard, but now we will have to deal with an octopus, doesn’t seem to follow the less cord trend that apple pushes. I think Thunderbolt is great/fast solution for many small peripherals and drives, but it is a complete step backwards when looking at the new mac pro as a pci replacement. I work in high demand video processing and we needed more slots for video cards and rocket cards… disappointing. Honestly I hope apple adapts it into a network interface as they did with firewire, but seeing as it is not backwards compatible with the legacy stations we have at work and the lowering prices of GbE 10 I don’t think it will be adopted quickly. USB 3.0 or ESATA 6G has become the new fast standard out here.


  12. Beta was better than VHS. I understand where you are coming from. In this day and age I would think a product that cuts down on cables alone would make it a winner. I wish thunderbolt would have gotten the external graphics cards that were touted as the next big thing or at least shown off early on. It would be nice to be able to dock a laptop and instantly have more graphics power. I think for the average consumer usb 3 falls under the I understand what a usb device is and 3 is more than 2 so it must be great category. Also with cheaper costs usb 3.0 hits cheaper buying options and ends up in more devices. As we all know this isn’t always for the best but it’s what we end up with. I do believe thunderbolt will stick around though.


  13. USB’s biggest pain is the fact that you need to use a hub.
    Firewire and Thunderbolt connect in serial.

    If Thunderbolt fails to succeed it’s not going to be due to being an inferior than USB product. It will be due to public ignorance. Just like how VHS beat Betamax, when Betamax was a much better product.


  14. I don’t care what becomes the standard, USB, FireWire or Thunderbolt – I just wish I can back-up onto an external drive without the drive disconnecting from my MBP from time to time.


  15. Len Williams says:

    I always preferred and bought Firewire external drives for the speed. However, Apple made licensing Firewire expensive and so peripheral developers concentrated on USB. The same thing is happening with Thunderbolt. The general public doesn’t understand technical details and superiority. It largely only understands cost, so until and unless Intel and Apple bring down the price of Thunderbolt, it will remain a niche technology like Firewire, despite all of its benefits and superiority over USB 3.


  16. Monte Miller says:

    It’s very difficult to fight the mentality that goes I-need-port-for-hard-drive, I-need-port-for-monitor, I-need-port-for-mouse, etc. Thunderbolt is fighting the mindset that one port cannot do everything. How long did PC manufacturers include parallel and VGA ports? Forever. As usual, Macs will have the most advanced technology that the dunderheads have no idea what to do with.


  17. I’m struck by the way in which the processor (Multi-core Xeon + acceleration hardware) in the MacPro looks just like the compute node in a state-of-the-art super-computer. Now, if you could only join multiple MacPros together by Thunderbolt, you be able to build mini-supercomputers pretty easily.


  18. Ian Eberle says:

    I don’t see the Mac Pro saving it because the Mac Pro isn’t going to be a popular product – it’s way too expensive to be a mass-market product like the iPhone. However, if Apple does ever make a TV like the rumors say, they could use Thunderbolt instead of HDMI and all the other standard TV hookups. I can see it catching on then. Blu-Ray players and cable boxes would have to be updated to support Apple’s new Thunderbolt technology.


  19. gann (@gann) says:

    I think we are all underestimating the potential PCIe over a wire… with so many things going SoC, this ability to augment rather than replace is what’s next (imo).


  20. Thunderbolt cannot be a success while there is very little peripherals out there for it. I haven’t been following things, but I bet there aren’t out there more than 10 products altogether that are built for TB. Just like Bluetooth 4.0, manufacturers are very sheepish to release peripherals for it (I have yet to see a BT4.0 headset, for example, while there’s tons of smartphones out there that support it)

    As long as there is no comparable (in price) TB alternative to cheap USB drives between 16 and 128GB or cheap USB HDDs between 320 and 750GB, don’t expect much interest around it. TB is a niche and it will stay a niche, unfortunately methinks.

    Personally, I’d love to see a 64/128GB TB flash drive, but something tells me it’s only going to stay a dream


    • This “10 products” nonsense has gone on long enough.

      Apple’s website alone lists 40 different thunderbolt peripherals including portable, desktop, and RAID storage; Ethernet, SAN, Firewire, USB, and PCIe adaptors/expansion chassis.

      And that doesn’t include Apple’s own cables, adaptors, and Cinema Displays.

      And it doesn’t count any of the DOZENS of Pro video/audio oriented I/O from Blackmagic Design, Matrox, AJA, AVID, etc., etc.

      There’s LOADS of Thunderbolt out there if you’re in the market that it’s going to appeal to right now- power users for whom a doubling (and now quadrupling) in speed is a very meaningful benefit.


