Waiting for Thunderbolt--one port to rule them all
What if there was a connectivity standard faster than the fastest version of USB, worked with even more types of devices, and was even compatible with USB itself, via an adapter?
Since the late 1990s, USB has been the dominant PC connectivity technology for external devices. As anyone who remembers the era of parallel, serial, and PS/2 connectors can tell you, USB is a wonderful thing--whether you're talking about the original 1.1 standard, the now-pervasive 2.0 version, or the still-emerging 3.0 variant.
But what if there was a connectivity standard that was faster than the fastest version of USB? And what if it worked with even more types of devices, including displays? And what if it was even compatible with USB itself, through the use of an adapter?
It's no fantasy--it's Thunderbolt. And I have high hopes for this technology, even though it's not yet clear that it'll be anywhere near as universal as the good old Universal Serial Bus has been for years.
Thunderbolt originated in Intel's labs as a technology called Light Peak. It mashed up PCI Express (the zippy standard used for cards you install inside a PC) with DisplayPort (the new standard for connecting displays), and added compatibility with various other connection technologies. Apple got excited about the idea and began working with Intel to commercialize it. Under the catchier moniker Thunderbolt, it's now available on all new Macs except for Mac Pro desktops, and it's a safe bet that Apple will add it to those machines in their next big upgrade.
With two 10-Gbps channels, Thunderbolt is supercharged by any definition: USB 3.0, which is itself no slowpoke, provides a single 5-Gbps channel. (USB 2.0, still the most pervasive port out there, chugs along at a comparatively pokey 480-Mbps.)
That's great for hard disks and other devices that need all the speed they can get. But it's Thunderbolt's versatility that I find most promising. With the right cables or adapters, you can connect practically anything to a Thunderbolt port--not just fast new Thunderbolt displays, drives, and other devices, but also older monitors, USB and FireWire peripherals, and other Thunderbolt-equipped computers. And daisy-chaining lets you hook up multiple devices to one Thunderbolt connector.
By quickly moving to put the technology in most of its computers--as well as a new 27-inch Thunderbolt display--Apple is doing its part to get the technology rolling. Promise Technology has shipped , and LaCie has an external drive in the works. But Thunderbolt will be most powerful when it's more widely supported, and it won't be more widely supported until the cost of building it into products comes down. (Even a simple Apple Thunderbolt cable is currently an intimidating $49.)
Still, the odds seems excellent that Thunderbolt will eventually be at least as commonplace in the Apple ecosystem as Apple's FireWire once was. (Apple seemed to be itching to retire FireWire even before Thunderbolt came along and will presumably move even quicker to nuke it now that Thunderbolt is here.)
What's far less clear is whether the technology will catch on in the greater, Windows-centric PC industry outside of Apple's sphere of influence. As with any new interface, there's a basic chicken-and-egg conundrum: there's no need for PCs to have Thunderbolt ports unless there are peripheral devices to plug into them, and Thunderbolt peripheral devices are useless unless there are PCs that can accommodate them. So far both chickens and eggs are non-existent. And while Apple can pretty much decree that something's a standard by putting it on all Macs, standards only happen in the rest of the industry if Acer, Asus, Dell, Hewlett-Packard, Lenovo, Sony, and a bevy of other companies all think they're a good idea. Even when consensus comes, it takes time.
I'm hoping for good news in 2012, but I fret that Thunderbolt--even more than FireWire--will end up being "that connector that's a hit on Macs but an underutilized oddity on PCs." I keep meeting with companies in the PC hardware business and expressing my enthusiasm for it--and getting various combinations of skepticism, caution, and fear in return.
Tech execs point out to me that Thunderbolt is expensive to implement. They remind me that it competes with USB 3.0, another next-generation connector that's much further down the road to ubiquity. They tell me that they think it's mostly of interest to niche audiences such as video pros who need ultra-fast storage.
They may be right, and I may be disappointed in the long run. But I'm not giving up just yet. I remember when USB was expensive to implement, not widely supported, and--now that I think about it--kind of flaky. Eventually, it got affordable, commonplace, and indispensable. The rest is connectivity history. Here's hoping Thunderbolt ends up repeating it.