  21. ellzworth says:

    As of right now, Firewire 800 is still my hero. I currently own a 2010 15-inch Macbook Pro that has two USB 2 ports, one Firewire 800 port and one Thunderbolt port. Apple got rid of that amazing expresscard port which limits me to using what I have. I’m a graphic designer who travels for work. I need to carry my workstation with me including a display port to DVI port cable and a handful of hard drives. The problem that I have with my set-up and with what is currently offered in the market is the lack of multiple thunderbolt ports on my laptop and on portable drives (I’m aware that the new Macbook Retina now has 2 Thunderbolt ports.) Intel sold me with Thunderbolts ability to daisy chain multiple devices with it’s technology. So far, I haven’t found one reasonable portable thunderbolt drive that offers two ports for daisy chaining. Companies like G-tech and Lacie pawn off daisy-chaining with a small disclosure (as long as their drives are the last in the chaining).
    Someone earlier mention the fact that when HDMI cables first came out, they were $100 a pop. That’s a harder comparison because that was at a different level where HD DVD and Blu-ray players from Multiple companies were being released. Consumers were forced to purchase them in order to enjoy full hd quality (Smart move). In the entertainment and broadcasting world, they’ll pay X amount of dollars for something they know works. Thunderbolt has been availble for about 3-4 years now and still today, we purchase hard drives that have Firewire 800 and eSata connectivity. The studio I currently work in just purchase 10 12 Core Mac Pros and none of them have Thunderbolt ports. In order for a technology like this to succeed in today’s time, it almost needs to be given away or sold as inexpensive as the competition. I’m looking forward to the day where thunderbolt cables cost a few dollars and actually comes included with the hard drives. Hopefully soon.


  22. drtyrell969 says:

    Thunderbolt is a red DVD box.


  23. Thunderbolt like FireWire serves a niche market. A market for high-end users who make a living using these products as tools, or just guys who really love technology.

    If Thunderbolt is the next FireWire, great! What’s the big deal? It’s actually 20x better than FireWire ever was and is very expandable. The average person doesn’t need anything more than USB3 and most wouldn’t know the difference between USB 2 and 3 if their lives depended on it.

    Stupid article.


  24. One simple PCIe box for an external graphics card would be enough to have a Thunderbolt port on every notebook/laptop. The GPU makes them expensive and battery hungry and the external GPUs would solve the problem in a heart beat. Unfortunately, all we’ve seen up to now were only promises! *sigh*


  25. Intel has certified many new 3rd party OEM PC Vendors who have begun releasing Thunderbolt ready Motherboards:

    Among them are Gigabyte, MSI, ASUS, ASRock

    Gigabyte Dual Thunderbolt boards:


    ASUS: 4 Motherboards alone:


    Plenty of love is coming for PC Vendors to have T-Bolt ready solutions.

    Next up is Peripheral vendors to start dropping more products:

    NewEgg has its own section on Thunderbolt:


  26. For Thunderbolt to be successful on the same level as USB, it needs to be even faster, cheaper for consumers and licensors. They need to have a better argument to upgrade to Thunderbolt – unless you’re a professional its hard to see the need for it when it comes time to open your wallet. The existence of cheap tech, in terms of cheap netbooks, off-brand (knock-off) wires, headphones and so on – consumers believe that anything beyond that price point is just straight profit, not realising that the increased cost allows companies to create a better product. I am planning on picking up the new Haswell rMBP and I’ll be running a complete Thunderbolt setup – hoping the updated rMBP will include 2.0!


  27. kpom1 says:

    Why doesn’t Intel make it part of the requirements for the Ultrabook standard? That would do more to give it an immediate boost than 4K displays.


  28. thunderbolt failed to capture the attention of PC Graphics Card Markers, AMD uses Display Port, althought Thunderbolt is compbatibile with Display Port none of the Data Transfer features are there and without a New Mac Pro to take advantage of the technology it will go the way of Firewire. Forgotten and used by a small, shrinking userbase.


  29. Ryan Gibson says:

    Only a couple of months ago there were reports that Intel was once again developing a Thunderbolt PCIe card to bring Thunderbolt to more of the PC market. The hope was that older PC units would add the expansion card on their board.

    Does anyone have any update on whether that’s still in the works and possible release dates? Honestly, I love my Thunderbolt Display, but I went with FireWire back in the day big time and if TB is in a decline, I don’t want to keep investing in a tech that doesn’t have a mid to long term future.


  30. Doug Suiter says:

    Anyone who thinks in terms of Thunderbolt versus USB would do well to read the following article:


  31. Seta Soso says:

    Expensive cables, lack of affordable periferals => my 2012 Macbook Pro’s Thunderbolt port has never been used and is unlikely ever to be used. R.I.P. Thunderbolt.


  32. Paul Fawkes says:

    I was interested in Thunderbolt, as a music engineer for audio interfaces, external hard drives and perhaps a 4k display in the future but my Mid 2012 Mac Pro (12 Core, 64Gig of RAM) is not compatible! I don’t understand how you can’t get pcie cards with Thunderbolt ports! I refuse to pay thousands for a brand new computer that will be performance wise exactly the same as what I already have but with Thunderbolt connectivity.


  33. felrobert01 says:

    The only way that Thunderbolt will succeed if that Apple will allow PC’S gamers to connect to their Mac’s as a monitor in target display. Gaming industry is bigger than movie. I would imagine that would be some big crowd.